Environmental Monitoring Systems: a Geospatial Perspective

L1641_Frame_C01.fm Page 1 Tuesday, March 23, 2004 8:55 PM

Conceptual Basis 1 of Environmental Monitoring Systems: A Geospatial Perspective

D.A. Bruns and G.B. Wiersma

CONTENTS

1.1 Introduction ...... 2 1.2 General Monitoring Design Concepts from NRC Reports...... 3 1.3 Overview of Speciﬁc Conceptual Monitoring Design Components ...... 7 1.4 Conceptual Monitoring Design Components ...... 9 1.4.1 Conceptual Framework as Heuristic Tool ...... 10 1.4.2 Evaluation of Source–Receptor Relationships ...... 13 1.4.3 Multimedia Monitoring...... 14 1.4.4 Ecosystem Endpoints ...... 14 1.4.5 Data Integration...... 18 1.4.6 Landscape and Watershed Spatial Scaling...... 21 1.5 Synthesis and Future Directions in Monitoring Design...... 24 1.5.1 EPA BASINS ...... 25 1.5.2 SWAT ...... 26 1.5.3 CITYgreen Regional Analysis ...... 26 1.5.4 ATtILA ...... 27 1.5.5 Metadata Tools and Web-Based GIS...... 27 1.5.6 Homeland Security...... 27 1.6 Conclusion...... 28 Acknowledgments...... 29 References...... 30

L1641_Frame_C01.fm Page 2 Tuesday, March 23, 2004 8:55 PM

2 Environmental Monitoring

1.1 INTRODUCTION The importance of and need for integrated environmental monitoring systems is well established. The U.S. National Science Foundation’s (NSF) Long-Term Ecological Research (LTER) Program has 18 sites in the U.S., each with a study area which generally collects long-term descriptive measurements of air, water, soil, and biota, including data on forest or grassland stands, population and community inventories, and watershed–stream channel characteristics and habitats (e.g., Franklin et al.1). Originally, these observational data were intended to serve as the environmental context for basic ecosystem research conducted on an experimental basis to address the pattern and control of primary production and organic matter accumulation, nutrient cycling, population dynamics, and the pattern and frequency of site disturbance. In a somewhat similar fashion, but focused on atmospheric pollutants,2 the U.S. National Acid Precipitation Assessment Program (NAPAP) established a national network for long-term monitoring of wet and dry deposition of sulfates, nitrates, and “acid rain.” In addition, NAPAP-sponsored ecological surveys (e.g., ﬁsh, invertebrates, forest conditions, and stream and lake water chemistry) were often collected for critically “sensitive” regions and ecosystems but on a much more geographically limited scope than for atmospheric components. Another more recent program for integrated environmental monitoring, building in part on past and ongoing LTER- and NAPAP-related activities, is the NSF’s currently proposed National Ecological Observatory Network (NEON; see www.nsf.gov/bio/neon/start.htm). The U.S. EPA also maintains an integrated monitoring network with a research agenda focused on developing tools to monitor and assess the status and trends of national ecological resources. This program, known as the Environmental Monitor- ing and Assessment Program (EMAP), encompasses a comprehensive scope of ecosystems (forests, streams, lakes, arid lands, etc.3) and spatial scales (from local populations of plants and animals to watersheds and landscapes4). EMAP usually holds an annual technical symposium on ecological research on environmental monitoring. For example, in 1997, EMAP addressed “Monitoring Ecological Condition at Regional Scales” and published the symposium proceedings in Volume 51 (Numbers 1 and 2, 1998) of the international journal Environmental Monitoring and Assessment. The broadest and perhaps most compelling need for better and more integrated design principles for monitoring is based on the numerous and complex problems associated with global environmental change. This includes worldwide concern with climate change,5,6 loss of biotic diversity,7 nutrient (especially nitrogen via atmospheric deposition) enrichment to natural ecosystems,8 and the rapid pace and impact of land-use change on a global basis.9,10 The necessity for a comprehensive global monitoring system was recognized in early publications of the International Geosphere–Biosphere Program (IGBP) and in later global change program proposals and overviews.11–14 In particular, “geo-biosphere observatories” were proposed for representative biomes worldwide and would be the focus of coordinated physical, chemical and biological monitoring.12,15,16 Bruns et al.17–19 reviewed the concept of “biosphere observatories” and evaluated various aspects of monitoring programs for remote wilderness ecosystems and a geospatial watershed site for a designated American Heritage River in the context of global environmental change. These sites represent a broad spectrum of ecological conditions

L1641_Frame_C01.fm Page 3 Tuesday, March 23, 2004 8:55 PM

Conceptual Basis of Environmental Monitoring Systems 3

originally identified in the IGBP. Remote sites, especially at higher elevations, may be very sensitive to global factors like climate change, while the Heritage River watershed site is heavily impacted by regional scale “industrial metabolism.”12 The latter may provide an important “test-bed” for evaluation of geospatial technologies (see text below and References 19 and 20) and related spatial scales of land use change that might be applied later to more remote monitoring sites as part of long-term networks. A conceptual basis for the design of integrated monitoring systems and associated networks has received growing attention in the last two decades as part of scientific research to address these environmental problems from a local to global perspective. Early and ongoing efforts include those of Wiersma, Bruns, and colleagues17–27— most of whom focused on conceptual design issues or monitoring approaches employed and exemplified at specific sites. Others have conducted similar work in relation to global environmental monitoring and research programs.14,28,29 In addition, two major reports30,31 sponsored by the National Research Council (NRC) cover a broad range of environmental monitoring issues, including consideration of comprehensive design principles. The former deals with marine monitoring and the latter report is focused on case studies to address the challenge of combining diverse, multimedia environmental data; this latter report reviewed aspects of the LTER program (at the H.J. Andrews Experimental Forest site), the NAPAP (Aquatic Pro-

cesses and Effects), the Department of Energy’s (DOE) CO2 Program, and the first International Satellite Land Surface Climatology Project (ISLSCP) among others. In this context of national and international global change programs, and the range of complex environmental problems from a global perspective, our objective in this chapter is to delineate and develop basic components of a conceptual approach to designing integrated environmental monitoring systems. First, general concepts from the National Research Council reports are reviewed to illustrate a broad perspective on monitoring design. Second, we highlight aspects of our previous and ongoing research on environmental monitoring and assessment with a particular focus on six components in the design of a systems approach to environmental monitoring. These are more specific but have evolved in the context of general ideas that emerge from the NRC reports. In particular, we use examples from our remote (wilderness) site research in Wyoming and Chile contrasted with an ongoing GIS watershed assessment of an American Heritage River in northeastern Pennsylvania. These examples are intended to facilitate illustration of design concepts and data fusion methods as exemplified in the NRC reports.30,31 Third, we provide a general synthesis and overview of current general ideas and future directions and issues in environmental monitoring design. Finally, we wish to acknowledge the varied agencies and sponsors (see end of chapter) of our past and ongoing environmental research projects on which these conceptual design components are based.

1.2 GENERAL MONITORING DESIGN CONCEPTS FROM NRC REPORTS The design of an integrated environmental monitoring strategy starts with identifying resources as risk in order to initiate development of a conceptual model.30 This process of strategic planning is an iterative process whereby the model may be reﬁned,

L1641_Frame_C01.fm Page 4 Tuesday, March 23, 2004 8:55 PM

4 Environmental Monitoring

elaborated, or enhanced based on practical and technical considerations, available resources, and defined monitoring objectives. This broad strategic approach (Figure 1.1a) usually will culminate in the development of testable questions that feed into the specifics of a detailed sampling and measurement design with a focus on parameter selection, quantifying data variability, and setting up a sampling scheme. This is also an iterative process (Figure 1.1b) with feedback to reframe questions and refine technical components of monitoring design. Data quality and statistical models for analyses also are identified as key components of this strategy. Boesch et al. provide important insight into the use of conceptual models in monitoring design30 and indicate that the term is sometimes misunderstood. A “conceptual model” typically begins as a qualitative description of causal links in the system, based on best available technical knowledge. Such a model may refer to descriptions of causes and effects that define how environmental changes may occur. For example, in monitoring toxic effects of point sources of pollutants, a conceptual model would identify critical sources of contamination inputs to the ecosystem and define which ecological receptors or endpoints (e.g., a particular species, a physical ecosystem compartment, or a target organ system) are likely to be impacted, modified, or changed. As a monitoring system is better defined, a more quantitative model or a suite of models based on different approaches (e.g., kinetic vs. numerical vs. statistical) may be used effectively to address complementary aspects of monitoring objectives. Defining boundaries, addressing predictions and uncertainty, and evaluating the degree of natural variability are also broad concerns in the development of a monitoring strategy and sampling design.30 For example, in monitoring pollutant impacts to streams and rivers, watershed boundaries may need to be established since upstream sources of contamination may be transported downstream during storm events, which may add uncertainty in the timing and movement of materials within the natural seasonal or annual patterns in the hydrologic cycle. For these reasons, a monitoring program should be flexible and maintain a continuous process of evaluating and refining the sampling scheme on an iterative basis. Both NRC reports30,31 highlight the need to address issues of spatial and temporal scales. Most monitoring parameters will vary on space and time scales, and no one set of boundaries will be adequate for all parameters. Also, it is expected that events that occur over large areas will most likely happen over long time periods, and both will contribute to natural variability in monitoring parameters—a condition con- founding data interpretation.30 Wiersma et al. identify spatial and temporal scales as one of the most apparent barriers to effective integration and analysis of monitoring data.31 For example, geophysical and ecological processes may vary at different scales, and both can be examined from a variety of scales. No simple solution to scale effects has yet to emerge for monitoring design although a hierarchical approach to ecosystems and the use of appropriate information technologies like geographic information systems (GIS) and satellite remote sensing appear to be making progress on these issues.31–33 Rosswall et al. and Quattorchi and Goodchild cover various ecological scaling issues for terrestrial ecosystems and biomes,34,35 and Boesch et al. summarize a range of space and time scales30 for various marine

L1641_Frame_C01.fm Page 5 Tuesday, March 23, 2004 8:55 PM

Conceptual Basis of Environmental Monitoring Systems 5

Identify Resources at Risk

Modify Resources

Develop Conceptual Model

Adjust Boundaries No Have Appropriate Resources Been Selected?

Yes Refine Model Determine Appropriate Boundaries

Are Selected Boundaries Adequate?

Yes

Predict Responses and/or Changes

Are Predictions Reasonable?

Yes

Develop Testable Questions

FIGURE 1.1A Designing and implementing monitoring programs: iterative ﬂow diagram for deﬁning a monitoring study strategy. (From Boesch, D.F. et al., Managing Troubled Waters: The Role of Marine Environmental Monitoring, National Academies Press, Washington, D.C., 1990. Reprinted with permission from the National Academy of Sciences. Courtesy of the National Academies Press, Washington, D.C.)

L1641_Frame_C01.fm Page 6 Tuesday, March 23, 2004 8:55 PM

6 Environmental Monitoring

Develop Testable Questions

Quantify Variability

Identify Meaningful Levels of Changes Identify Logistical Constraints

Select What to Measure

Reframe Questions

Develop Monitoring Design

Specify Statistical Models Conduct Power Tests and Optimizations No Can Predicted Responses Be Seen?

Yes

Refine Technical Define Data Quality Design Objectives

Develop Sampling Design

No Is Design Adequate?

Yes

FIGURE 1.1B Designing and implementing monitoring programs: iterative ﬂow diagram for developing an environmental measurement design. (From Boesch, D.F. et al., Managing Troubled Waters: The Role of Marine Environmental Monitoring, National Academies Press, Washington, D.C., 1990. Reprinted with permission from the National Academy of Sciences. Courtesy of the National Academies Press, Washington, D.C.)

impacts ranging from power plants, outfalls, and marinas to ﬁshing, dredging, and natural events like storms and El Niño events. Another general but key aspect in the overall planning for monitoring design relates to data quality assurance.30,31 Boesch et al. highlight two aspects of quality

L1641_Frame_C01.fm Page 7 Tuesday, March 23, 2004 8:55 PM

Conceptual Basis of Environmental Monitoring Systems 7

assurance in the design of monitoring programs:30 quality control (QC) and quality assurance (QA). QC might be viewed as strategic in nature since it is intended to “ensure that the data collected are of adequate quality given study objectives and the specific hypotheses to be tested.”30 QA is somewhat more “tactical” and deals with the everyday aspects of documenting sample analysis quality by repetitive measurements, internal test samples, use of standards and reference materials, and audits; specifically, sample accuracy and precision needs to be assessed, applied to data analysis and interpretation, and documented for reference. Standard Methods for the Examination of Water and Wastewater Analysis36 is a well-known reference source for QA and QC procedures in microbiological and chemical laboratory analyses. In addition, QA/QC concepts and procedures are well addressed and documented in Keith37 for a variety of multimedia environmental sampling methods. And finally, metadata (i.e., data about data) has emerged as a QC/QA component to monitoring programs during the last decade, given the emergence of relational databases and GIS for regular applications in environmental monitoring and assessment.31 Later chapters in this book deal with detailed aspects of QA/QC and metadata and related data management tools are briefly addressed subsequently in this chapter under Data Integration.

1.3 OVERVIEW OF SPECIFIC CONCEPTUAL MONITORING DESIGN COMPONENTS Conceptual components of our approach to environmental monitoring design (and application) have been detailed in papers by Wiersma and colleagues.21,23,27 These components at that time included (1) application of a conceptual framework as a heuristic tool, (2) evaluation of source-receptor relationships, (3) multimedia sampling of air, water, soil, and biota as key component pathways through environmental systems, and (4) use of key ecosystem indicators to detect anthropogenic impacts and influences. This conceptual approach was intended to help identify critical environmental compartments (e.g., air, water, soil) of primary concern, to delineate potential pollutant pathways, and to focus on key ecosystem receptors sensitive to general or specific contaminant or anthropogenic affects. Also implicit in this monitoring design is a watershed or drainage basin perspective17,18,38 that emphasizes close coupling of terrestrial–aquatic linkages within ecosystems. Figure 1.2 summarizes these overall components of our approach,27 especially at our remote monitoring sites in Chile, Wyoming, and the Arctic Circle (Noatak site). Remote sites were utilized for baseline monitoring and testing of design criteria and parameters. These sites were less impacted by local or regional sources of pollution or land use change and were expected to be more indicative of baseline conditions (in the context of natural variation and cycles) that might best serve as an “early warning” signal of background global environmental change.18 In addition, field logistics were pronounced and rigorous at these remote sites for any type of permanent or portable monitoring devices and instrumentation. These conditions served as a good test of the practical limits and expectations of our monitoring design components.

L1641_Frame_C01.fm Page 8 Tuesday, March 23, 2004 8:55 PM

8 Environmental Monitoring

Integrated Ecosystem and Pollutant Measurements

Air quality Forest ecology Oxides of nitrogen and sulfur Growth rates Particulates (metals, sulfates, nitrates) Decomposition rates

Water quality Major ions Aquatic ecology Nutrients and metals Benthic communities Plankton communities Sites: Torres del Paine NP, Chile (United Nations) Wind River Mountains, WY (Forest Service) Noatak National Preserve, AL (International Man and Biosphere Program) DOE Research Parks: Historical Data Analysis/Monitoring Design

FIGURE 1.2 Conceptual design for global baseline monitoring of remote, wilderness ecosystems. (From Bruns, D.A., Wiersma, G.B., and Rykiel, E.J., Jr., Ecosystem monitoring at global baseline sites, Environ. Monit. Assess., 17, 3, 1991. With permission from Kluwer Academic Publishers.)

Our overall monitoring design concept (Figure 1.2) also served as a basis for evaluating historical monitoring data from seven DOE National Environmental Research Parks. This conceptual assessment highlighted the need and opportunity inherent in geospatial technologies and data like Geographic Information Systems (GIS), satellite remote sensing imagery (RS), and digital aerial photography. In conjunction with the report by Wiersma et al.,31 this DOE monitoring design assessment27 facilitated start up of the GIS watershed research program and GIS Center in the GeoEnvironmental Sciences and Engineering Department at Wilkes University.20 In addition, this general conceptual monitoring approach was used for: a regional land use plan for 16,000 acres of abandoned mine lands,19,20,39 a successful community- based proposal to designate a regional watershed as an American Heritage River (AHR, see www.epa.gov/rivers/98rivers/), a National Spatial Data Infrastructure Community Demonstration Project (www.fgdc.gov/nsdi) and recipient of a U.S. government Vice Presidential “Hammer Award” (www.pagis.org/CurrentWatershedHammer.htm), and a GIS Environmental Master Plan for the Upper Susquehanna–Lackawanna River.40 Figure 1.3 provides additional overall conceptualization of our monitoring design for remote wilderness ecosystem study sites. This heuristic tool22,41 highlights the atmospheric pathway for anthropogenic impacts to remote ecosystems and indicates the multimedia nature of our monitoring efforts based on ﬁeld tested protocols evaluated in our remote site research program.25,27 Details of this conceptual component of our monitoring design are provided below.

L1641_Frame_C01.fm Page 9 Tuesday, March 23, 2004 8:55 PM

Conceptual Basis of Environmental Monitoring Systems 9

Terrestrial Groundwater Fauna

Wet

Atmosphere Dry Vegetation Short- and Long- We Wet Dr t Surface Range y Sources Dry Water Soil Micro-, Litter / Macro- Humus Flora/Fauna

Aquatic Micro-Macro- Mineral Soil Flora/Fauna

Deeper Soil Sediment

FIGURE 1.3 Systems diagram and heuristic tool for conceptualization of monitoring design for sensitive wilderness ecosystems. (From Bruns, D.A., Wiersma, G.B., and Rykiel, E.J., Jr., Ecosystem monitoring at global baseline sites, Environ. Monit. Assess., 17, 3, 1991. With permission from Kluwer Academic Publishers.)

The NRC report by Wiersma et al.31 signiﬁcantly broadened our general conceptual approach to integrated environmental monitoring systems. This report’s focus on combining diverse (including multimedia) environmental data sets and the extensive geographic spatial extent of two of the case studies (the ISLSCP example noted above, and use of remote sensing for drought early warning in the Sahel region of Africa) resulted in adding two additional components19 to our design concepts: data integration with geospatial tools like GIS and remote sensing, and a landscape spatial scaling component, based in part again on GIS, but especially in the context of a hierarchical approach to ecosystems.42

1.4 CONCEPTUAL MONITORING DESIGN COMPONENTS We have tested and evaluated different aspects of our monitoring design concepts depending on a range of criteria, including study site location and proximity, degree of local and regional pollutant sources and land use perturbations, funding agency and mission, duration and scope of the study (funding limitations), and issues of degree of spatial and temporal scaling. Our work at the Wyoming and Chile sites has been proﬁled and described in several contexts: global baseline monitoring,25,27 freshwater ecosystems and global warming,18 and testing and evaluation of agency (U.S. Forest Service) wilderness monitoring protocols for energy development assessment.17,19 These are both remote, wilderness monitoring sites with the Torres del Paine Biosphere Reserve in southern Chile being one of the “cleanest” (and least disturbed globally), remote study areas from an atmospheric pathway; in contrast,

L1641_Frame_C01.fm Page 10 Tuesday, March 23, 2004 8:55 PM

10 Environmental Monitoring

the Wyoming site was “downwind” of significant ongoing and potential atmospheric emissions (oxides of sulfur and nitrogen) from regional energy development. Numerous multimedia parameters were measured and evaluated at the Wyoming site, especially from a sampling protocol perspective. Details of this work and descriptions of the study site are provided in Bruns et al.17,19 who focus on five evaluation criteria for monitoring design and implementation: ecosystem conceptual basis, data variability, uncertainty, usability, and cost-effectiveness. This study site gives the best perspective and detail on a wide range of monitoring parameters in our remote site work and serves as one of three (there is another remote site in southern Chile; for details on site conditions, see References 18, 25, and 27) examples of our conceptual approach to monitoring design. The third study site for a basis of contrast and comparison to our remote sites is in northeastern Pennsylvania. This represents a 2000-square-mile portion of a watershed designated in 1998 by President Clinton as one of 14 American Heritage Rivers. A GIS watershed approach was employed for research in monitoring and assessment with geospatial tools to address environmental impacts from urban stormwater runoff, combined sewer overflows, acid mine drainage, impacts from abandoned mining lands, and regional suburbanization and land use change. Cleanup and reclamation costs for mining alone approach $2 billion, based on Congressional hearings in 2000.40 As noted above, this site provides more perspective on geospatial tools and scaling issues vs. our earlier monitoring work at remote sites.

1.4.1 CONCEPTUAL FRAMEWORK AS HEURISTIC TOOL This component is generally considered as the starting point in monitoring design. It is not intended as a static or stand-alone element in the monitoring program. As a simple “box-and-arrow” diagram, it serves as an interdisciplinary approach to examine and identify key aspects of the monitoring program being designed. Basi- cally, principal investigators and their technical teams, along with responsible program managers and agencies, often across disciplines and/or institutions, can take this simple approach to focus discussion and design on answering key questions: Is the study area of concern being potentially impacted by air or water pollution sources? What are the relative contributions of point vs. nonpoint sources of water pollution? What are the primary pollutant pathways and critical ecosystem components at risk? How are critical linkages between ecosystem components addressed and measured? What is the relative importance of general impacts like land use change vs. media specific impacts like air, water, or subsurface (e.g., landfills) contamination sources? Figure 1.2 and Figure 1.3 as noted above illustrate the atmospheric route as primary disturbance and pollutant pathway to remote ecosystems such as those at our Wyoming and Chile study sites. In these cases, “wilderness” or “national park” status prevent immediate land use perturbations but atmospheric pollutants, either as a potential global background signal (e.g., particulates associated with “arctic haze”) or from regional point sources like coal-fired power plants,27 might be transported long distances and may affect remote ecosystems via wet and dry deposition processes.43,44

L1641_Frame_C01.fm Page 11 Tuesday, March 23, 2004 8:55 PM

Conceptual Basis of Environmental Monitoring Systems 11

As expected at remote monitoring sites, field logistics and/or regulatory restrictions on available sources of electricity or weather protection may prohibit a number of instrument approaches to monitoring methods and techniques. Figure 1.3 shows shaded components (soil, water, vegetation, and aquatic community) of the monitoring program that are easily measured with simple field sampling devices and procedures. Forest survey methods, soil sampling trowels, and aquatic kick nets allow for rapid field assessments and sampling as field restrictions or time limits may dictate. We have also used this approach successfully in even more remote sites like the Noatak Biosphere Reserve in the Arctic Circle of Alaska18,27 and in a mountainous “cloud forest” ecosystem of Fan Jing Shan Biosphere Reserve in south central China.19 At the Wyoming remote monitoring site, metals in vegetation (terrestrial mosses), aquatic macroinvertebrates, and stream (water chemistry) alkalinity all scored highest across our five evaluation criteria noted above.19 Figure 1.4 shows a similar “systems diagram” developed for the northeastern Pennsylvania study site with a major focus on regional mining impacts. The eastern anthracite (coal) fields of Pennsylvania cover a general area of about 3600 mi2, with about 2000 mi2 directly within the Susquehanna–Lackawanna (US-L) watershed study area.40 The watershed covers about an 11-county area with over 190 local forms of government or agencies and supports a regional population base of over 500,000 people. Due to the broad spatial extent of these impacts and complex set

Birch Locusts Aspen Groundwater

Atmosphere Vegetation

Culm and AMD in Mining Waste Piles Streams

Aquatic Mineral Soil Micro-Macro- Flora/Fauna

Mine Pool Iron Oxides

FIGURE 1.4 A GIS watershed systems approach to monitoring and assessment of regional mining impacts in Northeastern Pennsylvania. (From Bruns, D.A., Sweet, T., and Toothill, B., Upper Susquehanna–Lackawanna River Watershed, Section 206, Ecosystem Restoration Report, Phase I GIS Environmental Master Plan, Final Report to U.S. Army Corps of Engineers, Baltimore District, MD, 2001.

L1641_Frame_C01.fm Page 12 Tuesday, March 23, 2004 8:55 PM

12 Environmental Monitoring

of environmental conditions, we have employed a major geospatial (GIS based) technological approach to monitoring and assessment at this site.40,45 However, even here, the basic box-and-arrow diagram served a number of useful applications. First, we assembled an interdisciplinary team of almost 20 members from various institutions and state and federal agencies. Hydrologists, geochemists, river ecologists, GIS technicians, plant ecologists, soil scientists, and engineers were represented for a one-day workshop on which these concepts and components were proposed, discussed, evaluated and agreed upon as a GIS watershed approach to regional monitoring and assessment. Second, the general elements implicit in this conceptual framework allowed for scaling from local, site-specific and stream-reach applications (e.g., well-suited to local watershed groups) to broader watershed and landscape spatial scales (e.g., see the U.S. Environmental Protection Agency’s (EPA) GIS Mid-Atlantic Integrated Assessment over a five-state region4). Our watershed monitoring research with federal sponsorship (e.g., EPA and U.S. Department of Agriculture [USDA]) facilitated our use of GIS, RS, aerial photography, and the Global Positioning System (GPS)—all of which are not generally available to local watershed groups or local branches of relevant agencies. Therefore, we avoided duplication of field measurements at a local level and instead focused on a watershed (and sub-catchment) approach with GIS. We were able to coordinate with local groups in public meetings and technical approaches due to a common conceptual design of the environmental system. Third, a systems diagram of this nature also facilitated data analyses among key components, the pollutant sources, and the affected elements of the watershed and landscape. For example, the diagram in Figure 1.4 was used in setting out our statistical approach for prioritizing watershed indicators of potential use and identifying stream monitoring parameters for ranking of damaged ecosystems.40 This also allowed us to incorporate land use and land cover databases derived from satellite imagery and integrate it with point samples of water (chemistry) quality and stream community biodiversity via statistical analysis (Figure 1.5).

FIGURE 1.5 Statistical analysis of stream biodiversity vs. watershed area in mining land use.

L1641_Frame_C01.fm Page 13 Tuesday, March 23, 2004 8:55 PM

Conceptual Basis of Environmental Monitoring Systems 13

And fourth, both our EPA- and USDA-sponsored GIS research projects maintained a public outreach and environmental education component. The basic systems diagram shown in Figure 1.4 successfully enhanced our educational component in this regard, both with other technical participants, and, especially, with the public. These outreach and education activities involved various public meetings for dis- l cussion, input, and coordination, in addition to posting of information, data, GIS analysis, and environmental concepts at our Website (www.pagis.org) for GIS l research on the US-L watershed. Also, we participated in environmental education activities as part of public community riverfront park activities and high school student visits to campus for brieﬁngs on GIS, watershed analyses, and geospatial data applications.

1.4.2 EVALUATION OF SOURCE–RECEPTOR RELATIONSHIPS This element in the conceptual design of a monitoring program is also implicit in the diagrams of Figure 1.2 to Figure 1.4. However, this component also mandates an interdisciplinary approach to environmental monitoring since soil scientists, forest ecologists, and stream ecologists may not have typical expertise in various techniques of water, air, and soil pollution monitoring. Likewise, environmental engineers involved with the design of air and water pollution control and monitoring technologies may lack the needed ecological expertise for identifying and measuring the response of critically sensitive ecosystem receptors or endpoints. Our work at remote sites in Wyoming and Chile beneﬁted from a technical team approach since our research was sponsored during our employment with a DOE national laboratory where the necessary interdisciplinary mix of expertise was readily available to support work on remote site monitoring (e.g., see References 25 and 46.) For example, we had ready access to various scientists and staff with expertise in soils, forestry, river ecology, geology, air pollution, analytical chemistry, and general environmental science and engineering through various technical programs and organizations at the Idaho National Engineering Laboratory. For the Pennsylvania GIS watershed project, similar concerns and issues were addressed. In this case, expertise in mining, engineering, geochemistry, hydrology, GIS, and stream ecology was derived through public and state and federal agency outreach during the public sector portion of the long-term project. However, the ultimate selection of key pollutant sources and critical ecosystem receptors needs to be well focused, since both remote site and watershed approaches often end with long lists of parameters for potential implementation in a monitoring program. In this situation, logistics, ﬁnancial resources, and funding limitations require a subset of measurements that will allow assessment of the more important relationships. In some situations, peer-review by an outside panel,47 case study reports,31 or actual testing and evaluation of a range of parameters17,19 may help to resolve differences or professional preferences and result in a more cost-effective but focused set of monitoring parameters. More examples are discussed below for the other design components. Finally, it should be noted that data provided in Figure 1.5 represent one example of the successful selection of a key source of pollution with a sensitive ecological endpoint. In this case, the extent of mining disturbed lands within a watershed (or

L1641_Frame_C01.fm Page 14 Tuesday, March 23, 2004 8:55 PM

14 Environmental Monitoring

subwatersheds or “sampling” catchments) represented a major source of pollutant (and land use) impact—but derived from satellite imagery and ground-truthed with GPS.45 Disturbed mining lands are devoid of natural vegetation and soil horizons and are susceptible to extreme amounts of sediment loading to streams and rivers where aquatic habitats are destroyed due to sedimentation processes (see literature review40). In addition, atmospheric exposure of pyritic mining wastes can generate a considerable amount of acid mine drainage to streams via the hydrologic cycle, so additional geochemical impacts are evident due to the eco-toxicity of high acidity and mobilized heavy metals like Cu and Zn from waste materials. Part of our research in this GIS watershed study was to determine what water chemistry and stream biotic variables would be best associated with regional mining impacts. The biodiversity of stream macro- invertebrates40 was one of the better indicators in this regard as shown in Figure 1.5. As noted previously, stream macroinvertebrate parameters scored well over the ﬁve monitoring evaluation criteria employed for the Wyoming remote study site.17,19

1.4.3 MULTIMEDIA MONITORING The rationale for monitoring various environmental media encompassing air, water, soil, and biota is based on several factors. First, the physical and chemical properties of pollutants demonstrate a wide range of fate and transport mechanisms with different pathways and effects upon ecological receptors. This is supported both by multimedia modeling approaches48 and general estimation methods in ecotoxicology and environmental chemistry.49,50 Second, focused population and community studies on the fate of metals and organic contaminants relative to bioaccumulation and trophic-food web transfer pathways49,50 also indicate a need to approach monitoring design from a multimedia perspective. And third, larger-scale watershed and regional landscape investigations of particular pollutants like acid rain effects on freshwater ecosystems51 and air pollution impacts to forests52 should reinforce this design component if resource managers are to understand the fate and effects of pollutants in a holistic ecosystem framework. Methods of sampling and analysis on a multimedia basis are well established53 and detailed elsewhere in this volume. Our research on multimedia monitoring design has emphasized the testing and evaluation of methods for use in remote, wilderness ecosystems.25,27 Table 1.1 lists monitoring parameters of various physical, chemical, and biological characteristics of a high-elevation ecosystem in Wyoming from a multimedia standpoint, and includes a cataloging of appropriate methods for use under potentially harsh ﬁeld conditions. As indicated above, we have developed evaluation criteria for assessing the overall utility of these methods and the reader is referred to other reports and publications for more detailed consideration.17,19,46

1.4.4 ECOSYSTEM ENDPOINTS The search for key ecosystem parameters for environmental monitoring and assessment has received considerable attention over the past two decades. Earlier studies, more aligned with environmental toxicology research or assessment of sewage pollution in streams, focused on population inventories or surveys of “indicator” species. Indicator

Conceptual Basis of Environmental Monitoring Systems 15

TABLE 1.1 Integrated Multimedia Monitoring Parameters at the Wind Rivers Study Site

Measurement Method (Previously Evaluated)

Abiotic Measurements 54 SO4, NO3, HNO3, NH3, NO2, SO2 (atm) Transition ﬂow reactor (ﬁlter pack) — EPA Source-term particle analysis Scanning electron microscopy with energy dispersive x-ray analysis Ozone UV Photometry Meteorological parameters Standard sensors; plus dry depostion methods of Bruce Hicks/Oak Ridge National Laboratory Trace metals (atm) Low and high volume sampling Trace metals (in water, litter, soil, vegetation) Ecological sampling at study site25,55 Trace metals in snow Snow cores before runoff (later analysis with Standard Methods36) Soil (organic matter, exchange bases, and acidity, U.S. Forest Response Program47,56 pH, extractable sulfate) 57 NO3, PO4, SO4 (water) National Surface Water Survey Lake/streams water chemistry (cations and anions) National Surface Water Survey57

Biotic Measurements Lake chlorophyll a, zooplankton, benthic algae, U.S. Forest Service Wilderness Guidelines47 and ﬁshes, benthic macroinvertebrates Standard Methods36 Stream ecosystem analysis (macroinvertebrate River Continuum Concept38,58–60 functional feeding groups, periphyton, decomposition, benthic organic matter) Terrestrial (forest) ecosystem ˇ (productivity, Dr. Jerry Franklin; U.S. Department of needle retention, needle populations, litter Agriculture, Forest Service methods61 decomposition, litterfall, foliage elemental composition, community structure)

Source: From Bruns, D.A., Wiersma, G.B., and Rykiel, E.J., Jr., Ecosystem monitoring at global baseline sites, Environ. Monit. Assess., 17, 3, 1991. With permission from Kluwer Academic Publishers.

species may include those that are either tolerant (e.g., tubificids or “bloodworms” thrive at low levels of oxygen due to organic loading of aquatic systems) or intolerant (e.g., various species of mayflies and stoneflies that require more pristine conditions of stream habitat and associated chemical constituencies) of pollutant concentrations or habitat disruptions.50 However, Cairns62 has cautioned against reliance on single indicator species since their known response is often in relation to very particular kinds of pollutants and may not warrant objective assessment of general or varied contaminant impacts. In this context, Schindler63 has suggested that some individual species, like the crustacean Mysis relicta, may represent unique keystone species within aquatic food webs (i.e., occupying specialized niches) and are susceptible to a variety of stresses. In this case, a monitoring program would be more effective with the

16 Environmental Monitoring

inclusion of such a species, assuming resource managers have access to previous knowledge and available data for decision support in the design process. At present, ecological measurement and assessment methods encompass a hierarchical framework to ecosystem management. This seems due to the maturation of ecotoxicology as a science64 along with further developments of environmental monitoring principles. For example, ecotoxicology texts50 generally cover pollutant effects on individual organisms, populations, and communities, and ecosystem structure and function, and in some cases use an explicit hierarchical treatment to basic and applied concepts. In addition, resource managers are now responsible for and concerned about whole ecosystems—and the monitoring programs need to support these assessment objectives. Thus, scientists are being called upon to address multiple stresses on ecosystems.65 In this context, a hierarchical approach represents the best available conceptual framework for dealing with complexities of both ecosystems and associated impacts of pollutants and physical disturbance to environmental systems such as changing land use. Allen and Starr66 and O’Neill et al.67 originally developed the basic concepts of a hierarchical approach to understanding ecosystems. In its basic form, different levels of biological organization were recognized in a hierarchical fashion, with increasing degrees of ecological complexity. This hierarchy for either aquatic or terrestrial systems, from lowest to highest, included the following levels of biological organization: individual organisms (e.g., plants or wildlife), populations, communities, and ecosystems. The widespread use and availability of geospatial tools and data, like geographic information systems (GIS) and satellite remote sensing imagery, has facilitated further development of the hierarchical ecosystem concept that extends the scope of watershed and landscape spatial (and temporal) scales.19,31,33 However, these aspects are covered below in the next two components in the conceptual design of monitoring systems. A review of the ecological assessment literature (Table 1.2) indicated the use of a range of parameters across these various levels of ecological complexity.68 Such measurements encompass biomass for a population (e.g., trout), biodiversity (e.g., stream macroinvertebrates) as an indicator of community structure, and nutrient cycling (e.g., water chemistry) as an integrator of ecosystem function. In general, the most common parameters have included trophic relationships, species diversity, succession (temporal changes in composition), energy ﬂow, and nutrient cycling. Some investigators51,63 have indicated that functional responses of ecosystems may be more robust than structural changes due to “functional redundancy” and variation in pollutant sensitivity among species; for this reason, individual species and community level monitoring has been recommended for detecting ecological impacts. Our approach to ecosystem monitoring17,19,25,27 has been to include both structural and functional parameters for terrestrial and aquatic habitats and environments on a watershed basis (see Table 1.1 and Figure 1.2, Figure 1.3, and Figure 1.4 above). At present, few studies and monitoring programs have produced long-term data on both structure and function,63 and more research is needed before deﬁnitive guidelines can be set. Also, we have observed extreme impacts from land use change, like regional mining and urban stormwater runoff; while structural changes in these cases are more easily measured in the early stages of impact, we expect that in these

Conceptual Basis of Environmental Monitoring Systems 17

TABLE 1.2 Ecological Parameters: Recommendations for Monitoring and Assessment of Baseline Conditions and Human Impacts

Author(s) 69 70 71 72 73 74 75 67 76 77 78 79 80 Parameter Abundance ++++ +(a) + (biomass) Reproduction ++++ +(a) Behavior +++

Community Structure Trophic +++++−+ +(a) ++++ relationships Species diversity ++++++++ + Succession/ +++++++(a) +++ change in composition Size ++ relationships

Ecosystem Function Energy ﬂow ++++++−+ −(a) ++ +(t) Nutrient cycling +++++++(a) −(a) ++ + Decomposition/ +++++ +(a) −(a) ++ + respiration Biomass/nutrient +++ + + pools

Notes: (a) = aquatic ecosystem; (t) = terrestrial ecosystem; + = good potential for monitoring and assessment; and − = robust; not indicative of early impacts or stresses. Plus signs reﬂect generally positive view of a particular parameter as a key indicator of impact or at least its potential utility to detect anthropogenic perturbations. A negative sign means that an author has found this parameter to be a poor indicator of ecological impact; these parameters were found to be too robust and were not very sensitive to impacts.

Source: Bruns, D.A. et al., An ecosystem approach to ecological characterization in the NEPA process, in Environmental Analysis: The NEPA Experience, Hildebrand, S.G. and Cannon, J.B., Eds., Lewis Publishers, Boca Raton, FL, 1993, 103. With permission.

extreme cases, functional changes are needed to deﬁne the total system collapse that warrants immediate attention to ecosystem restoration and pollution mitigation. Also, our experiences in remote site monitoring concurrently for aquatic (streams and lakes) vs. terrestrial (forests) systems suggests that functional measures like forest productivity, litterfall, and leaf decay rates will better reﬂect short-term impacts of atmospheric than compositional changes, given the long life cycle of most tree species vs. short-lived aquatic species.

18 Environmental Monitoring

General reviews of the monitoring literature63,68,71,74,76 indicate that a number of important ecological impacts can be measured and assessed only at the ecosystem level. In addition, measuring many different parameters is not necessarily the optimal strategy for designing and implementing a monitoring program. In most cases, a selected subset of parameters can be deﬁned from a conceptual basis and principles as outlined above, viewed in conjunction with knowledge of the published literature.

1.4.5 DATA INTEGRATION This is one of the major challenges to implementing a well-designed monitoring program and cuts across all of the other components of our systems approach. Generally, this aspect of a monitoring program and its practical utility in the “real world” will be limited to the extent that these other components are ignored, relegated to a minor role, or inadequately developed or addressed. A conceptual model or framework with clearly identiﬁed sources of pollution, their pathways, and likely environmental endpoints provides the broad overview and context within which data sets will be processed, summarized, and evaluated (i.e., “data fusion”; see Wiersma 31 l et al. ). This framework will provide an initial set of testable hypotheses for trend analysis and the inference of potential effects on ecosystems from point pollution sources and/or more diffuse impacts from nonpoint sources that may include changing land use over larger environmental extents and spatial scales. Actually collecting multimedia data and measuring key sensitive ecosystem endpoints are needed if resource managers are to manage, protect, and sustain environmental systems in a holistic fashion. The NRC report by Wiersma et al.31 provides a comprehensive review of issues associated with fusing diverse sets of environmental data. The authors review a series of case studies that encompass predicting droughts in the Sahel, atmospheric depo-

sition in the U.S., the U.S. CO2 program, ISLSCP (noted above), and marine fisheries. Numerous recommendations are provided, based on practical problems encountered from specific case study programs. These include organizational, data characteristics, and technological impediments to data fusion efforts. In this chapter, we focus on the challenge of data integration with a selected view toward aspects of data characteristics and geospatial technologies identified in Wiersma et al.31 Organizational challenges, like agency mission, infrastructure, and coordination, are equally important but beyond the technical scope of this publication. The reader is referred to the original NRC report for more detailed information and insight into organizational factors in environmental monitoring design. Geospatial technologies like GIS are emerging as the major approach to data fusion efforts, ranging from “enterprise GIS” in the business world to the “geodatabase” model in environmental management systems (www.esri.com and see Ref- erence 81). The NRC report recommended using GIS and related technologies, like the GPS and satellite remote sensing imagery, in environmental monitoring and management programs to facilitate data acquisition (at various scales, see text below), data processing and analysis, and data dissemination to resource managers, political leaders, and the public. Concurrent with the NRC panel proceedings and

Conceptual Basis of Environmental Monitoring Systems 19

publication, an applied GIS watershed research program19,20,45,82 was being planned and implemented for the US-L—nationally designated in 1998 as 1 of 14 American Heritage Rivers. Four geospatial technologies were incorporated into this evolving program based in part on recommendations of the NRC report and geospatial data inventories at DOE national laboratories.27 These aspects are highlighted here to demonstrate one approach to data fusion efforts in the spirit of the NRC report. Figure 1.6 showcases how GIS is used to organize and integrate diverse environmental data sets to help solve environmental problems associated with past and on-going practices in land use. In particular, there is a $2-billion land reclamation and ecosystem restoration problem from over 100 years of regional coal mining. In addition, this watershed of 2000 square miles covers 196 local governments where urban stormwater runoff and combined sewer overflows (CSOs) have resulted in a $200 to 400 million aquatic pollution cleanup issue.19,20,40 Sampling of stream and river sites was of high priority given the nature of these environmental impacts to aquatic chemistry, habitats, and ecological communities (Fig- ure 1.4 and Figure 1.6). GPS was used to locate each site and delineate point source features (mine water outfalls or CSOs) of pollution. Although field sampling techniques were “low-tech” based on standard methods, all ecological data of this type were easily integrated into a relational data base as part of the GIS for the watershed. In addition, sampling sites integrated into the GIS allowed for delineation and digitization of sampling site subcatchments for data analysis and integration from a comparative watershed perspective. For example, Figure 1.6 also shows the utility of GIS in visualizing data on a comparative basis between two watersheds (GIS graphic charts in lower left of figure). The subwatershed in a rural setting with no mining had high-stream macroinvertebrate biodiversity (clean water species), low acidity, and land cover mostly in forests and grassland meadows and minimal development vs. a mining watershed with more than 30% of land cover as mining disturbed areas, and with only pollution-tolerant aquatic species and high acidity in surface water streams. In our Heritage River study area and region, satellite imagery (the Mid-Resolu- tion Land Characteristics or MRLC, e.g., see Reference 83) was processed for land cover to facilitate watershed characterization for relating land use practices and problems to ecological conditions along environmental gradients within the watersheds.40 Figure 1.5 and Figure 1.6 demonstrate one approach we used to data fusion by statistically relating stream biodiversity measures to mining land use (SPOT imagery shown in middle inset of Figure 1.6) within 18 delineated subwatersheds: (1) we used GPS to identify and locate point sampling sites on stream segments, (2) we digitized subwatersheds above each sampling point with a GIS data set of elevation contour lines, and (3) we processed SPOT imagery for land cover and land use84,85 and conducted extensive ground-truthing of classified imagery with GPS.20,45 Land use impacts to ecological systems are generally viewed to be as widespread and prevalent worldwide to warrant a higher risk to ecosystems than global warming.8–10 Satellite imagery also allows for a range of landscape4 and watershed indicators33 to be calculated for environmental monitoring and assessment at a broader spatial scale (see next section). Vogelmann et al.86 surveyed data users of Landsat Thematic Mapper data (known as the National Land Cover Data set, or NLCD) from the early 1990s and found 19 different categories of application including land

Conceptual Basis of Environmental Monitoring Systems 20 ver watershed: GIS, GPS, remote watershed: ver Integrated use of geospatial technologies for environmental monitoring on the heritage ri use of geospatial technologies for environmental Integrated sensing imagery, and orthoimagery. sensing imagery, FIGURE 1.6 FIGURE © 2004 by CRC Press LLC L1641_Frame_C01.fm Page 21 Tuesday, March 23, 2004 8:55 PM

Conceptual Basis of Environmental Monitoring Systems 21

cover change assessment, hydrologic-watershed modeling, environmental impact statements, water quality and runoff studies, and wildlife habitat assessments. In addition to the integrated use of GIS, GPS, and remote sensing imagery in our PA Heritage River watershed research project (Figure 1.6), we have employed digital aerial photography as the fourth geospatial data source and technology.20,40,84 Also known as orthoimagery, these data have been identified by the U.S. Federal Geographic Data Committee (FGDC)87 as one the fundamental “framework” geo- data sets for the National Spatial Data Infrastructure in the U.S. In this context, we surveyed 196 different local governments and regional state and federal agency offices within the 2000-square-mile watershed of the US–L River and found 10 of 11 counties lacking in local scale orthoimagery needed for tax assessment, land use and planning, emergency management, environmental cleanup, land and deeds records, ecological protection and monitoring, and floodplain management (see GIS watershed plan40). Falkner88 provides an overview to methods and applications of aerial mapping from orthoimagery. Applications include mapping of geographically extensive wetlands,89 cartographic support to management of state aquatic resources,90 and floodplain management.40 A final element to data integration is the importance of QA and QC for the data sources themselves, along with metadata on all aspects of data development, processing and integration, and analysis.31 Methods of multimedia field sampling and laboratory analysis (see references in Table 1.1) generally deal with adequate and established standard procedures of accuracy and precision (see also Reference 37 for general QA/QC issues). In contrast, geospatial metadata methods are still in various stages of development. GPS is generally accepted for most environmental applications in field mapping and is now commonly used for on-board aerial photography88 and later aerotriangulation and accuracy calculations that require positional data as a replacement to conventional ground control surveys. In turn, either GPS91 or accurate, geo- referenced orthoimagery83,86 may be used in accuracy assessments of remote sensed data classified for land use and land cover. Bruns and Yang45 used GPS to conduct regional accuracy assessments on four such databases used in landscape–watershed analyses and reviewed general methods of accuracy assessment.92,93

1.4.6 LANDSCAPE AND WATERSHED SPATIAL SCALING The scope and extent of environmental contaminants in ecosystems, their potential for long-range transport through complex pathways, and their impact beyond simply local conditions, all dictate that environmental monitoring programs address pollution sources and effects from geographically extensive landscape and watershed perspectives. Although we have only recently added this ﬁnal component to our conceptual design for monitoring systems,19 there has been well over a decade of ecological research that serves as a foundation for successful inclusion of this element in monitoring programs. The success of this approach is supported from several standpoints. As indicated above, landscape ecology has been well developed and investigated as part of a hierarchical perspective to ecosystem analysis.42,66,67,94,95 Landscape parameters and indicators include dominance and diversity indices, shape metrics, fragmentation

22 Environmental Monitoring

indices, and scale metrics, and are routinely incorporated into natural resource management texts on GIS and the emerging field of landscape ecotoxicology.96 In a similar fashion, a hierarchical approach to spatial scales in environmental analysis42 has been developed for both terrestrial and aquatic ecosystems, usually on an integrated basis relative to either a landscape or watershed context. Hunsaker and Levine97 used GIS and remote sensing of land use in a hierarchy of 47 watersheds to assess water quality in the Wabash River System in Illinois. In this study, water quality monitoring sites were linked to their respective watershed segment in the hierarchy to address issues of terrestrial processes in the landscape and evaluate their relevance to environmental management practices. This GIS and hierarchical approach facilitated identification of water quality conditions at several spatial scales and provided resource managers with tools to enhance decision support and data maintenance. O’Neill et al.33 recommended the use of GIS and remote sensing data, along with recent developments in landscape ecology, to assess biotic diversity, watershed integrity, and landscape stability. These authors presented GIS watershed integrity results for the lower 48 states on the basis of 16 U.S. Geological Survey Water Resource Regions. In general, GIS and remote sensing imagery have strongly facilitated a hierarchical approach to spatial scale and watershed analysis. This has been due, in part, to the better availability of geospatial data and technology, but this also is based on the relevancy of these regional environmental assessments for broad geographic extents.33,97 A spatial hierarchy to watersheds has been employed in four other examples relevant to design principles for environmental monitoring. In the first example, Preston and Brakebill98 developed a spatially referenced regression model of watershed attributes to assess nitrogen loading in the entire Chesapeake Bay watershed. These investigators used the EPA River Reach File to generate a spatial network composed of 1408 stream reaches and watershed segments for their regional analysis. From their GIS visual maps of the watershed, point sources of high nitrogen loading could be associated with specific urbanized areas of the Bay watershed and allowed the authors to acknowledge and identify large sewage-treatment plants as discharge points to stream reaches. In the second example, an Interagency Stream Restoration Working Group (15 federal agencies of the U.S.) recently developed a guidance manual for use in stream restoration99 based on a hierarchical approach to watersheds at multiple scales. This team recognized ecosystems at five different spatial scales from regional landscape to local stream reach, and stated that watershed units can be delineated at each of these scales—depending on the focus of the analysis and availability of data. Spatial scales of watershed ecosystems from the stream restoration guidance manual (Figure 1.7) were used in the GIS watershed master plan40 and served as the basis for our regional heritage river designation and approach to the first steps of assessing environmental conditions in the US-L watershed. The illustration of spatial scale shown in Figure 1.7 is based on an example of ecosystem hierarchy in the overall Chesapeake Bay watershed. This hierarchy was employed in our study design and tributary analysis of the US-L watershed40 that ranged from a regional landscape watershed to a local stream reach along a linear segment of stream or river corridor (see text below).

Conceptual Basis of Environmental Monitoring Systems 23

FIGURE 1.7 Ecosystems at multiple scales and used as the basis for regional to local GIS watershed analysis40 for the Upper Susquehanna–Lackawanna American Heritage River. (Modiﬁed from the Interagency Stream Restoration Manual, Interagency Team, Stream Res- toration Guidance Manual, Stream Corridor Restoration: Principles, Processes, and Prac- tices, Federal Interagency Stream Restoration Working Group (FISRWG) (15 federal agencies of the U.S. government), GPO Item No. 0120-A; SuDocs ISBN-0-934213-59-3, Washington, D.C., 1998.)

24 Environmental Monitoring

The third example is the EPA EMAP ecological assessment of the U.S. Mid- Atlantic Region, including the Chesapeake Bay watershed plus portions of other watersheds like the Ohio and Delaware rivers over a five-state area.4 These investigators developed 33 different watershed indicators of ecological conditions in 123 watershed units throughout the study area. This watershed assessment methodology is based on concepts developed earlier by O’Neill et al.33 and relies strongly on land use and land cover data derived from satellite imagery, along with a number of complementary geospatial datasets like population density, roads, and hydrography. The US-L watershed is one of the 123 watershed units from the EPA Mid- Atlantic regional study and we used their information to provide a regional context for our more detailed spatial analyses.40 Thus, our fourth example is from the Heritage River watershed study and comprehensive GIS environmental master plan40 where we conducted a GIS tributary analysis (on ten watershed indicators) and focused on 42 ecologically defined tributaries to the Susquehanna River (within the regional US-L watershed) including the river corridor segment of the mainstem. We highlighted results from five land cover classes (Multi-Resolution Land Character- istics or MRLC) that were evaluated in the watershed analysis. Of these, the mining land cover class was most effective (high) in detecting statistically significant impact differences between four (reference) rural watersheds without mining from another 12 tributaries with visibly observable mining/urban impacts.40 Agricultural land cover was rated “medium” in its effectiveness in this regard. It was generally difficult to use forest, urban, and wetland cover classes to differentiate between mining/urban and “non-mining” rural reference tributaries. Other watershed parameters developed specially to address our regional environmental problems included the number of CSOs in a tributary subwatershed, the number of acid mining outfalls in a subwatershed, iron loading rates (from outfalls), and hydrogen ion (acidity) loading (also from mining outfalls). Statistical tests indicated significant differences in ranks between watershed conditions for CSOs, hydrogen ion loading, iron (Fe) loading, AMD outfalls, and an index based on the average of all seven watershed indicators.40

1.5 SYNTHESIS AND FUTURE DIRECTIONS IN MONITORING DESIGN Our approach to monitoring design has been in continuous evolution, dependent on the growing awareness, concern, and the expanding scientiﬁc research and literature on regional environmental change, climate warming, and pollutants and land use change on a global basis.14 The ﬁrst four design components generally apply to our work at the southern Chile Biosphere Reserve, the Wind River Mountains high- elevation monitoring site, and the Pennsylvania Heritage River watershed. These components are based on long-founded interdisciplinary principles in the environmental sciences and ecology. For example, the conceptual framework of terrestrial-aquatic linkages within a river drainage basin38,58 has been a key feature in river ecology research since the mid-1970s. Likewise, the use of heuristic ecosystem diagrams focused on source-receptor pathways of pollutant contaminants has been applied to biosphere research for the United Nations Global Environmental Monitoring System (GEMS) since the early and mid-1980s.22,23

Conceptual Basis of Environmental Monitoring Systems 25

The third and fourth design components, multimedia monitoring and ecosystem endpoints, are based on well-established methods in pollution measurements and field ecology but require an interdisciplinary teamwork approach. Long-standing monitoring programs like the LTER program at NSF, the NAPAP program for acid rain in the U.S., and EPA’s EMAP program all exemplify these principles. In many ways, our work at the Wind Rivers site incorporated aspects of all of these programs17–19 but pushed the limits of standardized instrumentation and methodology since most of our field measurements occurred above 10,000 feet elevation in alpine and subalpine forests, streams, lakes, and watersheds. Technological innovations100 and data availability,4,87 NRC panel reports,30,31 and the general spatial extent and complexity of regional and global environmental change, especially global land use change,9,10,14 have all had a direct effect on our current focus on the last two recent components of an integrated conceptual approach to environmental monitoring. These encompass data integration with GIS and image processing software,84,85,101 and landscape-watershed spatial scaling based on remote sensing imagery (SPOT, Landsat) and a landscape sampling design.29,100–102 These have been more applicable to our research and work at the Chile and Pennsylvania sites since these have been ongoing, evolving, and have benefited from the availability of GIS, GPS, and remote sensing technologies. And finally, from an overall systems viewpoint, our approach has been influenced by and consistent with the landscape and watershed methods of O’Neill et al.32,33 and the watershed analyses of the EPA Mid-Atlantic Landscape Atlas.4 Given the opportunities inherent in geospatial technologies and GIS data for environmental monitoring,101 we are currently examining a number of watershed and landscape tools that would seem helpful to resource managers designing integrated monitoring programs. We predict that these and related GIS watershed and landscape software tools will facilitate ongoing and future efforts to monitor and assess environmental systems on a geographically extensive basis. These are briefly described below and highlighted in terms of their potential applicability to monitoring design considerations.

1.5.1 EPA BASINS Better Assessment Science Integrating Point and Nonpoint Sources (BASINS) is a multipurpose environmental analysis system for use by natural resource agencies to perform watershed- and water-quality-based studies. It was developed by EPA to address three objectives:

• To facilitate examination of environmental information • To support analysis of environmental systems, and • To provide a framework for examining management alternatives

BASINS103 is intended to support a watershed-based approach to environmental and ecological studies in a watershed context. As such, the system has been designed to be ﬂexible with a capability to support analysis at a variety of scales, using tools that range from simple to sophisticated. Comprehensive multimedia data sets compiled

26 Environmental Monitoring

by EPA are provided in BASINS and one can access and query the datasets through a GIS data mining tool; other assessment tools include TARGET and ASSESS which allow one to evaluate water quality concerns and impacts on either a watershed or sampling point basis, respectively. BASINS requires a GIS platform as a separate software component but makes ample use of GIS utilities for data integration, analyses, graphic visualizations, and modeling. A number of models are also available, mostly for point sources of pollution regarding industrial discharges to surface waters. BASINS has been used in various capacities as part of a comprehensive GIS environmental plan for the US-L watershed and has helped contribute to all design components of a community water quality monitoring program (see Reference 40 and EMPACT Website at wilkes.edu/~gisriver/).

1.5.2 SWAT The Soil Water Assessment Tool (SWAT) is now part of BASINS103 but was originally developed by the U.S. Department of Agriculture104 for management and assessment of agricultural and rangelands. SWAT is a GIS model that incorporates geospatial data on climate, rainfall, soils, slopes, and land cover and provides for a procedure to address changing land use (land cover) conditions in a watershed. Runoff and sediment yield from different patterns of land cover in a watershed are provided as model output105 and as such give the resource manager a better idea of source- receptor issues in the design of a monitoring program. In addition, the hydrologic model is useful for evaluating conceptual relationships on a watershed basis, facil- itating data integration, and predicting impacts to ecological endpoints.

1.5.3 CITYGREEN REGIONAL ANALYSIS CITYgreen is a GIS software tool for regional, local, and watershed–landscape analysis on the environmental function and economic value of trees and forests, especially in urban areas.106 This is an environmental planning tool capable of using detailed forest (and tree) stand data locally or regional satellite imagery classiﬁed for land cover. The software uses readily available data on soils (STATSGO), slopes (STATSGO), and regional rainfall zones and precipitation—all of which are provided with the software program and user’s manual. Model output includes runoff, carbon sequestration rates and storage (due to forests), and air pollution (sulfur dioxide, carbon monoxide, ozone, and nitrogen dioxide) removal potential due to deposition to forest vegetative surfaces. The software allows managers and planners to predict the outcome of various development scenarios by easily modifying (via user menus or software “wizards”) the amount and type of land use change for a delineated study area.107 CITYgreen appears to be a useful community planning tool that allows for data integration and especially predictive modeling of multimedia pathways and affects — both of which are useful to resource managers developing monitoring programs. The relative ease of operation for planning would seem to facilitate the iterative process necessary to monitoring design processes. We are currently starting analyses with CITYgreen to evaluate the potential for carbon sequestration if extensive re-forestation and ecosystem restoration efforts were to be implemented on regional areas of barren mining lands.

Conceptual Basis of Environmental Monitoring Systems 27

1.5.4 ATtILA ATtILA refers to an EPA-developed GIS software component known as “Analytical Tools Interface for Landscape Assessments.” This is a GIS software extension that allows users to easily calculate many common landscape metrics regardless of their level of GIS knowledge. Four metric groups are currently included in ATtILA108: landscape characteristics (e.g., percentage in forests or grasslands, landscape diversity, forest patch size), riparian characteristics (e.g., percentage of stream length adjacent to forests or urban land cover) human stresses (e.g., population numbers and change), and physical characteristics (e.g., average elevation, precipitation, total stream length). At present, ATtILA is available for R&D efforts mostly within the agency but we have examined the potential of ATtILA for application of landscape l and watershed indicators in environmental monitoring and assessment. Again, due to the nature of a GIS approach, data integration, conceptual relationships, and assessment of landscape and watershed ecological endpoints would seem to be appropriate beneﬁts to resource managers and researchers who need to address broad scale impacts from regional land use change and watershed modiﬁcation.

1.5.5 METADATA TOOLS AND WEB-BASED GIS As a final note to future directions, we point out the importance of GIS metadata tools and the potential of Web-based GIS for data integration and showcasing environmental relationships at different spatial scales — local to global. The latter will also be extremely important for dissemination of data to resource managers, government leaders, scientists, industry and business leaders, and the public at large. The importance of metadata for GIS and environmental monitoring in general has been identified by the NRC report of Wiersma et al.31 and briefly addressed above. ESRI’s (Environmental Systems Research Institute) ArcGIS software now provides a metadata tool consistent with FGDC geospatial data standards. Metadata are crucial to QA and QC issues with environmental data of all types but are necessary also for accessing and querying databases on the Web for incorporation into a monitoring program or for a GIS decision support system for monitoring design. ESRI’s ArcIMS (Internet Map Server) software and its Geography Network provide mechanisms for scientists, researchers, and resource managers to access a variety of environmental data for assessment, monitoring, and the design of a field measurement program. The reader is referred to the ESRI general Web page (www.esri.com) for additional information on software applications in this regard. Reports for the US-L American Heritage River watershed plan40 provide diagrams and software architecture design components for how a GIS Web-based decentralized data distribution system can be used for community environmental programs, including monitoring of water quality and the status of watersheds in the region.

1.5.6 HOMELAND SECURITY The events of September 11, 2001 have focused detailed attention to the vulnerability of the nation’s resources. An NRC report has recently addressed this issue109 that has become the top priority to federal, state, and local government leaders and

28 Environmental Monitoring

agencies, and the public at large. The NRC report highlights the need for a national program aimed at making the nation safer and outlines the technical elements necessary for homeland security. Geographic information systems (GIS) are identiﬁed as a critical component to this program within the context of information technology (IT) in general. Other reports by the National Center for Environmental Assessment (EPA web documents at http://www.epa.gov/nheerl/wtc/index.html) address the human and environmental risks associated with toxic chemical releases to the atmosphere associated with the destruction, combustion, and collapse of the New York World Trade Center towers. In the future, geospatial data and technologies are expected to be crucial analytical tools in both the collection of relevant environmental data and in making data available to emergency management personnel, government leaders, cleanup crews, environmental and public health managers and organizations, and the public at large. Many of the nation’s resources that are at high risk as identiﬁed in the NAS report on homeland security occur in rural communities, watersheds, and landscapes. h These include energy, nuclear plants, groundwater resources, surface waters, food, agricultural soils, utilities and corridors, and rural transportation arteries between major cities and urban centers. In addition, EPA has a “Strategic Plan for Homeland Security” and the agency is especially responsible for security issues110 regarding water resources, water supply, wastewater treatment facilities, and facilities of the chemical and energy industries. Although “Homeland Security” in its many complex facets is beyond the scope of our objectives in the paper, it should be indicated that the ecological consequences of intentional releases of chemical and biological agents warrant a comprehensive systems approach to monitoring the fate and effects of these contaminants in the environment. In some cases, monitoring methods and measurements have direct applicable from more conventional programs of environmental management but the conceptual basis and six components outlined here are still intended to be useful starting points to support these newer efforts as part of national security.

1.6 CONCLUSION The principles and concepts of environmental monitoring design are dynamic and iterative in nature. We have attempted to outline the key components of our approach to these concepts within the context of recommendations and reports of subcommit- tees on integrated monitoring sponsored by the NRC. The need has never been greater for integrated programs that collect, analyze, evaluate, and disseminate relevant environmental measurements at various scales on an ecosystem and multimedia basis. Local communities now grapple with issues of land use planning and change since most economic and environmental decisions are made at this local level of government. Such communities are often ill-equipped with the data and decision tools to make informed decisions on environmental health and the quality of life in general. State and federal agencies, research universities, and international organizations all maintain numerous environmental networks, but data may not be available to the general public or their use is beyond their resources for interpretation and decision making. Land use change is now viewed from a worldwide basis as

Conceptual Basis of Environmental Monitoring Systems 29

more prevalent and critical (at least short-term) than global climate warming.9,10 Other aspects of global environmental change14 have important implications for natural and managed ecosystems worldwide and provide critical challenges to those professionals charged with designing and implementing integrated environmental monitoring programs. Global dispersal and deposition of air contaminants, toxics in the global environment, loss of biodiversity, and climate warming all warrant close attention to coordinated efforts and networks of monitoring programs. The proposal of a network of “biosphere observatories” as originally described in an IGBP report12 is still a viable concept to global environmental monitoring activities. The LTER program of NSF, EPA’s EMAP, and the proposed and emerging NEON program represent select examples of critical, ongoing monitoring efforts. In addition, the Ecological Society of America has proposed the Sustainable Bio- sphere Initiative111 with broad applicability and relevance to integrated monitoring programs along with recommendations for an aggressive research program to assess ecological responses to stress and loss of biodiversity both on a regional and global basis. Our conceptual approach to these issues is intended to help scientists, researchers, decision makers, and other leaders develop, implement, and maintain integrated monitoring on a comprehensive basis but relevant to local, regional, national, and global perspectives. Our work at remote sites provides insight to the challenges of monitoring and detecting early-warning indicators against a background of natural variability in ecosystems. In contrast, the Heritage River watershed project has been developed with the intended beneﬁt of geospatial technologies and the earlier design components from our remote site research. These sites provide a broad spectrum of applications ranging from baseline monitoring in relative un-impacted regions to regional watersheds where land use change and extensive mining has resulted in opportunities to assess and monitor the impacts of “industrial metabolism” (see References 12 and 19) from a landscape-watershed perspective with geospatial data and technologies. This collective work is intended to promote the biosphere observatory concept and to provide practical, real world “test-beds” for evaluating methods and approaches to integrated environmental monitoring.

ACKNOWLEDGMENTS We would like to thank our numerous colleagues for their many contributions to our research at remote study sites, and especially Greg White who headed up the forest ecosystem measurements component at the Wind Rivers study area. We also appre- ciate the support of our Chilean colleagues including Guillermo Santana, who hosted our work as Manager of Torres del Paine National Park. Xiaoming Yang, Sid Halsor, Bill Toothill, Brian Oram, Bill Feher, Dave Skoronski, and Jim Thomas all contrib- uted to various aspects of data collecting and processing for the Heritage River site; Kristopher Smith reﬁned graphical materials for the chapter. This paper is part of the ongoing GIS environmental master plan for the US-L Watershed and has been supported by various people and agencies. Alex Rogers, the AHR Navigator, provided support and coordination for the community outreach aspects of the master plan cited throughout this chapter. Dave Catlin of the EPA’s Ofﬁce of Environmental

30 Environmental Monitoring

Information enthusiastically provided support and collaborative energy to our Her- itage River site regarding application of geospatial data and technologies. Tom Sweet of the Pennsylvania GIS Consortium has provided insight to the use of local scale aerial photography data and how to incorporate that component into our conceptual design approach. Chris Cappelli of ESRI provided many hours of technical and conceptual support for the use of GIS, especially Web-based GIS, in data integration, analysis, and dissemination. Writing of aspects of this paper was funded and supported by the USDA (Cooperative State Research, Education, and Extension Service) Rural GIS program (some data analysis and manuscript writing), and EPA’s Ofﬁce of Environmental Information (writing and data analysis phases). And ﬁnally, we would like to thank Congressman Paul E. Kanjorski (U.S. 11th Congressional Dis- trict in Pennsylvania) and his staff for their long-term support of GIS technology to solve environmental problems; the Congressman was tireless in his leadership to facilitate designation of the US-L watershed as an American Heritage River.

REFERENCES

1. Franklin, J.F., Bledsoe, C.S., and Callahan, J.T., Contributions of the long-term ecological research program, BioScience, 40, 509, 1990. 2. National Acid Precipitation Assessment Program (U.S. NAPAP), 1990 Integrated Assessment Report, NAPAP Office of the Director, Washington, D.C., 1991. 3. Summers, J.K. and Tonnessen, K.E., Linking monitoring and effects research: EMAP’s intensive site network program, Environ. Monit. Assess., 51, 369, 1998. 4. Jones, K.B. et al., An Ecological Assessment of the United States Mid-Atlantic Region: A Landscape Atlas, U.S. EPA, Office of Research and Development, Washington, D.C., EPA/600/R-97/130, 1997. 5. Houghton, R.A. and Woodwell, G.M., Global climate change, Sci. Am., 260, 36, 1989. 6. Houghton, R.A. et al., Climate Change 1995: The Science of Climate Change, Cambridge University Press, Cambridge, U.K., 1996. 7. Pimm, S.L. et al., The future of biodiversity, Science, 269, 347, 1995. 8. Vitousek, P.M. et al., Human domination of earth’s ecosystems, Science, 277, 494, 1997. 9. Carpenter, S.R. et al., Global change and freshwater ecosystems, Annu. Rev. Ecol. Syst. 23, 19, 1992. 10. Chapin, F.S., III et al., Biotic control over the functioning of ecosystems, Science, 277, 500, 1997. 11. National Research Council, Global Change in the Geosphere–Biosphere: Initial Pri- orities for an IGBP, National Academies Press, Washington, D.C., 1986. 12. National Research Council, Toward an Understanding of Global Change: Initial Priorities for U.S. Contributions to the International Geosphere–Biosphere Program, National Academies Press, Washington, D.C., 1988. 13. Long-Term Ecological Research Program (NSF-LTER Network Office), 1990’s Glo- bal Change Action Plan, Long-Term Ecological Research Office, Seattle, WA, 1989. 14. Walker, B. et al., Eds., The Terrestrial Biosphere and Global Change: Implications for Natural and Managed Ecosystems, Cambridge University Press, Cambridge, U.K., 1999. 15. Scientific Committee on Problems of the Environment/Man and the Biosphere Pro- gramme (SCOPE/MAB), Definition and Description of Biosphere Observatories for Studying Global Change, Draft Report, SCOPE/MAB Workshop, Paris, France, 1987.

Conceptual Basis of Environmental Monitoring Systems 31

16. University Corporation for Atmospheric Research, Opportunities for Research at the Atmosphere/Biosphere Interface, Report of a Workshop, Boulder, CO, 1985. 17. Bruns, D.A., Wiersma, G.B., and Minshall, G.W., Evaluation of community and ecosystem monitoring parameters at a high-elevation Rocky Mountain study site, Environ. Toxicol. Chem., 11, 359, 1992. 18. Bruns, D.A., Wiersma, G.B., and Minshall, G.W., Lotic ecosystems and long-term monitoring for global change, in Global Warming and Freshwater Ecosystems, Firth, P. and Fisher, S.G., Eds., Springer-Verlag, New York, 1992, chap. 14. 19. Bruns, D.A., Wiersma, G.B., and White, G.J., Testing and application of ecosystem monitoring parameters, Toxicol. Environ. Chem., 62, 169, 1997. 20. Bruns, D.A. et al., System for Environmental Survey and Land Use (GIS Analysis of watershed impacts from regional coal mining), Final Report to Earth Conservancy (Ashley, PA) and Department of Defense, Advanced Research Program Agency, Washington, D.C., 1997. 21. Wiersma, G.B., Conceptual basis for environmental monitoring programs, Toxicol. Environ. Chem., 27, 241, 1990. 22. Wiersma, G.B. et al., The use of simple kinetic models to help design environmental monitoring systems, Environ. Monit. Assess., 4, 233, 1984. 23. Wiersma, G.B. et al., Integrated global background monitoring network, in Monitor- ing and Managing Environmental Impact: American and Soviet Perspectives, Sch- weitzer, G.E. and Philips, A.S., Eds., Proceedings of the Fifth United States/Union of Soviet Socialist Republic Symposium on Comprehensive Analysis of the Environ- ment, National Academies Press, Washington, D.C., 1986, 246. 24. Wiersma, G.B. et al., Reconnaissance of Noatak National Preserve and Biosphere Reserve as a potential site for inclusion in the Integrated Global Background Mon- itoring Network, Report for U.S. Man and the Biosphere Program, NTIS PB 88- 100037, 1986. 25. Wiersma, G.B. et al., Integrated Monitoring Project at Torres del Paine National Park, Chile, Methodology and Data Report—1984 to 1986, EG&G Idaho, Inc., Informal Report, EGG-EES-7966, 1988. Idaho Falls, ID. 26. Wiersma, G.B. and Bruns, D.A., Monitoring for ecological assessment, in North American Workshop on Monitoring for Ecological Assessment of Terrestrial and Aquatic Ecosystems, Bravo, C.A., Ed., USDA Technical Report RM-GTR-284, 1996, 31. 27. Bruns, D.A., Wiersma, G.B., and Rykiel, E.J., Jr., Ecosystem monitoring at global baseline sites, Environ. Monit. Assess., 17, 3, 1991. 28. Nilhgard, B. and Pylvanainen, M., Eds., Evaluation of Integrated Monitoring Pro- gramme in Terrestrial Reference Areas in Europe and North America: The Pilot Programme 1989–91, Environmental Data Centre, National Board of Waters and the Environment, Helsinki, 1992. 29. Heal, O.W., Menaut, J., and Steffen, W.L., Eds., Towards a Global Terrestrial Observ- ing System (GTOS): Detecting and Monitoring Change in Terrestrial Ecosystems, MAB Digest 14 and IGBP Global Change Report 26, UNESCO, Paris, and IGBP, Stockholm, 1993. 30. Boesch, D.F. et al., Managing Troubled Waters: The Role of Marine Environmental Monitoring, National Academies Press, Washington, D.C., 1990. 31. Wiersma, G.B. et al., Finding the Forest in the Trees: The Challenge of Combining Diverse Environmental Data, National Academies Press, Washington, D.C., 1995. 32. O’Neill, R.V. et al., Indices of landscape pattern, Landscape Ecol., 1, 153, 1988. 33. O’Neill, R.V. et al., Monitoring environmental quality at the landscape scale, Bio- Science, 47, 513, 1997.

32 Environmental Monitoring

34. Rosswall, T., Woodmansee, R.G., and Risser, P.G., Eds., Scales and Global Change: Spatial and Temporal Variability in Biospheric and Geospheric Processes, SCOPE 35, John Wiley & Sons, New York, 1988. 35. Quattrochi, D.A. and Goodchild, M.F., Eds., Scale in Remote Sensing and GIS, Lewis Publishers, Boca Raton, FL, 1997. 36. Clesceri, L.S., Greenberg, A.E., and Eaton, A.D., Eds., Standard Methods for the Examination of Water and Wastewater, American Public Health Association (APHA), Washington, D.C., 1998. 37. Keith, L.H., Ed., Principles of Environmental Sampling, American Chemical Society, Washington, D.C., 1988. 38. Minshall, G.W. et al., Developments in stream ecosystem theory, Can. J. Fish. Aquat. Sci., 42, 1045, 1985. 39. EDAW, Inc., Earth Conservancy Land Use Plan, Luzerne County, Pennsylvania, Final Report to Earth Conservancy, Ashley, PA, 1996. 40. Bruns, D.A., Sweet, T., and Toothill, B., Upper Susquehanna–Lackawanna River Watershed, Section 206, Ecosystem Restoration Report, Phase I GIS Environmental Master Plan, Final Report to U.S. Army Corps of Engineers, Baltimore District, MD, 2001. 41. Wiersma, G.B. and Otis, M.D., Multimedia design principles applied to the development of the global integrated monitoring network, in Pollutants in a Multimedia Environment, Cohen, Y., Ed., Plenum Publishing, New York, 1986, 317. 42. O’Neill, R.V., Hierarchy theory and global change, in Scales and Global Change: Spatial and Temporal Variability in Biospheric and Geospheric Processes, Rosswall, T., Woodmansee, R.G., and Risser, P.G., Eds., SCOPE 35, John Wiley & Sons, New York, 1988, 29. 43. Olson, R.K. and Lefohn, A.S., Eds., Effects of Air Pollution on Western Forests, Transactions Series No. 16, ISSN 1040-8177, 1989. 44. Bohm, M. and Vandetta, T., Atlas of Air Quality and Deposition in or near Forests of the Western United States, U.S. Environmental Protection Agency, EPA/600/3- 90/081, 1990. 45. Bruns, D.A. and Yang, X., An accuracy assessment of satellite imagery used in landscape–watershed assessments: a comparison of four databases, in Papers and Proceedings of the Applied Geography Conferences, Vol. 25, Montz, B.E. and Tobin, G.A., Eds., Applied Geography Conferences, Binghamton, NY, 2002, 230. 46. White, G.J. et al., Evaluation of U.S. Forest Service Document: Guidelines for Mea- suring the Physical, Chemical, and Biological Condition of Wilderness Areas, Final Report, EG&G Idaho, Inc., Informal Report, EGG-BE-9929, DOE National Engi- neering Lab., Idaho Falls, ID, 1991. 47. Fox, D.G., Bernabo, J.C., and Hood, B., Guidelines for Measuring the Physical, Chemical, and Biological Condition of Wilderness Areas, General Technical Report RM-146, U.S. Department of Agriculture, Forest Service Rocky Mountain Forest and Range Experimental Station, Fort Collins, CO, 1987. 48. Mackay, D., Multimedia Environmental Models, the Fugacity Approach, Lewis Pub- lishers, Boca Raton, FL, 1991. 49. Landrum, P.F., Harkey, G.A., and Kukkonen, J., Evaluation of organic contaminant exposure in aquatic organisms: the signiﬁcance of bioconcentration and bioaccumulation, in Ecotoxicology: A Hierarchical Treatment, Newman, M.C. and Jagoe, C.H., Eds., CRC Press, Boca Raton, FL, 1996, chap. 4. 50. Newman, M.C., Fundamentals of Ecotoxicology, Ann Arbor Press, Chelsea, MI, 1998. 51. Schindler, D.W., Effects of acid rain on freshwater ecosystems, Science, 239, 149, 1988.

Conceptual Basis of Environmental Monitoring Systems 33

52. Taylor, G.E., Jr., Forest ecosystems and air pollution: the importance of multiple stress interactions on a regional and global scale, in Multiple Stresses in Ecosystems, Cech, J.J., Jr., Wilson, B.W., and Crosby, D.G., Eds., Lewis Publishers, Boca Raton, FL, 1998, chap. 4. 53. Shaw, I.C. and Chadwick, J., Principles of Environmental Toxicology, Taylor and Francis, Bristol, PA, 1998. 54. Environmental Protection Agency (EPA), Protocol for the Transition-Flow Reactor Concentration Monitor: Determination of Atmospheric Concentrations of Gaseous

HNO3, SO2, NO2, and NH3 and of Fine Particulate Nitrate, Sulfate, and Ammonium Ion, Atmospheric Sciences Research Laboratory, Ofﬁce of Research and Develop- ment, Research Triangle Park, NC, 1987. 55. Wiersma, G.B., Recommended Integrated Monitoring System for Pollutants in U.S. National Parks Designated as Biosphere Reserves, EG&G Idaho Informal Report, EGG-PBS-6721, Idaho Falls, ID, 1985. 56. Robarge, W.P. and Fernandez, I., Quality Assurance Methods Manual for Laboratory Analytical Techniques, Draft EPA Report, Corvallis Environmental Research Labo- ratory, Corvallis, OR, 1986. 57. National Surface Water Survey (NSWS), Analytical Methods Manual for the National Surface Water Survey, Environmental Protection Agency, Environmental Monitoring Systems Laboratory, Las Vegas, NV, 1986. 58. Vannote, R.L. et al., The river continuum concept, Can. J. Fish. Aquat. Sci., 37, 130, 1980. 59. Bruns, D.A. et al., Ordination of functional groups and organic matter parameters from the middle fork of the Salmon River, Idaho, Freshw. Invertebr. Biol., 1, 2, 1982. 60. Bruns, D.A. et al., Tributaries as modiﬁers of the river continuum concept: analysis by polar ordination and regression models, Arch. Hydrobiol., 99, 208, 1984. 61. Baker, G.A., Harmon, M.E., and Greene, S.E., A study of selected ecosystem processes potentially sensitive to airborne pollutants, in Proceedings of a Conference on Science in the National Parks: Physical Processes and Water Resources, Flug, M., Ed., G. Wright Society and Colorado State University, Fort Collins, CO, 1986, 119. 62. Cairns, J., Jr., The myth of the most sensitive species, BioScience, 36, 670, 1986. 63. Schindler, D.W., Ecosystems and ecotoxicology: a personal perspective, in Ecotoxi- cology: A Hierarchical Treatment, Newman, M.C. and Jagoe, C.H., Eds., Lewis Publishers, Boca Raton, FL, 1996, chap. 13. 64. Newman, M.C. and Jagoe, C.H., Eds., Ecotoxicology: A Hierarchical Treatment, Lewis Publishers, Boca Raton, FL, 1996. 65. Cech, J.J., Jr., Wilson, B.W., and Crosby, D.G., Eds., Multiple Stresses in Ecosystems, Lewis Publishers, Boca Raton, FL, 1998. 66. Allen, T.F.H. and Starr, T.B., Hierarchy: Perspective for Ecological Complexity, Uni- versity of Chicago Press, Chicago, IL, 1982. 67. O’Neill, R.V. et al., A Hierarchical Concept of Ecosystems, Princeton University Press, Princeton, NJ, 1986. 68. Bruns, D.A. et al., An ecosystem approach to ecological characterization in the NEPA process, in Environmental Analysis: The NEPA Experience, Hildebrand, S.G. and Cannon, J.B., Eds., Lewis Publishers, Boca Raton, FL, 1993, 103. 69. Rapport, D.J., Regier, H.A., and Hutchinson, T.C., Ecosystem behavior under stress, Am. Nat., 125, 617, 1985. 70. Odum, E.P., Trends expected in stressed ecosystems, BioScience, 35, 419, 1985. 71. Schaeffer, D.J., Herricks, E.E., and Kerster, H.W., Ecosystem health: I. Measuring ecosystem health, Environ. Manage., 12, 445, 1988.

34 Environmental Monitoring

72. National Research Council, Testing for the Effects of Chemicals on Ecosystems, National Academy Press, Washington, D.C., 1981. 73. Sheehan, P.J. et al., Eds., Effects of Pollutants at the Ecosystem Level, SCOPE 22, John Wiley & Sons, New York, 1984. 74. Taub, G.B., Indicators of change in natural and human-impacted ecosystems: status, in Preserving Ecological Systems: The Agenda for Long-Term Research and Devel- opment, Draggan, S., Cohrssen, J.J., and Morrison, R.E., Eds., Praeger, NY, 1987, 115. 75. Sigal, L.L. and Suter, G.W., II, Evaluation of methods for determining adverse impacts of air pollution on terrestrial ecosystems, Environ. Manage., 11, 675, 1987. 76. Schindler, D.W., Detecting ecosystem responses to anthropogenic stress, Can. J. Fish. Aquat. Sci., 44, 6, 1987. 77. Hinds, W.T., Towards monitoring of long-term trends in terrestrial ecosystems, Envi- ron. Conserv., 11, 11, 1984. 78. Beanlands, G.E. and Duinker, P.N., An ecological framework for environmental impact assessment, J. Environ. Manage., 18, 267, 1984. 79. Karr, J.R., Biological monitoring and environmental assessment: a conceptual framework, Environ. Manage., 11, 249, 1987. 80. Ausmus, B.S., An argument for ecosystem level monitoring, Environ. Monit. Assess., 4, 275, 1984. 81. Zeiler, M., Modeling our World: The ESRI Guide to Geodatabase Design, ESRI Press, Redlands, CA, 1999. 82. Bruns, D.A. and Sweet, T., Tackling Environmental Clean-up with GIS: Regionally Coordinated Geographic Information Systems Provide Solutions to Susquehanna– Lackawanna Watershed Pollution Problems, University of Wisconsin, Land Informa- tion Bulletin, USDA Rural GIS, Madison, WI, 2001. 83. Smith, J.H. et al., Impacts of patch size and land cover heterogeneity on thematic image classiﬁcation accuracy, Photogr. Eng. Rem. Sens., 68, 65, 2002. 84. Avery, T.E. and Berlin, G.L., Fundamentals of Remote Sensing and Airphoto Inter- pretation, Macmillan, NY, 1992. 85. Lunetta, R.S. et al., Remote sensing and geographic information systems data integration: error sources and research issues, Photogr. Eng. Rem. Sens., 57, 677, 1991. 86. Vogelmann, J.E. et al., Completion of the 1990s National Land Cover Data Set for the conterminous United States from Landsat Thematic Mapper data and ancillary data sources, Photogr. Eng. Rem. Sens., 67, 650, 2001. 87. Federal Geographic Data Committee, Framework Introduction and Guide, Federal Geographic Data Committee, Washington, D.C., 1997. 88. Falkner, E., Aerial Mapping Methods and Applications, Lewis Publishers, Boca Raton, FL, 1995. 89. Lyon, J.G., Wetland Landscape Characterization: Techniques and Applications for GIS, Mapping, Remote Sensing, and Image Analysis, Ann Arbor Press, Chelsea, MI, 2001. 90. Dahlman, B.N. and Lanzer, E.L., Cartographic support for managing Washington State’s aquatic resources, in GIS Solutions in Natural Resource Management: Bal- ancing the Technical–Political Equation, Morain, S., Ed., OnWord Press, Santa Fe, NM, 1999, 115. 91. Brondizio, E. et al., Land cover in the Amazon estuary: linking of the Thematic Mapper with botanical and historical data, Photogr. Eng. Rem. Sens., 62, 921, 1996. 92. Congalton, R.G., A review of assessing the accuracy of classiﬁcations of remotely sensed data, Rem. Sens. Environ., 37, 35, 1991.

Conceptual Basis of Environmental Monitoring Systems 35

93. Congalton, R.G. and Green, K., Assessing the Accuracy of Remotely Sensed Data: Principles and Practices, Lewis Publishers, Boca Raton, FL, 1999. 94. Turner, M.G., Landscape ecology: the effect of pattern on process, Annu. Rev. Ecol. Syst., 20, 171, 1989. 95. Turner, M.G. and Gardner, R.H., Eds., Quantitative Methods in Landscape Ecology, Springer-Verlag, New York, 1991. 96. Holl, K.D. and Cairns, J., Jr., Landscape indicators in ecotoxicology, in Handbook of Ecotoxicology, Hoffman, D.J., Eds., Lewis Publishers, Boca Raton, FL, 1994, 185. 97. Hunsaker, C.T. and Levine, D.A., Hierarchical approaches to the study of water quality in rivers, BioScience, 45, 193, 1995. 98. Preston, S.D. and Brakebill, J.W., Application of Spatially Referenced Regression Modeling for the Evaluation of Total Nitrogen Loading in the Chesapeake Bay Watershed, USGS Water-Resources Investigations Report 99-4054, U.S. Department of the Interior/U.S. Geological Survey, Reston, VA, 1999. 99. Interagency Team, Stream Restoration Guidance Manual, Stream Corridor Restora- tion: Principles, Processes, and Practices, Federal Interagency Stream Restoration Working Group (FISRWG) (15 federal agencies of the U.S. government), GPO Item No. 0120-A; SuDocs ISBN-0-934213-59-3, Washington, D.C., 1998. 100. Lillesand, T.M. and Kieffer, R.W., Remote Sensing and Image Interpretation, John Wiley & Sons, New York, 1987. 101. Lunetta, R.S. and Elvidge, C.D., Eds., Remote Sensing Change Detection: Environ- mental Methods and Applications, Ann Arbor Press, Chelsea, MI, 1998. 102. Wessman, C.A., Spatial scales and global change: bridging the gap from plots to GCM grid cells, Annu. Rev. Ecol. Syst., 23, 175, 1992. 103. Environmental Protection Agency (EPA), Better Assessment Science Integrating Point and Nonpoint Sources (BASINS), User’s Manual, Version 3.0, U.S. EPA Ofﬁce of Water (4305), Washington, D.C., EPA-823-H-01-001, 2001. 104. Arnold, J.G. et al., SWAT — Soil Water Assessment Tool, USDA, Agricultural Research Service, Grassland, Soil and Water Research Laboratory, Temple, TX, 1994. 105. Hernandez, M. et al., Modeling runoff response to land cover and rainfall spatial variability in semi-arid watersheds, Environ. Monit. Assess., 64, 285, 2000. 106. American Forests, CITYgreen: Calculating the Value of Nature, Version 5.0 User’s Manual, American Forests, Washington, D.C., 2002. 107. Carl, J.R., DeAngelo, M.M., and Ruskey, D.C., Sustainable development plan and runoff study for a proposed development in the Hick’s Creek watershed, in Papers and Proceedings of the Applied Geography Conferences, Vol. 25, Montz, B.E. and Tobin, G.A., Eds., Applied Geography Conferences, Inc., Binghamton, l NY, 2002, 9. 108. Ebert, D.W. et al., Analytical Tools Interface for Landscape Assessments (ATtILA), Quick Start Guide, Version 3.0, Draft, Environmental Protection Agency, Ofﬁce of Research and Development, Las Vegas, NV, 2001. 109. National Research Council, Making the Nation Safer: The Role of Science and Technology in Countering Terrorism, National Academies Press, Washington, D.C., 2002. 110. Environmental Protection Agency, Strategic Plan for Homeland Security, Web site at http://www.epa.gov/epahome/downloads/epa_homeland_security_strategic_plan.pdf, 2002. 111. Lubchenco, J. et al., The sustainable biosphere initiative: an ecological research agenda, Ecology, 72, 371, 1991.

Review Agricultural modiﬁcations of hydrological ﬂows create ecological surprises

Line J. Gordon1, Garry D. Peterson2 and Elena M. Bennett3

1 Stockholm Resilience Centre and Stockholm Environment Institute, Stockholm University, S-10691 Stockholm, Sweden 2 Department of Geography and McGill School of Environment, McGill University, 805 Sherbrooke Street West, Montreal, QC H3A 2K6, Canada 3 Department of Natural Resource Sciences and McGill School of Environment, McGill University, 21111 Lakeshore Road, Ste-Anne-de-Bellevue, QC H9X 3V9, Canada

Agricultural expansion and intensification have altered (CA) [7]. Their reviews of research needs revealed that the quantity and quality of global water flows. Research although knowledge of these trade-offs has increased, we suggests that these changes have increased the risk of lack an integrated understanding of how agricultural catastrophic ecosystem regime shifts. We identify and modifications of the hydrological cycle regulate the preva- review evidence for agriculture-related regime shifts in lence and severity of surprising nonlinear change in eco- three parts of the hydrological cycle: interactions be- systems [7,8]. tween agriculture and aquatic systems, agriculture and Some of the most catastrophic changes in ecosystem soil, and agriculture and the atmosphere. We describe services are a result of nonlinear, abrupt shifts between the processes that shape these regime shifts and the different ecosystem regimes (Box 2). Regime shifts are scales at which they operate. As global demands for frequently surprising and difficult to reverse, presenting agriculture and water continue to grow, it is increasingly a substantial challenge to ecosystem management and urgent for ecologists to develop new ways of anticipat- development goals [9–11]. A rapidly growing body of evi- ing, analyzing and managing nonlinear changes across dence suggests that agricultural modification of the quality scales in human-dominated landscapes. and quantity of hydrological flows can increase the risk of ecological regime shifts [12–17]. Humans have modified the water cycle through An improved, synthetic understanding of how such agriculture regime shifts are produced is particularly urgent now Human transformation of global water flows has dramatic- because of growing demand for water, agricultural pro- ally impacted ecosystems and the services they generate. ducts such as food and biofuels, and other ecosystem Through water withdrawals, land use and land cover services such as carbon sequestration, climate moderation, changes, agriculture, which covers almost 40% of the ter- erosion control and opportunities for recreation. Climate restrial surface [1], is arguably the major way in which change that is expected to generate unprecedented altera- humans change water quantity and quality (Box 1). Water tions in precipitation, soil moisture and runoff will make for irrigation accounts for 66% of societal water withdra- negotiating the complex hydrology-related ecological wals, reducing water availability for downstream ecosys- trade-offs of agriculture even more challenging. In this tems [2]. Irrigation and deforestation for agriculture have paper, we review how agricultural modification of the redistributed global evapotranspiration, altering the hydrological cycle can produce regime shifts. We empha- regional climate [3]. Nutrient runoff from agricultural size the key agricultural and hydrological processes fertilizer use has decreased water quality in aquatic eco- involved in regime shifts and the spatial and temporal systems around the world [4,5]. These changes have driven scales at which these processes operate. We conclude by rapid declines in nonagricultural ecosystem services, such discussing management challenges for building resilience as fisheries, flood regulation and downstream recreational against undesired regime shifts, and highlight important opportunities [6]. Despite these impacts, increases in research questions that need to be addressed to reduce the agricultural production have reduced malnutrition and risk of future agriculture–water surprises. hunger, and agriculture has been an engine of economic growth in many countries. The complex trade-offs between Three parts of the hydrological cycle where agriculture increased agricultural production and declines in other can trigger regime shifts ecosystem services as caused by agricultural changes to The hydrological cycle can be seen as the ‘bloodstream of the hydrological cycle have been reviewed by the Millen- the biosphere’ [18], because runoff, groundwater and eva- nium Ecosystem Assessment (MA) [6] and the Compre- potranspiration move materials among different ecosys- hensive Assessment of Water Management in Agriculture tems and alter energy balances in landscapes. This paper examines how agricultural changes across the whole Corresponding author: Gordon, L.J. ([email protected]). hydrological cycle can produce regime shifts. We classify

Box 1. Historical and future demand of land and water for agriculture

Agricultural production, and its related hydrological changes, has for 2050, suggesting that depending on investment strategies, trade greatly increased during the 20th century. Figure I shows (i) and improvement of technology, water demands for agriculture will agricultural land use, including croplands and rangelands over the rise 30%–54%, while agricultural area will expand between 9% and past 300 years [2], (ii) the amount of societal water withdrawal for 38% [7]. irrigation since 1900 [2] and (iii) the increase in agricultural fertilizer The CA stressed the importance of improving water productivity, use since 1960 [64]. which is most often measured as m3 of evapotranspiration used per These changes are expected to continue in the 21st century. ton of grains. Water productivity can be increased by reducing Population growth, the production of biofuels and increased meat unproductive losses of water, such as soil evaporation, while consumption are driving increased agricultural demands. The increasing productive water flows (transpiration). The highest gains Millennium Ecosystem Assessment scenarios estimated that future in water productivity can come from improvements in low-yielding agricultural expansion would convert between 10% and 20% of agriculture when supplemental irrigation is combined with improved existing forest and grassland to cropland, with this conversion tillage and nutrient management [66]. The CA scenarios showed that expected to be concentrated in low-income countries and in drylands increased water productivity has the potential to reduce future water [6]. The nations of the world have pledged in the UN Millennium needs by more than 50%, compared to future water needs without Development Goals to halve the proportion of people who suffer from any improvements in water productivity [7]. However, the potential malnutrition by 2015, and to eradicate hunger by 2030. Meeting these trade-offs between locally improved water productivity and potential goals in targeted countries will require a 50% increase of food downstream water quality decline from increased use of fertilizers production by 2015 and at least a doubling at 2050, with water and pesticides due to increased water productivity are scarcely consumption (in terms of evapotranspiration) increasing from addressed in the literature and illustrate the need to enhance our 4500 km3/yr to 7300 km3/yr in 2050 [65]. The Comprehensive Assess- capacity to analyze multiple trade-offs. ment of Water Management in Agriculture (CA) developed scenarios

Figure I. Agriculture’s extent and modification of the quantity and quality of hydrological flows have increased over the past centuries. agriculture–water regime shifts into three categories organic matter produced by excess nutrients in estuaries depending on where they occur in relation to the hydro- can lead to areas of depleted oxygen called hypoxic zones logical cycle (Figure 1). These categories are: [21]. These can have alternative regimes due to recycling of (i) agriculture and aquatic systems, including changes in nitrogen stored in sediments during previous years. This runoff quality and quantity that lead to regime shifts feedback can, however, be weakened owing to tempera- in downstream aquatic systems (Figure 1a); ture, water movement and the availability of other nutri- (ii) agriculture and soil, in which changes in inﬁltration ents [22]. Freshwater and estuarine eutrophication regime and soil moisture result in terrestrial regime shifts shifts caused by phosphorus can be a result of accumu- (Figure 1b); and lation of phosphorus in aquatic sediments as well as in (iii) agriculture and the atmosphere, in which changes in agricultural soils. Accumulated phosphorus in sediments evapotranspiration result in regime shifts in the can be released to the water column under the low-oxygen climatic system itself or in terrestrial ecosystems as a conditions common to eutrophication, setting in motion a consequence of climatic changes (Figure 1c). positive feedback cycle in which low oxygen caused by high nutrient levels leads to additional phosphorus released to the water column [19,23]. Furthermore, phosphorus can Agriculture and aquatic systems also be accumulated in agricultural soils when not all of the Agriculturally driven change in water ﬂows, nutrient applied phosphorus is taken up by plants [5]. Because soil levels and sediment loads can produce regime shifts in phosphorus concentrations change slowly, this accumu- downstream aquatic systems in at least three different lation might set the stage for decades, if not centuries, ways (Table 1). Nutrients such as phosphorus and nitrogen of impaired water quality, even after phosphorus input to in fertilizers can move downhill in runoff, leading to fresh- terrestrial systems has stopped [20]. water and estuarine eutrophication [19] that might be Another aquatic system regime shift is changes in river reversible only after massive reductions of nutrients for channel position [24], common in river deltas when agricul- decades or longer [20]. Decomposition of large amounts of tural land use affects sediment loading and channel veg-

212 Review Trends in Ecology and Evolution Vol.23 No.4

Box 2. Ecosystem regime shifts

Ecological dynamics are defined by both internal dynamics (such as thus often referred to as ‘slow variables.’ If a valley in the stability vegetation growth) and external forces (such as precipitation and landscape completely vanishes, or an external shock pushes the droughts). Regime shifts occur when external forces or gradual system from one valley to another, the system undergoes a regime internal changes alter a system so that its organization shifts from shift. Equilibria diagrams (Figure I, lower panel) summarize how the being organized around one set of mutually reinforcing processes dynamics of an ecosystem change as ecological drivers change. The (e.g. vegetation enhancing precipitation) to another (e.g. less vegeta- colored dotted lines show where each stability landscape is located tion, no effect on precipitation) [10,11,15]. Resilience is related to the on each equilibria diagram. concept of regime shifts and is defined as the capacity to deal with Gradual change (Figure Ia) occurs when precipitation increases change and disturbances while retaining essentially the same from drier to wetter conditions in a landscape that contains species functions and processes [11]. whose growth responses to increasing vegetation are diverse and Figure I illustrates the differences between gradual ecological relatively evenly distributed so that vegetation cover gradually change and three different types of regime shifts using precipitation– increases with precipitation. Threshold change (Figure Ib) occurs vegetation interactions as an example [15,17,42]. The feedbacks that when vegetation cover rapidly increases at a specific amount of maintain a system in a given regime are represented as the shape of precipitation. Changes in external drivers can in systems with stability landscapes (Figure I, upper panel), with the configuration of thresholds push a system from dense to spare vegetation cover. the system represented by a ball. Multiple valleys in the landscape Hysteresis (Figure Ic) can occur if there is strong moisture recycling represent the potential of having alternative regimes. External from vegetation that stimulates precipitation; this implies that the shocks (such as a drought or an intense rainfall event) can change precipitation thresholds at which vegetation will quickly increase the system configuration, moving the ball across the landscape. (recovery) are higher than precipitation levels at which vegetation will However, the stability landscape itself can also change as the forces collapse. Irreversible change (Figure Id) is a stronger form of defining the feedback processes in the system change, for example, hysteresis, where vegetation is unable to recover to its pre-collapse through vegetation clearing or changes in soil moisture holding levels, even if rainfall increases substantially. This type of dynamic capacity. Changes in internal variables that alter the feedback can occur when the ability of vegetation to recover is lost from the processes that define a regime are often relatively slower than other system during a collapse. system dynamics that people monitor, such as yield levels. They are

Figure I. The differences between gradual ecological change (a) and three different types of regime shifts (b–d) using precipitation–vegetation interactions as an example. etation [25]. Rivers overloaded with sediment from eroded Grazing can shift systems to a less productive regime by agricultural soils can experience channel incision [26],in reducing vegetation cover, setting in motion a feedback which increased scouring causes loss of channel vegetation, that decreases nutrient and water accumulation [29,30]. which can, in turn, lead to the abrupt formation of highly Similarly, in arid to semiarid savanna systems, regime eroded channels [27]. In other cases, sediment load blocks shifts between grass and shrub domination can occur if the the path of the river, causing the river to suddenly form a existing competition between grasses and shrubs for water new course. It can take centuries or longer for the river to in the root zone is destabilized. This competition can be return to its original course, if it ever does so [24]. altered by grazing that removes less drought-sensitive species, by differing patterns and intensity of fire and by Agriculture and soil systems changes in drought occurrence [31]. Interactions between vegetation and soil water, through A second type of regime shift occurs when a rise in the effects on infiltration, soil water holding capacity and root water table triggers salinization. Water table rise occurs water uptake cause at least three types of regime shifts when more water infiltrates the soil, owing to irrigation or leading to land degradation and loss of productivity increases in precipitation, or when less water is removed (Table 1). The first is related to vegetation patchiness that due to reductions in evapotranspiration when deep-rooted can impede the flow of surface water, creating spatially trees are replaced with annual crops or grasses. The concentrated patterns of infiltration that locally increase resulting water table rise can mobilize salts in the soil. nutrient and water accumulation, which in turn sustains The process speeds up when a water table has risen to vegetation growth and landscape patchiness [14,28,29]. within 2 m of the surface, because capillary action

213 Review Trends in Ecology and Evolution Vol.23 No.4

Figure 1. Locations of regime shifts in the hydrological cycle. Conceptual diagram showing three main flows in the hydrological cycle (precipitation, runoff and evapotranspiration) and where in the hydrological cycle three potential regime shifts produced by couplings between agriculture and water flows can occur: (a) agriculture and aquatic systems, (b) agriculture and soil and (c) agriculture and atmosphere regime shifts. through fine pores can then move water to the surface, Agriculture and the atmosphere carrying salt with it. Salts present in the soil become Land cover conversion for agriculture impacts precipi- concentrated in the plant rooting zone making the soil tation and can produce at least four different regime shifts. inhospitable to vegetation, resulting in regime shifts diffi- Many regional to global studies of vegetation–climate cult to reverse or irreversible [32,33]. Finally, it has been interactions have shown that changes in vegetation cover suggested that reductions in the length of fallow periods in can alter precipitation [38,39]. Some of these studies have semiarid croplands can result in rapid yield declines once assessed whether these interactions can produce regime fallow periods fall below a critical threshold [34]. Fallow shifts [40]. Theory suggests that regime shifts can occur if periods allow for restoration of soil organic matter and two conditions are fulfilled [17,41].First,thevegetation nutrient levels. Reduced fallows can lead to changes in soil cover has to respond nonlinearly to changes in precipi- structure including compaction and crusting that renders tation (Box 2, Figure Ib–d) and second, vegetation has to soils more drought sensitive by decreasing soil moisture have a sufficiently strong effect on precipitation that it holding capacity [35]. Soil water availability is crucial to can, in turn, alter the amount of vegetation cover (Box 2). sustain yields and productivity, because highly variable The strongest evidence for land–atmosphere regime shifts precipitation in drylands produces recurrent dry spells and comes from studies of the shifts between wet savanna droughts [36]. The evidence for this type of regime shift is systems and dry savanna systems in the Sahara and relatively weak. However, although soil structure regime the Sahel. Paleoecological records show that vegetation shifts might be biophysically quite easy to reverse, decreas- collapsed relatively abruptly 5500 years ago, demon- ing incentives to invest in degraded soils implies that there strating that regime shifts can occur in this region [42]. can be stronger economic than biophysical thresholds for Modern empirical evidence has shown that vegetation in the reversibility of these shifts [37]. the Sahel responds nonlinearly to changes in precipi-

214 Review Trends in Ecology and Evolution Vol.23 No.4

Table 1. Regime shifts from agriculture changes in water quality and quantity, showing alternative regimes, consequences of the regime shift, key internal variables, agricultural drivers of change, other drivers and assessment of the evidence for the reality of each shift Internal slow Regime shift Regime A Regime B Impacts of shift variable Agricultural driver Other drivers Evidence from A to B Refs Freshwater Eutrophic Non- Reduced access to Sediment and Nutrient and soil Flooding, Strong [19,20,23] eutrophication eutrophic recreation, drinking watershed soil management landslides water, risk of fish loss phosphorus Coastal Hypoxic Not hypoxic Fishery decline, Aquatic Nutrient and soil Flooding Strong [4,21,22] hypoxic zones loss of marine biodiversity management biodiversity, toxic algae River channel Old channel New channel Damage to trade and River channel Erosion, river Extreme floods, Strong [24,25,27] position infrastructure shape channelization climate Vegetation Spatial No spatial Productivity Vegetation Grazing, land Fires, droughts Medium [14,28,29] patchiness pattern pattern declines, erosion pattern clearing Salinization High Low Yield declines, Water table salt Reduced Wetter climate Strong [32,33] productivity productivity salt damage to accumulation evapotranspiration, infrastructure irrigation and ecosystems, contamination of drinking water Soil structure High Low Yield decline, Soil organic Biomass removal, Droughts, dry Weak [35–37] productivity productivity reduced drought matter fallow frequency spells tolerance Wet savanna- Wet Dry savanna Loss of productivity, Moisture Reduced net Droughts, fires Medium [42–44] dry savanna savanna or desert yield declines, recycling, primary droughts/dry spells energy balance production and evapotranspiration Monsoon Monsoon Weak or no Risk for crop failures, Energy balance, Land cover Change in sea Weak [42,50] circulation monsoon changed advective change, irrigation surface climate variability moisture flows temperature Forest- Forest Savanna Loss of biodiversity, Moisture Reduced net Fires Weak [31,41, savanna changed suitability recycling, primary 45,46] for agricultural energy balance production production Cloud forest Cloud forest Woodland Loss of productivity, Leaf area Land clearing Fog frequency Medium [51,52,60] reduced runoff, biodiversity loss tation, and recent modeling studies have suggested that persist despite low precipitation. Without large enough leaf these changes in vegetation can produce precipitation area to intercept fog, there is insufficient moisture capture to feedbacks to vegetation that could enable regime shifts establish vegetation. Clearing of vegetation can thus result [17,42–44]. in a regime shift to a savanna or shrubland [51,52]. Transitions of forests to savannas has been suggested as a regime shift in the Amazon region, although the evidence There is large variation in the spatial scales and for this type of shift is weaker [45]. The possibility of reversibility of regime shifts alternative regimes in the Amazon has been suggested by The regime shifts we have identified as related to agricul- several studies of vegetation–climate interactions, in- tural changes to water flows (Table 1) operate at a wide cluding both models and statistical analysis of observations range of spatial scales and are reversible at different [41,45,46]. Amazonian vegetation has in other studies temporal scales (Figure 2). Agriculture–aquatic system shown a surprising resilience to drought [47], but these regime shifts occur at the watershed to river basin scales studies did not include the impact of agricultural fires, which but vary from years to millennia in their reversibility. For can reduce the resilience of forest to drought [48]. Another example, freshwater eutrophication is often irreversible, or regime shift with even less evidence is shifts in monsoon only reversible after massive reductions of phosphorus behavior as a consequence of land cover change. It has been inputs for decades or longer, owing to internal cycling of suggested that monsoon systems can exhibit regime shifts phosphorus within the lake system and accumulation of where the strength of the monsoon can influence vegetation, phosphorus in watershed soils [20]. Agriculture–soil and where the strength of the monsoon also depends upon regime shifts tend to operate at field to landscape scale that vegetation [40,49]. Monsoon collapse has been with varying degrees of reversibility. Although soil struc- suggested to be a driving force in the Sahel regime shift ture regime shifts occur at small spatial scales, their [40,42] and vegetation regime shifts in Australia [50]. impact can cascade across the landscape, as exemplified At a smaller scale, agriculture–atmosphere regime shifts by the development of the Dust Bowl in the 1930s in the also can occur in cloud or fog forests, where moisture that US. The Dust Bowl started at the scale of individual fields vegetation intercepts from fog allows the vegetation to and expanded nonlinearly to impact the agricultural

215 Review Trends in Ecology and Evolution Vol.23 No.4

Figure 2. Estimates of the spatial and temporal scales at which regime shifts operate. Blue indicates agriculture and aquatic systems, white indicates agriculture and soil, and green indicates agriculture and atmosphere regime shifts. regions of the US [12]. Broad-scale weather patterns these examples show, hydrological consequences of climate caused individual fields to become highly connected, creat- change might move ecosystems closer to some critical ing massive dust storms that nonlinearly aggravated the thresholds and away from others, thereby influencing their situation [12]. vulnerability to other agriculturally induced changes in Finally, agriculture–atmosphere regime shifts tend to hydrology. operate at relatively large spatial and temporal scales, although uncertainty remains about the important scales Enhancing resilience of agricultural landscapes of land–atmosphere feedbacks. For example, although Hydrological alterations due to growing agricultural evapotranspiration from the forests is the main source of demands (Box 1) are likely to increase the risk of surprising water for precipitation in the Amazon, patchy regional regime shifts unless management practices are changed. deforestation that increases landscape heterogeneity can The expected alterations can be reduced by improving the contribute to an increase in rainfall through the establish- productivity of water in agriculture (Box 1), but enhancing ment of anomalous convective circulations, whereas large- resilience to the regime shifts discussed here requires scale deforestation would substantially decrease precipi- active management of ecosystem processes across agricul- tation even in very distant places [53]. tural landscapes. Avoiding the discussed regime shifts is thus not only a question of improving management of Agriculture interacts with other drives to produce water or agriculture. Here we identify some general strat- water-related regime shifts egies of enhancing ecosystem resilience to water–agricul- Regime shifts are triggered by the interaction of changes ture regime shifts by focusing on ecology as a vital part of internally in a system and changes in external drivers (Box resilience building. 2). In Table 1, we identify critical internal ‘slow variables’ that strongly influence the vulnerability of an ecosystem to From crop optimization to functional diversity regime shifts. Managing these slow variables to maintain The management of agricultural systems has tended to resilience can thus be an important management strategy. focus on maximization of yield, emphasizing the pro- We also identify the external drivers that can produce duction of a single ecosystem service at the expense of regime shifts, separating these into agricultural and non- other ecosystem services. Changing the focus of agricul- agricultural drivers. In many cases, agricultural drivers tural management to how to reliably and profitably pro- interact with other external drivers, such as climate duce food while also producing other ecosystem services change, in triggering regime shifts. For example, reduced was a key recommendation of the MA and CA [1,7,8]. This soil organic matter, a critical slow variable, can lead to change in focus could help increase resilience to regime decreased water holding capacity. Less water in the soil shifts by maintaining or enhancing functional and reduces capacity to cope with a high frequency of dry spells response diversity in agricultural landscapes [54]. Func- [35,36]. In the Mississippi Valley, increasing precipitation tional diversity is the species diversity that maintains a in the late fall and spring influences nitrogen runoff, which specific ecosystem function, whereas response diversity is expands the size of the hypoxic zone in the Gulf of Mexico the diversity of responses that different species have to [22]. It has been suggested that the recent drought in the variations, and they are particularly important for reorga- Murray-Darling Basin in Australia has reduced the rate of nization after disturbance [10,54]. One of the major ways in dryland salinization expansion because the drought has which regime shifts can become irreversible is if the eco- kept water tables low. Consequently, the return of wetter logical processes that maintain a regime vanish from the conditions could have disastrous consequences [33].As landscape. For example, fragmentation of the Australian

216 Review Trends in Ecology and Evolution Vol.23 No.4

Box 3. Locating regime shifts in the Earth system

Agricultural and other land use activities have often been seen as a various attempts to identify regions vulnerable to changes in land– local management issue despite pervasive global impacts [1]. Most atmosphere interactions [39,68,69]. One of the factors that appear to evidence for regime shifts comes from local case studies. There are control vulnerability to agriculture–atmosphere regime shifts is the few analyses of the larger-scale distribution of regime shifts and their extent to which local evapotranspiration versus evapotranspiration potential implications on the Earth system, despite a need for from oceans influences precipitation at different scales. In Figure I,we improved capacity of analyzing regime shifts at the global scale. show where areas of deforestation and irrigation already have Hotspots of expected aquatic regime shifts, driven by increased modified evapotranspiration [3], and where these areas coincide with fertilizer use, can be found around the globe in areas that combine regions where soil moisture has been suggested to have a substantial intensive agriculture, high fertilizer use, high animal density, naturally impact on rainfall [69]. Changes in evapotranspiration owing to nutrient-rich soils, high rainfall and high rates of soil erosion [21]. All deforestation for agriculture or irrigation are shown in yellow to red the soil regime shifts occur in drylands. Drylands are also the areas for decrease, blue for increase and green for no change (where gray is with the largest water challenges for reaching the United Nations nonagricultural lands) [3]. The regions where precipitation (summer Millennium Development Goals on hunger and malnutrition [65], and rainfall) is suggested to be strongly affected by soil moisture have been identified as particularly vulnerable to a complex set of anomalies was identified in an assessment of land–atmosphere interacting problems including land degradation, persistent poverty, coupling strength across 12 global land–atmosphere climate models population growth and remoteness of decision making and infra- [69]. The circles in the figure indicate where these regions coincide structure [6,67]. Hotspots related to soil regime shifts are thus likely to with areas of changes in evapotranspiration. These overlaps suggest occur in drylands with large populations (or population growth), high regions in the Earth system that could be the focus for more in-depth poverty and large needs-increased food production. There have been analysis.

Figure I. Regions where deforestation and irrigation already have modified evapotranspiration, and where they coincide with areas where soil moisture has been suggested to have a substantial impact on rainfall.

landscape altered hydrology producing dryland salinity, Spatial heterogeneity can, for example, play a vital role in but fragmentation also reduced plant reproduction and identifying and managing particularly vulnerable regions propagule dispersal, reducing the ability of the systems as well as critical areas that have a disproportionately high to recover from salinization when hydrology is restored influence on the risk for regime shifts in the larger region. [32]. This illustrates the importance of sustained habitats Because of variation in ecological processes across a land- for species that can connect the landscape (so-called mobile scape, some locations will be more vulnerable to regime links) by providing ecological functions such as seed dis- shifts than others. By identifying locations where the persal [55]. forces that maintain an existing regime are weak, managers can identify where regime shifts are likely to occur From homogeneous fields to heterogeneous landscapes [56]. For example, some savanna systems can exhibit We showed earlier that changes in vegetation patterns regime shifts between forest and savanna, whereas other from overgrazing can increase the risk for regime shifts in sites that are wetter or drier only have one possible regime arid to semiarid systems [30], whereas stimulating water [31].InBox 3, we discuss vulnerable areas at a global scale. infiltration at critical spots can help restore the system Identifying critical areas can allow managers to focus their [29]. Spatial heterogeneity can also be an important factor rehabilitation or monitoring efforts where they are likely to in other types of agricultural landscapes, and for building have the greatest effect. For example, relatively small resilience to other types of water-related regime shifts. areas in watersheds that combine high soil phosphorus

217 Review Trends in Ecology and Evolution Vol.23 No.4 concentrations and high runoff potential are disproportio- statistical methods specifically designed to detect evidence nately responsible for the majority of phosphorus runoff of abrupt shifts in long-term data [63]. Ecologists need to be into freshwater lakes [57]. Water quality can thus be involved in making sure these goals are met. Although this substantially improved by managing these ‘critical source is a complex and difficult endeavor, one way to start is to areas’ rather than managing an entire catchment [58]. identify key regions based on where we expect regime shifts to be more likely or more common, and focus research From stability to dynamics and management efforts on these (Box 3). Ecosystems change as a result of internal and external processes, but management and policy often assume them Conclusions and research challenges to be relatively stable, neglecting to plan or manage for There is strong evidence that agricultural modification of disturbance and reorganization [11]. Management that water flows can produce a variety of ecological regime shifts adapts to variation in external drivers can increase resili- that operate across a range of spatial and temporal scales. In ence. For example, grazing management that accounts for a world of growing demands for water, agricultural products variability of rainfall rather than considering only average and other ecosystem services, there will inevitably be eco- conditions can increase rangeland productivity [59]. Alter- logical surprises. Preparing for these surprises is essential natively, extreme events can be used to engineer a regime to maintain ecosystem services of importance for human shift. For example, rainy periods associated with El Nino well-being. Preparation requires understanding the forces can be used in combination with grazing reduction to that drive these regime shifts, as well as better methods restore degraded ecosystems, whereas grazing reduction designed specifically to anticipate and analyze them. Iden- alone is insufficient to achieve this goal [60]. tifying ways of building resilience to these shifts illustrates Changes in internal variables that alter the processes the need to manage agriculture as an embedded part of that define a regime (Box 2) are often relatively slower than larger landscapes, with special attention to the internal and other monitored system dynamics, such as yield levels. external dynamics that drive change in inter-linked agricul- Change in these slow variables might go unnoticed over tural, hydrological and ecological processes. Achieving this long time periods, because researchers tend not to measure will require increased scientific and policy collaborations them until the system shifts to a new regime. For example, among ecologists, agronomists, hydrologists and global despite some early indications of how clearing of woody change researchers. vegetation could increase risks of salinization in Australia already in the 19th century, Australians were in general Acknowledgements unaware of this phenomenon until salinization started to L.J.G.’s research was funded by the Swedish Research Council for the become widespread [33]. Identifying and monitoring slowly Environment, Agricultural Sciences and Spatial Planning (Formas), and G.D.P.’s work was funded by the Canada Research Chairs Program. We changing key ecological processes, such as the water table wouldalsoliketothankElinEnfors,Fre´de´ric Guichard, Navin rise in Australia, can be used to predict the likelihood of a Ramankutty and Maria Tengo¨ for comments on the manuscript, and regime shift. The ability to manage systems to avoid Tove Gordon and Robert Kautsky for help with figures. regime shifts in a world of high variability would be improved if we could predict how close an ecosystem is References to critical thresholds. The creation of methods that provide 1 Foley, J.A. et al. (2005) Global consequences of land use. Science 309, early warning of regime shifts is an important scientific 570–574 challenge, and research suggests that changes in pattern 2 Scanlon, B.R. et al. (2007) Global impacts of conversions from natural to formation and rising variance could be used to detect when agricultural ecosystems on water resources: quantity versus quality. Water Resour. Res. 43, W03437 DOI: 10.1029/2006WR005486 (http:// and where systems are becoming more likely to experience www.agu.org/journals/wr) regime shifts [30,61]. 3 Gordon, L.J. et al. (2005) Human modification of global water vapor flows from the land surface. Proc. Natl. Acad. Sci. U. S. A. 102, 7612–7617 From disciplinary divides to hydrological integration 4 Galloway, J.N. et al. (2004) Nitrogen cycles: past, present, and future. Biogeochemistry 70, 153–226 The identified regime shifts (Table 1) created by agricul- 5 Bennett, E.M. et al. (2001) Human impact on erodable phosphorus and tural modification of the hydrological cycle present a chal- eutrophication: a global perspective. Bioscience 51, 227–234 lenge to ecosystem governance, because water often 6 Millennium Ecosystem Assessment (2005) Ecosystems and Human transmits the consequences of change to locations which Well-Being: Synthesis, Island Press are spatially and temporally separated from the place 7 Comprehensive Assessment of Water Management in Agriculture (2007) Water for Food, Water for Life: A Comprehensive Assessment where the change occurred (Figure 2). Coping with these of Water Management in Agriculture, Earthscan Publications disconnects can be improved by governance systems that 8 Carpenter, S.R. et al. (2006) Millennium ecosystem assessment: connect local agricultural management practices to the research needs. Science 314, 257–258 scales that key ecological processes operate [62]. Because 9 Peterson, G.D. et al. (2003) Uncertainty and the management of regime shifts are often a surprising outcome of slow multistate ecosystems: an apparently rational route to collapse. Ecology 84, 1403–1411 change, governance that is able to learn how to effectively 10 Folke, C. et al. (2004) Regime shifts, resilience, and biodiversity anticipate, avoid and respond to abrupt ecological change in ecosystem management. Ann. Rev. Ecol. Evol. Syst. 35, will be better prepared for future surprises [62]. Achieving 557–581 this goal requires better biophysical understanding of the 11 Gunderson, L. and Holling, C., eds (2002) Panarchy: Understanding key feedback processes that connect regime shifts at differ- Transformations in Human and Natural Systems, Island Press 12 Peters, D.P.C. et al. (2004) Cross-scale interactions, nonlinearities, and ent scales [17,43,44]. Discovering regime shifts in empiri- forecasting catastrophic events. Proc. Natl. Acad. Sci. U. S. A. 101, cal data also requires the development and use of 15130–15135

218 Review Trends in Ecology and Evolution Vol.23 No.4

13 Rial, J.A. et al. (2004) Nonlinearities, feedbacks and critical thresholds 42 Foley, J.A. et al. (2003) Regime shifts in the Sahara and Sahel: within the Earth’s climate system. Clim. Change 65, 11–38 interactions between ecological and climatic systems in northern 14 Rietkerk, M. et al. (2004) Self-organized patchiness and catastrophic Africa. Ecosystems (N. Y., Print) 6, 524–539 shifts in ecosystems. Science 305, 1926–1929 43 Dekker, S.C. et al. (2007) Coupling microscale vegetation-soil water 15 Scheffer, M. et al. (2001) Catastrophic shifts in ecosystems. Nature 413, and macroscale vegetation-precipitation feedbacks in semiarid 591–596 ecosystems. Glob. Change Biol. 13, 671–678 16 Walker, B. and Meyers, J.A. (2004) Thresholds in ecological and social- 44 Los, S.O. et al. (2006) An observation-based estimate of the strength of ecological systems: a developing database. Ecol. Soc. 9, 3 rainfall-vegetation interactions in the Sahel. Geophys. Res. Lett. 33, 17 Scheffer, M. et al. (2005) Synergy between small- and large-scale L16402 DOI: 10.1029/2006GL027065 (http://www.agu.org/journals/gl) feedbacks of vegetation on the water cycle. Glob. Change Biol. 11, 45 Oyama, M.D. and Nobre, C.A. (2003) A new climate-vegetation 1003–1012 equilibrium state for tropical South America. Geophys. Res. Lett. 30, 18 Ripl, W. (2003) Water: the bloodstream of the biosphere. Philos. Trans. 2199 DOI: 10.1029/2003GL018600 (http://www.agu.org/journals/gl) R. Soc. Lond. B Biol. Sci. 358, 1921–1934 46 Hutyra, L.R. et al. (2005) Climatic variability and vegetation 19 Carpenter, S.R. (2003) Regime Shifts in Lake Ecosystems: Pattern and vulnerability in Amazonia. Geophys. Res. Lett. 32, L24712 DOI: Variation, Ecology Institute 10.1029/2005GL024981 (http://www.agu.org/journals/gl) 20 Carpenter, S.R. (2005) Eutrophication of aquatic ecosystems: 47 Cowling, S.A. and Shin, Y. (2006) Simulated ecosystem bistability and soil phosphorus. Proc. Natl. Acad. Sci. U. S. A. 102, threshold responses to co-varying temperature, precipitation and

10002–10005 atmospheric CO2 within a region of Amazonia. Glob. Ecol. 21 Diaz, R.J. (2001) Overview of hypoxia around the world. J. Environ. Biogeogr. 15, 553–566 Qual. 30, 275–281 48 Nepstad, D. et al. (2004) Amazon drought and its implications for forest 22 Donner, S.D. and Scavia, D. (2007) How climate controls the flux of flammability and tree growth: a basin-wide analysis. Glob. Change nitrogen by the Mississippi River and the development of hypoxia in Biol. 10, 704–717 the Gulf of Mexico. Limnol. Oceanogr. 52, 856–861 49 Zickfeld, K. et al. (2005) Is the Indian summer monsoon stable 23 Scheffer, M. and van Nes, E.H. (2007) Shallow lakes theory revisited: against global change? Geophys. Res. Lett. 32, L15707 DOI: 10.1029/ various alternative regimes driven by climate, nutrients, depth and 2005GL022771 (http://www.agu.org/journals/gl) lake size. Hydrobiologia 584, 455–466 50 Miller, G. et al. (2005) Sensitivity of the Australian monsoon to 24 Hooke, J. (2003) River meander behaviour and instability: a framework insolation and vegetation: implications for human impact on for analysis. Trans. Inst. Brit. Geograph. 28, 238–253 continental moisture balance. Geology 33, 65–68 25 Knox, J.C. (2006) Floodplain sedimentation in the Upper Mississippi 51 del-Val, E. et al. (2006) Rain forest islands in the Chilean semiarid Valley: natural versus human accelerated. Geomorphology 79, 286–310 region: fog-dependency, ecosystem persistence and tree regeneration. 26 Simon, A. and Rinaldi, M. (2006) Disturbance, stream incision, and Ecosystems (N. Y., Print) 9, 598–608 channel evolution: the roles of excess transport capacity and boundary 52 Dawson, T.E. (1998) Fog in the California redwood forest: ecosystem materials in controlling channel response. Geomorphology 79, 361–383 inputs and use by plants. Oecologia 117, 476–485 27 Dent, C.L. et al. (2002) Multiple states in river and lake ecosystems. 53 D’Almeida, C. et al. (2007) The effects of deforestation on the Philos. Trans. R. Soc. Lond. B Biol. Sci. 257, 635–645 hydrological cycle in Amazonia: a review on scale and resolution. 28 Peters, D.P.C. et al. (2006) Disentangling complex landscapes: new Int. J. Climatol. 27, 633–647 insights into arid and semiarid system dynamics. Bioscience 56, 491–501 54 Elmqvist, T. et al. (2003) Response diversity, ecosystem change, and 29 Ludwig, J.A. et al. (2005) Vegetation patches and runoff-erosion as resilience. Front. Ecol. Environ. 1, 488–494 interacting ecohydrological processes in semiarid landscapes. Ecology 55 Lundberg, J. and Moberg, F. (2003) Mobile link organisms and 86, 288–297 ecosystem functioning: implications for ecosystem resilience and 30 Kefi, S. et al. (2007) Spatial vegetation patterns and imminent management.. Ecosystems (N. Y., Print) 6, 87–98 desertification in Mediterranean arid ecosystems. Nature 449, 213–217 56 Peterson, G.D. (2002) Estimating resilience across landscapes. Cons. 31 Sankaran, M. et al. (2005) Determinants of woody cover in African Ecol. 6, 17 savannas. Nature 438, 846–849 57 Nowak, P. et al. (2006) Disproportionality as a framework for linking 32 Cramer, V.A. and Hobbs, R.J. (2005) Assessing the ecological risk from social and biophysical systems. Soc. Nat. Resour. 19, 153–173 secondary salinity: a framework addressing questions of scale and 58 Sharpley, A.N. et al. (2001) Assessing site vulnerability to threshold responses. Austral. Ecol. 30, 537–545 phosphorus loss in an agricultural watershed. J. Environ. Qual. 30, 33 Anderies, J.M. et al. (2006) Loss of resilience, crisis, and institutional 2026–2036 change: lessons from an intensive agricultural system in southeastern 59 Janssen, M.A. et al. (2004) Robust strategies for managing rangelands Australia. Ecosystems (N. Y., Print) 9, 865–878 with multiple stable attractors. J. Environ. Econ. Manag. 47, 140–162 34 Fernandez, R.J. et al. (2002) Degradation and recovery in socio- 60 Holmgren, M. et al. (2006) Extreme climatic events shape arid and ecological systems. In Global Desertification: Do Humans Cause semiarid ecosystems. Front. Ecol. Environ. 4, 87–95 Deserts? (Reynolds, J.F. and Smith, D.M.S., eds), pp. 297–323, 61 Carpenter, S.R. and Brock, W.A. (2006) Rising variance: a leading Dahlem University Press indicator of ecological transition. Ecol. Lett. 9, 308–315 35 Bossio, D. et al. (2007) Conserving land – protecting water. In Water for 62 Folke, C. et al. (2005) Adaptive governance of social-ecological systems. Food, Water for Life: A Comprehensive Assessment of Water Ann. Rev. Environ. Res. 30, 441–473 Management (Molden, D., ed.), pp. 551–583, Earthscan 63 Scheffer, M. and Carpenter, S.R. (2003) Catastrophic regime shifts in 36 Enfors, E.I. and Gordon, L.J. (2007) Analysing resilience in dryland ecosystems: linking theory to observation. Trends Ecol. Evol. 18, 648– agro-ecosystems: a case study of the Makanya catchment in Tanzania 656 over the past 50 years. Land Degrad. Dev. 18, 680–696 64 Food and Agriculture Organization of the United Nations (2007) 37 Antle, J.M. et al. (2006) Multiple equilibria, soil conservation FAOSTAT Database on Agriculture, FAO (http://faostat.fao.org) investments, and the resilience of agricultural systems. Environ. 65 Rockstrom, J. et al. (2007) Assessing the water challenge of a new green Dev. Econ. 11, 477–492 revolution in developing countries. Proc. Natl. Acad. Sci. U. S. A. 104, 38 Makarieva, A.M. and Gorshkov, V.G. (2007) Biotic pump of 6253–6260 atmospheric moisture as driver of the hydrological cycle on land. 66 Rockstrom, J. and Barron, J. (2007) Water productivity in rainfed Hydrol. Earth Syst. Sci. 11, 1013–1033 systems: overview of challenges and analysis of opportunities in 39 Feddema, J.J. et al. (2005) The importance of land-cover change in water scarcity prone savannahs. Irrig. Sci. 25, 299–311 simulating future climates. Science 310, 1674–1678 67 Reynolds, J.F. et al. (2007) Global desertification: building a science for 40 Higgins, P.A.T. et al. (2002) Dynamics of climate and ecosystem dryland development. Science 316, 847–851 coupling: abrupt changes and multiple equilibria. Philos. Trans. R. 68 Delire, C. et al. (2004) Long-term variability in a coupled atmosphere- Soc. Lond. B Biol. Sci. 357, 647–655 biosphere model. J. Clim. 17, 3947–3959 41 Sternberg, L.D.L. (2001) Savanna-forest hysteresis in the tropics. Glob. 69 Koster, R.D. et al. (2004) Regions of strong coupling between soil Ecol. Biogeogr. 10, 369–378 moisture and precipitation. Science 305, 1138–1140

219 A review of the regeneration dynamics of North American boreal forest tree sp... D F Greene; J C Zasada; L Sirois; D Kneeshaw; et al Canadian Journal of Forest Research; Jun 1999; 29, 6; CBCA Reference and Current Events pg. 824

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Creating New Landscapes and Ecosystems The Alberta Oil Sands

E.A. Johnsona and K. Miyanishib aBiogeoscience Institute and Department of Biological Sciences, University of Calgary, Calgary, Alberta, Canada bDepartment of Geography, University of Guelph, Ontario, Canada

Extraction of oil from the Alberta Oil Sands through surface mining involves the removal of the overburden and oil sand to a depth of up to 100 m and over extremely large areas. While the operation of the bitumen processing plants has serious environmental impacts on downstream habitats, this article focuses on the reclamation of areas from which the oil sands have been removed, processed, and returned. This reclamation following closure of the mines will entail the complete re-creation of landforms and ecosystems at a landscape scale, with the goal of producing suitable habitats for plants, animals, and people. Such projects will require a reasonable understanding of the geophysical and ecological processes that operate at a wide range of scales. Some information is provided on the climate, hydrology, vegetation, and land use (past and current) of the Oil Sands area, situated within the Boreal Plain ecozone, to provide a framework for discussion of issues to be addressed in, and proposed guidelines for, such large-scale reclamation. Although none of the mines has yet closed, numerous consultant reports have been produced with recommendations for various aspects of such reclamation projects (e.g., wetland hydrology, vegetation, wildlife habitat). The scientiﬁc basis of such reports is found to vary with respect to depth of understanding of the relevant processes.

Key words: restoration; surface mining; tailings; ecohydrology; boreal forest; wetlands; peatlands; landscape ecology; oil development; heavy oil; meromictic lakes; natural disturbance paradigm

Introduction large-scale impacts are contemplated or carried out. Despite landscape ecology’s attempts Conservation biology and restoration ecol- at pattern analysis in landscapes and conser- ogy have for decades tried to alert us to prob- vation biology’s recent interest in large-scale lems and provide solutions based on some sci- conservation planning and strategies, the phys- ence, expert opinion, experience, and a logical ical environment template at large scales has feeling for the ecological system. Most conser- been mostly explored by ecohydrologists, geo- vation and restoration efforts have been focused morphologists, biogeochemists, and meteorolo- on single species, particularly mammals, birds, gists/climatologists. Conservation and restora- and plants, speciﬁc taxonomic assemblages tionatlandscapescaleshaveoftenassumedthat (e.g., neotropical migrants), or certain habitats the degradation is temporary and can be re- (e.g., wetlands). Nothing is wrong with this ap- paired over relatively short time scales of 50 or proach, but it presents obvious limitations when so years. What about the creation of completely new landscapes over time spans of 100 or more years? Address for correspondence: E.A. Johnson, Department of Biological Sciences, University of Calgary, Calgary, AB T2N 1N4, Canada. Voice: The Alberta Oil Sands is a large-scale de- +1-403-220-7635; fax: +1-403-289-9311. [email protected] velopment that will require reconstruction of

Ann. N.Y. Acad. Sci. 1134: 120–145 (2008). C 2008 New York Academy of Sciences. doi: 10.1196/annals.1439.007 120 Johnson & Miyanishi: Creating New Landscapes at Alberta Oil Sands 121

Figure 1. Location and distribution of oil sands in northern Alberta. Part of the Athabasca deposit has reserves shallow enough to make surface mining feasible (mineable area in black). Oil from the deeper deposits (core deposits in gray) is removed using in situ processes involving steam injection through wells into the oil sand. Source: Schneider (2002). ecosystems at the scale of whole landscapes. km2 currently under lease. Most of the areas The heavy oil sands are located in the Cold with underlying oil sands are also under For- Lake, Peace River, and Athabasca regions of est Management Agreement (FMA) leases to northern Alberta (Fig. 1) and cover approxi- forestry companies (Fig. 2). Thus, since 1967 mately 140,000 km2, an area larger than the and particularly in the last 20 years, oil sands state of Florida, or 23% of the Province (Oil exploitation and forestry leases have been se- Sands Consultations 2007). Prior to recent oil riously fragmenting the landscape (Schneider and forestry developments, the area of the Oil 2002). Climate change will further complicate Sands (except some areas around Peace River) any changes in the future. was largely unfragmented and undisturbed by With improved technology, the operating Europeans. It was still largely inhabited by in- cost of producing oil sand oil has gone from digenous (First Nation) peoples who tried to about $35 per barrel in 1980 to $20 per bar- follow a lifestyle in which their environment rel in recent years. Both the lowered cost of played an important practical and spiritual role. production and the increasing price of oil (re- Development for oil extraction was begun in cently surpassing $100 per barrel) have resulted 1967 by the Great Canadian Oil Sands Com- in an increased demand for this oil source. The pany (now Suncor). There are now ﬁve leases Alberta Oil Sands supplies one-third of the oil and several under consideration with 43,000 production of Canada, and Canada is the main 122 Annals of the New York Academy of Sciences

gave control of the land to the government and resulted in the establishment of First Nation reserves while allowing continued use of nonre- serve crown lands for subsistence activities such as hunting, fishing, and trapping. The reserves are under federal jurisdiction and, at present, only one First Nation group has requested and received federal approval for development of an oil sands mine on its reserve (Denstedt & Jamieson 2007). The removal of oil sands oil is by two methods. The near-surface deposits (classified as “mineable” in Fig. 1) are recovered by removing up to 100 m of surface material (overburden) to access the oil-bearing sedimentary layers, processing the oil sands, and then eventually replacing the mine tailings and overburden. About 2 tons of oil sands are required to produce a barrel of oil (Oil Sands Consultations 2007). Deeper deposits are removed in situ by steam assisted gravity drainage (SAGD). Both processes extract bitumen (a heavy carbon- Figure 2. Boundaries of the Forest Management Agreement (FMA) leases in Alberta as of 2006 (gray rich, hydrogen-poor hydrocarbon), which is up- areas). The lease labeled 2 (shown in dark gray) is graded to synthetic crude by removing carbon held by Alberta-Pacific, and the one labeled 4 is the and sulfur and adding hydrogen. Production of Cold Lake Air Weapons Range. The area labeled 1m3 of bitumen (1 m3 = 6.29 barrels) requires WBNP (north of the Oil Sands) refers to Wood Buf- approximately 125 and 214 m3 of natural gas falo National Park. The striped areas are classified by and 2–4.5 and 0.2 m3 of water by mining and Alberta as the White Zone where agriculture is permitted and most human settlement is concentrated. SAGD methods, respectively (National Energy Source: Alberta Sustainable Resource Development Board 2007). The lower water use by SAGD (2007). is due to the 90–95% recycling of the steam water. The chemical processes converting bi- supplier of oil to the United States (National tumen to synthetic crude oil also release SO2, Energy Board 2007). NOx, and volatile organics (Golder Associates In Canada, the Provinces have greater con- Ltd. 2003; RWDI AIR Inc. 2005a, 2005b; trol over nonrenewable and renewable re- Canadian Centre for Energy Information sources in their jurisdiction than does the fed- 2007). In this chapter, we focus solely on mineral government, but they must share part of the ing of near-surface deposits for two reasons: it is revenue from these resources with the rest of the currently responsible for most of the Oil Sands country.Also, as is the case with the Alberta Oil oil production, and it is the method that has Sands, most of the mineral rights are owned by the most complete and long-term impact on the Province and leased to private companies. ecosystems. As mentioned previously, the Oil Sands area The Alberta government mandates that in- includes land traditionally used by First Nation dustry is to create functioning ecosystems from peoples. All but one of the First Nation bands this landscape of disrupted ground and sur- in northern Alberta signed a Treaty Land En- face water hydrology, changed biogeochem- titlement Settlement Agreement in 1899 that istry, and totally removed boreal ecosystems Johnson & Miyanishi: Creating New Landscapes at Alberta Oil Sands 123

(Alberta Environment 1999). Given the large area of each mine (ca. 100 km2)eventually to be stripped and mined, this is not a restoration problem but the engineering of new ecosystems. The re-creation must achieve maintenance-free, self-sustaining ecosystems with capabilities equivalent to or better than the predisturbance conditions (Alberta Environ- ment 1999). Predisturbance land-use capabilities include timber harvesting, wildlife habitat, watershed functions, wetlands, sources of traditional foods and medicinal plants, and recreation (Oil Sands Vegetation Reclamation Com- mittee 1998). The industry must report each Figure 3. Distribution of the Boreal Plain in western Canada (gray) and location of the Alberta Oil year to both Alberta Environment and Alberta Sands area within the Boreal Plain. Source: Environ- Sustainable Resource Development, and plans ment Canada (2007). must address the final land-use stakeholders’ needs and consult with them. hundred years. As mentioned, reclamation of Two points should be made clear at the out- the Oil Sands mined areas will not be an eco- set. First, we will not make an argument here for logical restoration project but will require the a return to some presettlement or more pristine re-creation of the physical template and the environment, mostly because we feel that this ecosystems on it. is probably impossible considering the issues of In this chapter we will limit ourselves, as the both the scale of the development and climate title states, to discussion of the ideas and lim- change. Paleoecology has long recognized that ited results available on re-creating the land- ecosystems have never been the same through scape and ecosystems on the near-surface mine time but have constantly changed in composi- deposits. Up to this time, no mines have been tion as they adjust to their changing environ- closed so no re-creation has begun although ments (Davis 1976; Williams et al. 2007). This some small-scale pilot studies and reclamation makes it difficult to set benchmarks, particu- projects on mine tailings have been started. larly those based on some past or present time Most mines will operate for up to 40 years. with some presumed composition and environ- Further, while we are aware that the bitumen ment. In the past, ecosystems have been assem- processing plants produce important environ- bled and reassembled with relative ease in re- mental impacts downstream in rivers and in the sponse to environmental changes. However, as airshed due to pollution as well as energy and is implied in the government mandate above, water use, we will not consider these here but conditions under which natural processes are focus instead on the surface mining to extract allowed to recover must be re-created and also bitumen. ecosystem services must be maintained. The trade-offs are as yet unknown. Second, the time over which ecosystems have made adjustments to major changes in the past Boreal Plain Ecosystem has been long, not the short time over which economies usually adjust to changes in markets. The Oil Sands region is located in the North The re-creation of the landscape in the Alberta American Boreal Plain (Fig. 3), a dissected, rel- Oil Sands region will take a longer time than atively flat (400–800 m asl) region that was seems to be planned—probably more than a covered by the Laurentide ice sheet 10,000 to 124 Annals of the New York Academy of Sciences

12,000 years ago. The surficial glacial deposits slopes have white spruce and balsam fir (Abies are deep (30–200 m), loamy till and gravel–sand balsamea (L.) Mill.), while the wet basal slopes glaciofluvial and lacustrine deposits. These de- have black spruce (Picea mariana (Mill.) B.S.P.). posits overlie the Mesozoic- and Cenozoic-age On the better-drained and more nutrient-poor sedimentary rocks (largely carbonate) compris- glaciofluvial hillslopes, the dry hilltops are dom- ing most of the bedrock and containing the oil inated by jack pine (Pinus banksiana Lamb.) and deposits. The landscape has numerous lakes, the basal slopes by black spruce. Actual slopes ponds, and peatlands. are often more heterogeneous than this ideal- The climate is subhumid and midboreal, ized gradient because of factors, such as prox- with mean temperatures of −2to+1◦C; sum- imity of seed sources, effects of past distur- mers average 13 to 15◦C and winters −18 to bances (see below), and variable substrate and −14◦C. Precipitation is 300 to 600 mm with groundwater flow patterns. about 70% as wet precipitation. Annual poten- Poorly drained lowlands are usually covered tial evapotranspiration is greater than precipi- by peat. Shrub fens are dominated by willows tation. Potential evapotranspiration is relatively (Salix spp.) and sedges (Carex spp.), and forest constant while precipitation varies from year to fens by tamarack (Larix laricina (Du Roi) K. year (Bothe & Abraham 1993). Stream flow is Koch) and black spruce. Patterned fens are very low and varies from year to year due to variable common. Bogs, which occur mostly as islands water storage. The annual water deficit is from in large fens or in small potholes, are dominated 40 to 60 mm. Since peak precipitation occurs by short black spruce and Sphagnum moss. Peat- between June and August when the vegetation lands cover 103,200 km2 of Alberta (16.3% of is actively growing and transpiring, there is little the land base) and are most extensive in the saturated overland flow (Devito et al. 2005b). northern two-thirds of the Province (Vitt et al. The shape and composition of the land- 1996). They cover about 30% of the Oil Sands. scape and its climate and hydrology determine Wildfires were the principal determinant of the vegetation composition. The age of the the forest age mosaic before logging. Most forest is largely determined by natural distur- of the area burned by wildfire in the Boreal bances, usually insect outbreaks and wildfire. Plain is caused by lightning fires (Nash & John- Landscapes are made up of stream courses, son 1996). These wildfires occur mostly dur- depressions, and ridgelines with hillslopes being persistent high pressure systems that last tween them. Hillslopes are generally drier at longer than 10 to 15 days. The most common the top and wetter at the bases due to differ- persistent high pressures are associated with ences in the water-contributing area. The sub- the Pacific North America pattern (Johnson strate composition (e.g., till versus glaciofluvial & Wowchuk 1993) which is characterized by deposits) determines the ease of movement of anomalous low pressure in the Gulf of Alaska, water through the soil. Consequently, on hill- high pressure over western Canada, and low slopes the factors of contributing area, slope pressure over southeastern Canada and ad- angle, and substrate transmissivity in general cre- jacent United States. Over decadal scales, ate moisture gradients along which vegetation episodes of large area burned are associated is arranged, based on species’ moisture toler- with the positive mode of the Pacific Decadal ances. In the subdued landscape of the Bo- Oscillation (PDO) (Macias Fauria & Johnson real Plain, this pattern generally determines 2006; Skinner et al. 2006). The persistent posi- the upland vegetation (Bridge and Johnson tive mode of the PDO since 1977 has led to an 2000). Thus, on relatively nutrient-rich glacial increase in area burned compared to the very till hillslopes, the dry hilltops have a mixture low area burned during the immediately prior of aspen (Populus tremuloides Michx.) and white period 1950–1976, when the negative mode of spruce (Picea glauca (Moench) Voss); the mid- the PDO was more persistent. Johnson & Miyanishi: Creating New Landscapes at Alberta Oil Sands 125

The average fire frequency has changed at bedrock or impermeable layers in the glacioflu- least three times in the last 350 years due to vial deposits. large-scale climate changes related particularly The water budget of peatlands is determined to the Little Ice Age (Bergeron & Archambault by their vadose zone storage capacity because 1993; Weir et al. 2000). The fire frequency has of their more negative heat budget due to the always been shorter than the life span of the low thermal capacity of organic matter,the veg- trees (Johnson et al. 1995; Weir et al. 2000). The etation cover, and the low lateral and vertical forests adjacent to and up to 50 km away from conductivity of the peat (Price 2003; Price et al. agricultural settlements had much shorter fire 2005). Peatlands account for 25 to 50% of the frequencies in the early 1900s due to the spread landscape in the Boreal Plain (NWWG 1988). of settlement fires (Weir et al. 2000; Tchir et al. Hillslopes contribute little runoff to the peat- 2004). Most of the area is burned in high inten- lands. During dry weather,water may flow from sity crown fires that kill most of the trees and the peat into the mineral soil upland (Devito remove large amounts of the duff (F and H) et al. 1997). layers of the soil by smoldering combustion; Although the state of knowledge is still very such duff consumption is required for good incomplete, peatlands are a major store of car- tree regeneration (Charron & Greene 2002; bon in the Boreal Plain (Tarnocai et al. 2005). Miyanishi & Johnson 2002). Both the number Peatland surfaces are quite variable over rela- and total area of unburned patches in a burn tively short distances, making it difficult to esti- follow a power-law relationship to the burn mate the flux of carbon dioxide, methane, and area; that is, large fires have more unburned dissolved (in water) carbon (Waddington et al. patches with larger areas than if they were just 1998; Waddington & Roulet 2000). However, it smallfiresscaledup(Johnsonet al. 2003). Old- is clear at this time that peat moisture and tem- growth forests make up a small percentage of perature on different kinds of peat surfaces (for the landscape because of the relatively short example, lawns, pools, and plateaus) influence average interval between fires (Johnson et al. whether carbon is stored or released (Wadding- 1995). ton & Roulet 2000). Many peatlands in the Bo- The hydrology of the Boreal Plain is com- real Plain contain permafrost, which has been plex and does not conform easily to the usual degrading over the last several decades due to idea of simple topographic control (Smerdon global warming (Vitt et al. 1994; Halsey et al. et al. 2005). The general hierarchy of factors 1995; Camill & Clark 1998; Beilman et al. 2001; that control the hydrology in the Boreal Plain Camill 2005). As a result of water conditions is presented in Table 1 (Devito et al. 2005a). The and higher net productivity of Sphagnum moss low ratio of runoff to precipitation (<20%) is on the areas of permafrost melting, these areas a result of the unsaturated (vadose) zone stor- become carbon sinks. However methane emis- age, evapotranspiration (precipitation < po- sions in the decades following permafrost melt- tential evapotranspiration), and vertical flow. ing offset the carbon sink (Turetsky et al. 2002). The glacial substrates have important influ- Also, wildfires may release large amounts of car- ences on the hydrology. Fine texture (gener- bon from peatlands, further offsetting them as ally glacial till) substrates have low permeabil- a sink, particularly if peatland burning were to ity and low infiltration into the unsaturated increase by approximately 17% (Turetsky et al. surface soil; precipitation remains in the root- 2002; Johansson et al. 2006). ing zone and so is easily lost by evapotranspiration. Coarse texture (generally glaciofluvial) Forestry in the Oil Sands Region substrates have high permeability and high infiltration to the water table that reflects the Forestry operations occur on much underlying impermeable layer which may be of the surface of the Oil Sands region; 126 Annals of the New York Academy of Sciences

TABLE 1. Hierarchical classiﬁcation to generalize the dominant controls on water cycling and indices to deﬁne effective hydrologic response units Factor Range of factor Scale

Climate Dry, arid to sub-humid Wet, humid Continental to (P < PET) (P > PET) local • RpoorlycorrelatedwithP • R closely correlated with P • storage or uptake dominates • runoff dominates • tendency for vertical flow • tendency for lateral flow Bedrock geology Permeable bedrock Impermeable bedrock Continental to regional • intermediate to regional flow systems • characterized by local to intermediate flow systems • lack of topographic control on • topographic control on direction of direction of local flow local flow • vertical flow dominates in surface • lateral flow dominates in surface substrate substrate Bedrock slope perpendicular Bedrock slope parallel to land to land surface surface • complex watershed boundaries • simple watershed boundaries • regional aquifer definition needed to determine flow direction Surficial geology Deep substrates Shallow substrates Regional to local • intermediate to regional flow • local flow most probable (but see bedrock geology) Coarse texture Finer texture • vertical flow • lateral flow • deeper subsurface flow • depression storage and/or surface and shallow subsurface flow Spatially heterogeneous Spatially homogeneous deposits deposits • complex groundwater flow systems • simple groundwater flow systems • groundwater flow modeling • surface flow modeling important important Soil type and depth Upland mineral soils Lowland organic soils Local to regional • subsurface flow dominates • return flow and surface overland flow pathways dominate • slow flow generation (matrix flow) • quick flow generation (return flow saturation overland flow) Storage Storage • deeper soils with large water storage • shallower soils with small water potential storage potential • lower specific yield of organic soils and compression leads to surface saturation Transpiration Transpiration • deep roots access stored water • shallower roots limit access to stored water • P ≈ AET during dry periods • AET < PET during dry periods Topography and Gentle slopes Steep slopes Local to regional drainage network • disorganized, inefficient drainage • organized, efficient drainage network network • large groundwater recharge • small groundwater recharge • small, variable runoff yield • large, uniform runoff yield

Source: Devito et al. (2005a). R, runoff; LP, precipitation; AET, actual evapotranspiration; PET, potential evapotranspiration. Johnson & Miyanishi: Creating New Landscapes at Alberta Oil Sands 127

Figures 1 and 2 show the distribution of the ulated forest, the idea was to have the harvest oil sands and the FMA leases in northern Al- not exceed the total of the forest growth minus berta. These operations create a competing the loss from natural causes (e.g., fires, insect land-use and ecological effect. Forestry before outbreaks) (Schneider 2002). Over time, the the 1950s consisted of selective cutting of large Province put in place further limitations with and straight trees in accessible areas adjacent respect to stream and river courses, wetlands, to streams and rivers. In the mostly even-aged wildlife (game), and so on. stands of the boreal forest, such harvesting does In the 1980s two factors led to an increase not lead to good regeneration. Also, before in forestry leases in northern Alberta. The first 1950 there was little regulation and record of was advancements in pulp technology that al- cutting. However, logging was limited in scope lowed the use of hardwood species, not just and slow because of the equipment, which conifers. This opened pulp production to the consisted mostly of handsaws, axes, and horse use of aspen. The second factor was the grow- hauling. ing world demand for pulp products (Pratt & In 1948 the Province of Alberta established Urquhart 1994). Also, at this time both the the Green Zone, restricting agriculture from forestry companies’ and the provincial govern- lands that could best be used in forest pro- ment’s ideas of sustainability became more in- duction (Schneider 2002). Lands in the Green clusive of other parts of the forest ecosystem Zone were retained largely by the Province (Alberta Forest Conservation Strategy 1997). (i.e., Crown land). The Forest Act was pro- Alberta-Pacific’s FMA lease (Fig. 2) was mulgated in 1949 and instituted 20-year leases issued in 1990. Most of the Oil Sands with sustained yield as the objective. By the developments are in the Alberta-Pacific lease 1950s both natural silviculture and sustained- of 58,000 km2 (see Figs. 1 and 2). The com- yield forest management had been introduced. pany harvests about 16,000 ha per year in the These changes were a result of some un- forest part of its lease and 11,000 ha per year derstanding of the forest ecology but mostly in the oil part of its lease. Alberta-Pacific has based on experience in regeneration, growth, implemented an ecosystem management ap- and yield, and silvicultural methods that in- proach to maintaining biodiversity on its lease. creased forest productivity. Forest research us- The forest management system is similar to the ing permanent plots and other experimental usual timber supply model used in sustainable designs started at this time. Sustainable forest forestry up to this time but with more (but not management in these early years was usually complete) spatially explicit control and the in- modeled on the regulated forest idea. In the clusion of disturbances by fire, insects, and hu- western Canadian boreal, a regulated forest mans (e.g., seismic lines, well sites, roads, and management area is one in which all parts of the timber harvesting). productive forest (i.e., forests with growth rates Alberta-Pacific, like many other forestry that produce merchantable trees within the ro- companies in the 1990s, has used the “natu- tation age) have a specified period in which ral disturbance hypothesis” to help it along the they will be harvested, either all at once (clear- path to ecosystem management (Burton et al. cut) or in two passes (the faster-growing tree 2003). The natural disturbance model is sim- species, usually pines, are cut in the first pass ply the idea that forest management should be and some years later the slower tree species, made to mimic natural disturbances to be more usually spruce and fir, that have been released sustainable. This idea is still controversial; how- and grown to mature size are cut in the second ever, some parts of it may certainly be useful pass). The strategy varies considerably,depend- once they are more completely tested and/or ing on the exact combination of saw timber grounded in more experience. Central to the and pulp being managed. Overall in the reg- natural disturbance hypothesis is knowledge of 128 Annals of the New York Academy of Sciences what the age structure of the forest, its compo- Most have involved maintaining a tail of some sition, and spatial pattern of different ages and sort on the age distribution, even if it is not compositions should be. Once these are known, as old as some natural ones, and creating old- forestry operations could mimic them or, if the growth characteristics (e.g., Franklin et al. 1981) natural disturbance is not completely stopped in a younger forest (Bergeron et al. 1999). How (Larsen 1997; Johnson et al. 1998; Campbell & to maintain the forest composition and biodi- Campbell 2000), some combination of human versity within some range of variation that re- and natural disturbances could be incorporated flects the distribution of physical site conditions into the forest plan. These three characteris- of the landscape, the climate conditions, and tics of desired age structure, species composi- the disturbance regimes is still largely an open tion, and spatial patterning are by no means yet question. clearly articulated. Originally it was suggested that some presettlement state be the benchmark, but clearly the past and future dynamics Oil Sands Stripping—Extraction of climate, if nothing else, make this an unwise and Engineering New Ecosystems approach. To determine the age-class distribution, the The strategy of engineering new ecosystems processes that create it must be understood. is not to re-create the landscape as it existed This has usually been approached by under- before but to construct a landscape in which standing how the disturbances, usually fire, the physical environment of geomorphic, hy- caused the age distribution. If the fire frequency drological, and biogeochemical processes will had been constant this would have been easy, provide habitat for plants, fungi, and animals but we have already learned (e.g., Weir et al. to develop and be sustainable. The first decades 2000) that this has not been the case. Conse- have been spent in suggesting important pro- quently, over time there have been pulses of cesses to be considered, design ideas, and gen- age-classes recruited, reflecting usually periods eral objectives of what could be done based on when large areas burned. These pulses are a the literature, previous practice elsewhere, and result of the fact that 95% or more of the area preliminary results from small-scale, short-term burned results from a few large fires (Johnson studies. Only a small number of these studies et al. 1998). These pulses of large fires are associ- are currently in the refereed literature. ated with periods of large-scale climate patterns In recent years this work has been primar- conducive to wildfires (Macias Fauria & John- ily but not exclusively through two organi- son 2006; Skinner et al. 2006). Mimicking these zations. The Canadian Oil Sands Network pulses of age-classes is difficult under the pre- for Research and Development (CONRAD) vailing goals of sustainability for fiber and the is a collection of industry, consultants, aca- business model. demics, and regulatory agencies whose role The elimination of old-growth has been one is to coordinate reclamation research and to of the problems of regulated forests. The old- develop a model that will allow prediction of age tail of the forest age distribution would be ecosystem development and sustainability. In eliminated in a fully regulated forest. That is addition, the Cumulative Environmental Man- to say, all of the productive forest above the agement Association (CEMA) is a nongovern- rotation age of approximately 80 to 100 years mental association of more than 40 industry, would be removed (but note that in any lease First Nation, government, community, health, only part of the land-base is productive, e.g., and environmental groups who are interested about 48% in Alberta-Pacific’s lease). Several in protecting the environment of the Oil Sands ideas have been put forward on how to main- region. CEMA has established a number of tain this old-age tail of the age distribution. working groups and subgroups to look into Johnson & Miyanishi: Creating New Landscapes at Alberta Oil Sands 129 various aspects of reclamation: NOx SO2 Man- on any scaling ideas from geomorphology, hy- agement Working Group, Sustainable Ecosys- drology, or ecosystem science. tems Working Group, Reclamation Working In the Oil Sands, after the stripping of soil, Group, Trace Metals and Air Contaminant peat, and surficial deposits from the underly- Working Group, and Traditional Environmen- ing oil sands and the extraction of bitumen, the tal Knowledge. landscape is left as very large pits up to 100 The overall strategy of landscape ecosystem m deep, dumps of mine tailings, and piles of re-creation is best understood by examining overburden comprised of recent fluvial deposits the Landscape Design Working Subgroup pro- and a variety of glacial deposits and bedrock posal (CEMA-RWG Landscape Design Sub- formations. The mine tailings produced dur- group 2005). This document gives a flowchart ing bitumen extraction are a mixture of wa- of landscape design (Fig. 4) as well as a deter, clay, sand, and residual bitumen. The sand sign checklist and goals. This strategy is to be is used to construct dikes to form impound- used for landscapes yet to be developed. It does ments where the tailings are stored. The ma- not give explicit information on how the land- ture fine tailings (MFT) left after sand removal scape is to be constructed but simply what goals are a stable clay/water suspension that would are to be achieved in the end. As an example take centuries to settle. To speed the settling of use of the checklist, a mine planner pro- process, gypsum is added to bind the clay and poses new landforms consistent with the over- produce consolidated/composite tailings (CT) all closure plan for the lease, and the check- (Suncor Energy 2005). Thus, the end substrates list is then used by the landscape design team for reclamation are overburden, tailings sand, to attempt to satisfy each goal and to opti- and consolidated/CT; soil and peat from the mize the landscape goals in an iterative pro- original stripped sites as well as from neigh- cess. The development of this engineered land- boring undisturbed areas are mixed and used scape would take several decades of design, to cover these substrates prior to revegetation construction, regrading, cover soiling, reveg- (Oil Sands Vegetation Reclamation Committee etation, reclamation certification, and custo- 1998). dial transfers. The checklist gives the design The first problem is how the landscape issues to be addressed in the planning (e.g., template should be constructed in order to technology selection, footprint—size/location, restore/create hydrological, geomorphic, and mass balances, design for closure), desired char- ecosystem processes that could be considered acteristics/goals for various ecosystem aspects equivalent to what was there previously. Reveg- (e.g., soils, vegetation, wildlife, slope stability, etation is required for erosion control and slope trafficability/bearing capacity, natural appear- stability as well as provision of wildlife habitat. ance, seepage and groundwater, surface wa- Since successful revegetation requires sufficient ter hydrology), and the processes (natural haz- water for plant establishment, creation of func- ards and disturbing forces; erosion, transport, tioning hydrological systems is fundamental to and sedimentation; settlement of fills). The Oil any ecosystem reconstruction (Elshorbagy et al. Sands region is at 100-km scale and includes the 2005). The results of the mining process deter- disturbed landscape and the cumulative effects. mine to some extent this strategy.The large pits Landscapes are what one can see from a vantage contain highly saline water which initially will point, usually 10 to 100 km. Landscapes will serve as a treatment system. This dictates the typically contain 10 to 20 landforms. The land- contributing areas of the surrounding hillslopes form has a scale of 1 to 10 km and includes all that will determine the quantity and quality of parts of the ecosystem, both above and below the water. Also, the routing of the water from ground. These definitions given in the Land- these pits through the downstream landscapes scape Design Checklist seem not to be based must be considered. 130 Annals of the New York Academy of Sciences

Figure 4. Flow chart of how the Cumulative Environmental Management Association’s Landscape Design Checklist may be used. EUB, Alberta Energy and Utilities Board; AENV, Alberta Environment; SRD, Alberta Sustainable Resource Development; EIA, Environmental impact assessment; EPEA, Environmental Protection and Enhancement Act. Source: CEMA-RWG Landscape Design Subgroup (2005).

End Pit Lakes wetlands. The end pit lake should have vegetation around the perimeter and the epilimnion An end pit lake is the final mine pit in a mined should maintain levels of oxygen sufficient for subwatershed. The overburden and tailings will natural biodegradation of organic chemicals re- be placed in the pit. The lake will still be pro- leased from the mining. cessing consolidated tailings flux water, precip- Because the end pit lake is deep (ca. 20 m) itation, and surface runoff (Fig. 5). Chemical and large, it will be density stratified vertically flux will come from the reclaimed surrounding into epilimnion and hypolimnion as a result watershed. Some water will come from created of salinity and thermal differences. Mixing of Johnson & Miyanishi: Creating New Landscapes at Alberta Oil Sands 131

Figure 5. Conceptual model of an end pit lake: CT, consolidated tailings; MFT, mature fine tailings; TT, thickened tailings; NST, non-segregated tailings. Source: Golder Associates Ltd. (2005). these two layers occurs in spring and fall when the models cannot be calibrated and validated; wind and temperature gradients supply enough however, the models (RMA10, DYRESM) are energy. This mixing brings oxygen-rich epil- widely used mass-balance systems for simula- imnion waters to the normally oxygen-poor hy- tions in other stratified lakes (e.g., Balistrieri polimnion and increases biodegradation. How- et al. 2006; Bruce et al. 2006; Beutel et al. 2007; ever, the hypolimnion has poor water quality Castendyk & Webster-Brown 2007). because of its anaerobic conditions which will show up in the surface lake waters and in the Wetlands lake runoff. The concentration and biodegrad- ing of chemicals from the mines in the lake In 1999, several suggestions were made depend on the frequency of turnover and strat- in the “Landscape Design Considerations for ification; if the seepage enters the hypolimnion, Wetland Creation” document (Leskiw 1999). it will degrade slowly. If the seepage enters the Most of these were, as the title of the docu- epilimnion, it will degrade more rapidly. The ment states, considerations and not design re- passage of water through wetlands and other quirements. The Landscape Design Consider- pit lakes could improve the water quality. ations document presented a series of topics Starting in the 1970s there has been mod- that needed to be considered. It suggested that eling of the mass balance of the volume and some understanding was required of the ratio of concentration of chemicals in the input and upland to wetland area and that the drainage output of end pit lakes to determine the effects network should consist of a large number of of water recycling, anaerobic and aerobic con- small streams compared to a few big streams. ditions, lake turnover, chemical degradation, The document points out that most restoration oxygen depletion, groundwater seepage, and projects give little consideration to the need to tailing consolidation. Since there are no end create watersheds with functioning water bud- pit lakes at present in the Alberta Oil Sands, gets for the terrain, climate, and vegetation. 132 Annals of the New York Academy of Sciences

Further, the riparian and wetland areas must that this natural analog is not intended to re- provide corridors for wildlife. The diversity of construct exactly the previous landscapes or watersheds must also provide wetlands with watersheds but to give some idea of the kind variable water-fluctuation patterns that reflect of watersheds that can be constructed, what what might occur in this glaciated region nat- may develop naturally from certain constructed urally. The hillslopes surrounding the streams landforms, and what their spatial arrangements and wetlands must have the appropriate steep- could be in order to facilitate hydrological inter- ness, aspect, substrate, and erosion potential to actions in the landscape. Three preliminary ba- create the desired hydrological and geomorphic sic conceptual models of hydrological behavior processes which, in turn, will create the desired were proposed by Devito and Mendoza (2006) ecosystem processes and services. All of these as a basis for the design of reconstructed land- suggestions seem to take little guidance from scapes at Oil Sands mines based on substrate what is known about how drainage networks texture. develop in both glaciated and nonglaciated landscapes (e.g., Knighton 1984). Much more inclusive suggestions based on 1. Fine-grained deposits—overburden and research in the area were presented in the doc- MFT analog (Fig. 6A). Depressions in this ument “Maintenance and Dynamics of Natural landscape will generally be saturated and Wetlands and Western Boreal Forests: Synthesis form wetlands. However, these depres- of Current Understanding from the Utikuma sions will tend to be isolated, therefore Research Study Area” (Devito & Mendoza having limited conductivity and only lo- 2006). Most of the recent research on land- cal flow patterns. The wetlands are often scape hydrology in the Boreal Plain with partic- perched on topographic highs and may ular reference to wetlands has been done at the locally provide areas of recharge downs- Utikuma Research Study Area (URSA) by De- lope. Wetlands on flat terrain will require vito and colleagues. The URSA study was insti- peat to retain water on the landscape. tuted to understand the impact of forestry and 2. Coarse-grained deposits—sand tailings the oil industry on Boreal Plain catchments. analog (Fig. 6B). On upland areas, wet- This research has indicated that the groundwa- lands will only persist where fine-grained ter and surface water interaction is more com- confining layers are present and can ef- plicated than traditional topographic-based hy- fectively seal the depression. These de- drological models have suggested (Devito et al. pressions are not attached to the regional 2005a). With the subhumid climate and rolling groundwater and form isolated systems. (low-relief) glaciated topography of the Bo- Low areas will tend to be well connected real Plain, vertical water flow is a more im- to regional flow patterns; these will have portant process than lateral flow on hillslopes relatively constant water levels because of (Rodriguez-Iturbe 2000; Winter 2001b); unsat- their connection to a consistent ground- urated zone capacity and evapotranspiration water supply. These catchment areas can are key factors driving this dominance of the be quite large, depending on the size of vertical flow. the coarse-grained deposit. The guidelines that were developed from the 3. Coarse-grained veneer on fine-grained URSA research offer the Oil Sands mining a deposits—consolidated/CT analog (Fig. strategy for landform construction and wetland 6C). Groundwater flow will be near the maintenance (Devito & Mendoza 2006). The surface and responsive to climate in this approach is to use natural analogs or, in other system. Depressions and wetlands will be words, to try to follow the previous hydrolog- better connected often by shallow ground- ical patterns and processes in the area. Note water movement. Johnson & Miyanishi: Creating New Landscapes at Alberta Oil Sands 133

Figure 6. Conceptual model of dominant storage and water flow within the subhumid climate of the western Boreal Plain for: A) fine-grained deposits; B) coarse-grained deposits; and C) coarse-grained veneer on fine-grained deposits. P, precipitation; ET, evapotranspiration; E, evaporation; OLF, overland flow; SSSF, subsurface storage flow; GrW groundwater. Source: Devito & Mendoza (2006). 134 Annals of the New York Academy of Sciences

Figure 7. Plan of the three constructed slopes (D1, D2, and D3) on a reclaimed overburden pile at the Mildred Lake Mine in northern Alberta used to test sustainability of different reclamation strategies. Source: Elshorbagy et al. (2005).

Upland Hillslopes Hydrology at a 60◦ V-notch in the gully at the base of the slopes. A sample system also collected wa- In 1999 Syncrude Canada Ltd. started a ter from the peat and till only (i.e., above the project to re-create watersheds at their Mildred overburden of shale) at the toe of each slope. Lake Mine (Elshorbagy et al. 2005). The project The data from 2000 to 2004 were used in a consists of studies of what are thought to be the lumped parameter hydrological process model principal integrated hydrological mechanisms (Fig. 8) that involved feedback loops of water in responsible for hillslope moisture and thus veg- and between surface, peat, till, and shale layers etation recovery. Three north-facing hillslopes (Elshorbagy et al. 2005). Major steps of the mod- were constructed atop level terrain with areas eling involved understanding the system and its of 1 ha (200 m × 50 m) and slopes of 5:1 (Fig. 7). boundaries, identifying the key variables, repre- Slope D1 has a 30 cm till layer over the over- senting the physical processes through govern- burden of shale and a 20 cm peat layer, D2 ing equations, mapping the model structure, has 20 cm of till and 15 cm of peat, and D3 has and simulating the system to understand its be- 80 cm of till and 20 cm of peat (this treatment is havior. In the model, surface water depends approximately the Alberta Environment 1995 on the amount available as precipitation, sur- requirement). The slopes were seeded with bar- face overland flow, and infiltration into the peat ley and planted with white spruce and aspen layer. Peat (till) moisture depends on inter-flow, seedlings in June 1999. Soil moisture, soil tem- evapotranspiration, and infiltration into the till perature, and soil metric potential were mea- (shale) layer. The model assumes that the shale sured at midslope and at several depths on layerdoesnotcontributewatertothechan- each hillslope. Also at midslope, relative humid- nel at the base of the slope; that is, the water ity, air temperature, wind speed and direction, budget is determined by flow from the peat dew point, temperature, soil surface tempera- and till hillslope only. Interflow is determined ture, soil temperature gradient, net radiation, by the hydrologic conductivity, volumetric wa- and ground heat flux were measured. Over- ter content, hydrologic gradient, and soil type. land flow from each hillslope was determined Infiltration is determined by soil temperature, Johnson & Miyanishi: Creating New Landscapes at Alberta Oil Sands 135

Figure 8. Illustration of the hydrologic process model involving feedback loops of water in and between surface, peat, till, and shale layers. The + sign near the arrowheads indicate that variables at either end change in the same direction, while the − sign indicates the converse. + or − signs within loops indicate whether the loops are positive or negative. Source: Elshorbagy et al. (2005). hydrologic conductivity differences between in the mean relative error (MRE) of 3 to 8% and in each layer, and volumetric water con- and 5 to 9%, respectively.D2, the hillslope with tent. Evapotranspiration is determined by soil the thinnest layers, performed the poorest, with moisture and atmospheric parameters. The MRE of 11 to 21%. The evolution of the pa- model simulates daily hydrologic processes. rameters may be due to the drought conditions The model was calibrated using 2001 data and during the study period, changes in hillslope validated using 2003 data. Since some of the parameters, and, of course, the structure of the system parameters, such as saturated hydraulic model used. D3 had the largest changes in pa- conductivity, are not time invariant, each year rameters but also the highest growth of vegeta- the model parameters were tuned to that year’s tive biomass. The changes in hillslope parame- soil moisture in each layer and the outflow.This ters may be related to changes in till infiltration was required because the hillslopes are chang- from the peat perhaps due to decomposition ing over the short term and these changes in and compaction of the peat and changes in parameters were not known a priori. shale infiltration due to changes in hydraulic The results of this modeling exercise were conductivity caused by freeze-thaw cycles. D3 useful in understanding the changes occurring had the lowest moisture stress and was followed in the hillslopes immediately after their cre- by D1 and D2 with the highest moisture stress ation and the hillslopes’ responses to the dif- during the study period. ferent treatments. The model best fit the data The model can be used to test the ability of in hillslopes D3 and D1, the hillslopes with the the system to hold water and minimize deep deepest peat and till layers. This is reflected percolation to the shale layer under varying 136 Annals of the New York Academy of Sciences moisture conditions, estimate water tempo- ommendation has been to plant species that rally available for primary production, and de- are believed to represent different successional termine conditions leading to system failure stages to ensure that, as succession proceeds, (Elshorbagy et al. 2005). This model cannot, the later successional species would be available and was not intended to, describe long-term for colonization (Oil Sands Vegetation Recla- dynamics of these hillslopes; it simply gives a mation Committee 1998). The underlying gen- short-term understanding of the processes con- eral model of succession presented to the Soil trolling soil moisture under these three treat- and Vegetation Subgroup of CEMA is given ments. Notice that these are not watersheds in in Figure 10. The successional pathways can- that they do not consist of several hillslopes, not be derived from the gradient analysis study streams, and wetlands. but are generally inferred from chronosequence studies. However, as explained previously (see Boreal Plain Ecosystem section), the moisture– Vegetation Re-creation nutrient gradients and the upland plant com- The strategy for reclaiming the vegetation munities associated with positions along these was proposed in the Land Capability Classifi- gradients in the Boreal Plain have been shown cation System for Forest Ecosystems (Alberta to be linked to substrate and hillslope position Environment 1998) and by the Oil Sands Veg- rather than to time since the last disturbance etation Reclamation Committee (1998). These (Bridge & Johnson 2000). evolving strategies are to determine the prin- Some reclamation projects on the tailings cipal environmental gradients determining the sand dikes and overburden dumps have taken natural vegetation composition. The analysis of place since 1971 at Suncor and 1976 at Syn- the vegetation gradients is to provide species in- crude (Anderson et al. 1998). Initially, the fo- dicators and environmental conditions toward cus was on erosion control and the areas were which site trajectories should develop. Finally, seeded with grasses and legumes. While suc- this understanding of vegetation patterns and cessful in achieving this goal, the subsequent tolerances could help in the choice of overstory shift in objectives from erosion control to de- and understory plant species to be planted. velopment of self-sustaining ecosystems equiv- The vegetation and environmental gradients alent to predisturbance conditions required a were determined from previous boreal mixed- change in reclamation methods since the suc- wood studies (Beckingham & Archibald 1996). cessful establishment and growth of grasses in- Figure 9 gives the summary of both Becking- hibited tree establishment. ham and Archibald’s gradient analysis and the In 2000, a system of long-term monitoring conceptual model of the conditions limiting in plots was established in both natural and re- different parts of the moisture and nutrient gra- claimed sites. These monitored sites are to pro- dients (Geographic Dynamics Corp. 2002). In- vide a numerical index that can be used to dicator tree and shrub species were found using evaluate vegetation and soil/landscape proper- Dufrene and Legendre (1997). From the or- ties. The soil/landscape properties used are soil dination diagrams and species tolerances, po- moisture and nutrients (pH, soil structure and tential prescriptions have been developed for consistency, electrical conductivity, and sodium the plants best suited for different reclaimed absorption ratio) in the soil horizons. These sites. Furthermore, since the understanding ap- sites are divided into capacity classes with re- pears to be that the vegetation will develop spect to forest productivity even though this is along certain successional pathways (e.g., to not established at this time. This approach is the climax white spruce–balsam fir forest for thus seen at present as a site evaluation and uplands) (Smith & Ottenbreit 1998), the rec- reclamation tool only. Johnson & Miyanishi: Creating New Landscapes at Alberta Oil Sands 137

Figure 9. Distribution of conceptual site types and associated ecosites on the moisture– nutrient regime grid for the Oil Sands area on the Boreal Plain. Source: Geographic Dynamics Corp. (2006).

Wildlife Habitat are those of certain indicator organisms, such as caribou, moose, fisher, lynx, muskrat, and old- Restoration of wildlife in the mined areas is growth birds (Axys Environmental Consulting primarily a matter of creating suitable habitat. Ltd. 2003). Again, as we’ve seen above, sugges- These are usually defined as the pattern of up- tions are made based on what is known about land and lowland vegetation, connectivity of the natural history of the indicator species and this pattern, the role of disturbances (particu- what kinds of habitat and their patterns could larly fire) on the age of the forest, and finally be required. However, there are limited stud- the understory structure and shrub composi- ies at the scale at which Oil Sands mining will tion. The habitat patterns of particular interest occur. 138 Annals of the New York Academy of Sciences

Figure 10. Understanding of succession to climax as given in a report to the Oil Sands Soil and Vegetation Working Group. Source: Geographic Dynamics Corp. (2002).

Discussion The proposed goals of the Alberta government are to create functioning ecosystems that The development of the Oil Sands has in- are the equivalent of or better than the pre- creased rapidly in the last decade as the price disturbance ecosystems. However, similar to and demand for oil have increased. The speed other policy statements like forest health and of development has meant that studies can- ecosystem integrity, these are philosophical or not be carried out to see if the proposed re- policy goals, not scientific or engineering con- creation methods will, in fact, work over the cepts or theories. Thus, there are no explicit long run. Unfortunately,at present there are no re-creation designs or proven experiences upon well-accepted scaling laws in ecosystem restora- which to draw. Most of the environmental stud- tion and re-creation that would allow small- ies and field treatments to date have been con- scale model systems to be scaled up in either cerned with pollution from the bitumen pro- time or space. Past experience in restoration cessing plants, tailing ponds, and piles (Addison ecology has shown that the process generally & Puckett 1980; Koning & Hrudey 1992; Vitt takes longer than planned and requires con- et al. 2003). In this final section we will address siderably more attention and monitoring, of- some of the larger re-creation issues. This is by ten well past the original design and regula- no means a definitive discussion. tion period (Zedler & Callaway 1999). Given The exact composition and shape the re- the rate of change in developing boreal ecosys- created landscapes are to take are still very in- tems, the Oil Sands re-creation will probably completely known. This landscape re-creation take 100 or more years for the main biotic must be coordinated within a lease but also be- components of the system to mature and more tween leases. At present each lease is largely than 500 years for weathering and geomor- independent and the proposed re-created phic processes. Even simple questions, such as ecosystems are to be organized largely around when forestry can expect to be able to har- individual mines (leases). This strategy seems vest again, are unknown. Forest harvests on re- to be primarily driven by the sizes and engi- created ecosystems will probably be longer than neering considerations of the mining processes the present rotation of 80 to 100 years. and the manner in which the leases are issued. Johnson & Miyanishi: Creating New Landscapes at Alberta Oil Sands 139

Natural surficial landscapes evolved over time Nitrogen, on the other hand, is available only as a result of geomorphic, tectonic, and glacial by biological fixation or atmospheric deposi- processes. In the Boreal Plain the basement is a tion. The latter from human sources has be- sedimentary basin with a thick mantle of glacial come increasingly important in the last sev- material. The drainage network is incomplete, eral decades (Vitousek et al. 1997). Weathering with large areas (ca. 30%) covered by peatlands. of the substrate limits primary production in This surficial landscape has been weathered for ecosystems by limiting the nitrogen supply in about 12,000 years and the terrain is still not young soils, due in part to a lack of nitrogen- in equilibrium with erosional and weathering fixing organisms and the low soil carbon. Phos- forces. phorus will be higher early in the weathering Within leases, a return to the postglacial sequences as it is released from the unweath- landscape is not envisioned. However, there are ered primary minerals. However, with time the not yet sufficient studies or models of the geo- supply of unweathered primary minerals will morphic and hydrological processes to evaluate decrease and phosphorus will further be re- and monitor which of these new hillslope and tained in inorganic and recalcitrant organic stream-course designs will allow ecosystems to forms that decrease its biological availability develop that are acceptable to government re- (e.g., Wood et al. 1984). At some intermediate quirements. As a result of the work of Devito stage in the weathering sequence the phospho- and his colleagues (Devito 2005a, 2005b; rus/nitrogen ratio equilibrates, resulting in low Devito and Mendoza 2006), we have a good soil fertility as a result of the low phosphorus start toward an understanding of how land- availability. At the hillslope scale, the weather- scape hydrology operates in the Boreal Plain. ing and leaching processes change downslope, In particular, an understanding of the effect with greater weathering and leaching occur- of climate on the precipitation and evapotran- ring in the more acidic soils near ridgelines spiration balance and the importance of un- and less weathering and leaching occurring in saturated zones, wetlands, and surface-water the more basic soils at the bottom of hillslopes storage to the hydroperiod is central in any (Bouchard 1983; Bouchard & Jolicoeur 2000). landscape re-creation. Along with the moisture-contributing area re- The weathering process with its release of lationship on hillslopes, this explains the almost chemical ions into the developing soil pro- universal importance of moisture–nutrient gra- file and into lower-order streams is of central dients in vegetation composition (e.g., Bridge & importance to the developing ecosystems and Johnson 2000). The chemical characteristic of has not been adequately examined. Of par- first-order stream and wetland water is thus an ticular relevance to the Oil Sands is the cou- integration of the source of the water from the pling of weathering and ecosystem primary hillslopes. An understanding, even rudimen- productivity (Walker & Syers 1976; Vitousek tary, of the chemical evolution of a landscape, & Farrington 1997; Filippelli & Souch 1999; the flux of geologically produced solutes, and Hotchkiss et al. 2000; Nezat et al. 2004). In the biochemical cycles in the developing ecosys- terrestrial ecosystems, phosphorus and nitro- tem are essential in any landscape and ecosys- gen are two of the more important limiting tem re-creation in the Oil Sands. nutrients that control primary productivity of It appears that peatlands as they exist in ecosystems. Phosphorus and the other sedi- the predisturbance landscape are not to be re- mentary nutrients (e.g., calcium, magnesium, stored. The wetlands that are envisioned will potassium) are products of weathering release not have the deep layers of peat presently on into the soil solution, although there is some ev- the landscape or the patterned fens. It is not idence that mycorrhizae can extract calcium di- clear what effect the removal of the peatlands, rectly from primary minerals (Blum et al. 2002). its stockpiling for decades, and then mixing in 140 Annals of the New York Academy of Sciences with the overburden for surface covering will composition we see today is largely unique and have on the atmospheric gain and release of car- has no analogs in the past. The reason there bon. For a recent review of carbon stocks and are no analogs is that both the physical envi- fluxes in Canada, see Bridgham et al. (2006). ronment and the neighborhood of species at a The peatlands also play a significant role in the particular place and time are rarely the same ecohydrology of the region. for longer than 1000 years. This understand- At present the re-creation of the vegetation ing of how the boreal forest is organized and on the Oil Sands is rather simple. Informa- responds to the dynamics of its environment ex- tion from gradient analysis will be used to plains how it has been able to move from south choose indicator organisms to be planted, and of the glacial boundary in the United States the tolerance curves of these organisms and and to reinvade all of Canada and Alaska in the their position on the moisture–nutrient gra- last 12,000 years. Using this view of boreal for- dients will decide their abundance and where est dynamics and given the expected changes they will be placed on the landscape. This ap- due to climate warming (climate warming is proach is largely a reconnaissance method, and hardly ever considered in any of the Oil Sands the data were collected from undisturbed and discussions), one would expect a very different weathered soil landscapes. Thus, it is, in some composition of vegetation on these re-created sense, an equilibrium view between natural dis- sites. Thus one should be trying not to restore turbances. The Oil Sands vegetation, however, some previous composition of plants and ani- will grow on new surfaces that have not previ- mals but to guarantee that certain physical and ously had either physical weathering and ero- ecological processes (and services) are operating sion or plants of any kind. There seems to be within some expected bounds. It is these pro- a belief that some sort of orderly succession, cesses and services which must be determined as shown in Figure 10, will take place, allowing and the acceptable bounds defined. the ecosystem to eventually establish a “climax” In the Oil Sands studies of ecosystem revegetation. This simple sequence of succession creation there is a major difference between is not the current understanding of boreal forest the approaches of the physical process studies dynamics (Burton et al. 2003). (e.g., hydrology and geomorphology) and the The boreal forest is subject to a spectrum of vegetation and wildlife studies. The physical frequencies and types of natural disturbances process studies have created models of the pro- from the slow death of an individual tree over cesses they are trying to create to see how well a period of years to the rapid death of most they can validate their understanding and thus trees in a crown fire (White 1979). In gen- make predictions of the future. Even these de- eral, the small disturbances are frequent and manding approaches have difficulties in under- large disturbances are infrequent (Turner & standing how to model the transient behavior of Dale 1998). However, even the large infrequent these systems and their changing parameters. disturbances still occur in the life span of the The strength of this approach is that models longest-lived trees. Consequently, most com- identify processes that are functional compo- munities consist of the composition of plants nents of ecosystems which are desired in the whose life histories in some manner allow them government’s goals. Also, the success or fail- to just survive under a particular disturbance ure of this understanding can be examined regime, on a particular substrate, and in partic- against the present physical understanding of ular climate conditions. These conditions are the processes and the short-term re-creation relatively stationary for 1000 years or so and studies can be tested against the models’ rela- thus certain compositions of vegetation seem to tively quantitative predictions. recur frequently (Ritchie 1987). Paleoecology, On the other hand, the vegetation and on the other hand, has taught us that the species wildlife studies are pattern descriptions of Johnson & Miyanishi: Creating New Landscapes at Alberta Oil Sands 141 perhaps inappropriate data and do not try to seem to be limited mechanisms for creating understand the processes that give rise to these high-quality fundamental information on the patterns. These studies do not attempt to un- physical and ecosystem processes that land- derstand the energy flow and trophic structure use decisions will require. The Canadian Oil of the re-created ecosystems nor do they look Sands Network for Research and Development at nutrient cycling, population dynamics, or the (CONRAD) and CEMA have, in their short natural disturbance regimes. In both cases and lives, tried and made some notable contribu- particularly in the wildlife studies, the concept tions. However, they have not been integrated of indicator organisms is used. In many of these or expansive enough in their approaches. With- cases the use of indicators is based on either out a high-quality knowledge base, all the charismatic species, species that require large public consultation and industry/government areas of habitat, or ones that are felt to be in- planning are without a foundation. Develop- dicative of areas of high species richness. The ment of this knowledge base will take time. basis for any of these in science is often not Most geoscience and ecological studies take substantial (Andelman & Fagan 2000). years, if not decades, of continuous, high-quality research to develop a suitable base upon which Conclusions the changes and variation in the natural processes can be determined and a more secure The Oil Sands development process does base upon which to inform professional prac- not yet seem to fully incorporate the difficul- tice. This should come as no surprise when one ties of large-scale re-creation of physical land- considers how long (and ongoing) the develop- scapes and ecosystems. The traditional small- ment of the engineering and industrial methods scale methods of remediation and restoration of the mining and processing of the oil sands has are not designed for large-scale re-creations taken; the Alberta government in partnership of landscapes. Furthermore, the focus of such with an oil company began experimentation on reclamation methods is generally missing sev- oil sands extraction in 1944. eral key processes, such as biogeochemistry, the Second, the current laws, regulations, and loss of or dramatic shallowing of soils, and link- policies are unable to set landscape-scale ob- ages among terrestrial, wetland, and aquatic jectives. The current legal and policy arrange- systems. A knowledge base must be developed ments are designed for incremental decision- of how to re-create physical and ecological making over one lease and then by different processes so that they operate within certain government agencies. This means that deci- bounds that provide the biogeosystems and sions at the landscape scale of multiple leases ecosystem services desired. Without this kind are not easily possible. The cumulative impact of understanding, any restoration/reclamation is largely considered within a single lease so that is doomed to fail. In the almost 30 years of the culpability lies with the company or con- Oil Sands development, it has only been in the sortium that has the lease. Land-use planning last decade that we have begun to see interest and regulatory decisions are made for example in the re-creation of the biogeosystems. Except by Alberta Energy for mineral rights, Alberta for some interest in traditional knowledge (e.g., Sustainable Resource Development for timber, Smith 2006), almost nothing has been done on and Alberta Environment for water and air. ecosystem services. The concept of hydrological landscapes as pro- Because of the scale of Oil Sands develop- posed in Montgomery et al. (1995) and Winter ment at the lease level and over the whole (2001a) is perhaps the appropriate landscape– Oil Sands development, many of the prob- ecosystem scale for regional information and lems discussed in this chapter have to do with planning. Particularly in the case of the Oil the land-use regulation framework. First, there Sands, another serious limitation is the ability of 142 Annals of the New York Academy of Sciences environmental planning and decision-making Anderson, E.R., T.Coolern, S. Tuttle & C. Warner. 1998. to maintain both a monitoring and informa- A history of terrestrial reclamation in the Oil Sands tion flow so as to allow adaptive management Region. In Guidelines for Reclamation to Forest Vegetation in the Athabasca Oil Sands Region over the time scale of many decades that will be . Alberta Environmental Protection. Edmonton, AB. required, not just over the period in which the Axys Environmental Consulting Ltd. 2003. Literature Re- mine is operating and the period immediately view of Reclamation Techniques for Wildlife Habi- after closure. tats in the Boreal Forest. Biodiversity and Wildlife Subgroup of the Reclamation Working Group, Cu- mulative Effects Management Association, Fort Mc- Acknowledgments Murray, AB. http://www.cemaonline.ca/ (date accessed: 1/2/2008). We gratefully acknowledge D.R. Charlton, Balistrieri, L.S., R.N. Tempel, L.L. Stillings & L.A. G.I. Fryer, and an anonymous reviewer for their Shevenell. 2006. Modeling spatial and temporal vari- helpful comments on earlier versions of this ations in temperature and salinity during stratifica- manuscript as well as M. Puddister for prepar- tion and overturn in Dexter Pit Lake, Tuscarora, ing the figures. Nevada, USA. Appl. Geochem. 21: 1184–1203. Beckingham, J.D. & J.H. Archibald. 1996. Field Guide to Ecosites of Northern Alberta. Northern Forestry Centre. Conflicts of Interest Edmonton, AB. Beilman, D.W., D.H. Vitt & L.A. Halsey. 2001. Localized The authors declare no conflicts of interest. permafrost peatlands in western Canada: definitions, distributions and degradation. Arct. Antarct. Alp. Res. 33: 70–77. References Bergeron, Y. & S. Archambault. 1993. Decrease of forest fires in Quebec’s southern boreal zone and its relation to global warming since the “Little Ice Age.” Addison, P.A. & K.J. Puckett. 1980. Deposition of at- Holocene 3: 255–259. mospheric pollutants as measured by lichen element Bergeron, Y.,A. Leduc, B.D. Harvey & S. Gauthier. 1999. content in the Athabasca oil sands area. Can. J. Bot. Forest management guidelines based on natural dis- 58: 2323–2334. turbance dynamics: stand- and forest-level consider- Alberta Environment. 1998. Land Capability Classification for ations. Forest. Chron. 75: 49–54. Forest Ecosystems in the Oil Sands. ENV-61-P, Alberta Beutel, M., M. Hannoun, J. Pasek & K.B. Kavanagh. Environment, Government of Alberta. Edmonton, 2007. Evaluation of hypolimnetic oxygen demand AB. in a large eutrophic raw water reservoir. San Vicente Alberta Environment. 1999. Regional Sustainable Develop- Reservoir, Calif. J. Environ. Eng.-ASCE 133: 130–138. ment Strategy for the Athabasca Oil Sands Area. 1/754, Blum, J.D., A. Klaue, C.A. Nezat, et al. 2002. Mycor- Alberta Environment, Government of Alberta. Ed- rhizal weathering of apatite as an important calcium monton, AB. source in base-poor forest ecosystems. Nature 417: Alberta Forest Conservation Strategy. 1997. Alberta 729–731. Forest Conservation Strategy: A New Perspec- Bothe, R.A. & C. Abraham. 1993. Evaporation and Evap- tive on Sustaining Alberta’s Forests. Canada- otranspiration in Alberta 1986 to 1992 Addendum. Alberta Partnership Agreement in Forestry, http:// Surface Water Assessment Branch, Technical Ser- www.borealcentre.ca/reports/afcs.html# Principles vices & Monitoring Division, Water Resources Ser- (date accessed: 11/16/2007). vices, Alberta Environmental Protection, Edmonton, Alberta Sustainable Resource Development. 2007. AB. Map of Forest Management Agreements (last Bouchard, M. 1983. Stationary influences on the chem- review/updated: February 6, 2006). Sustain- ical alteration of soils derived from till (Sherbrooke, able Resource Development, Government of Al- Quebec, Canada). Catena 10: 363–382. berta, Edmonton, AB. http://www.srd.gov.ab.ca/ Bouchard, M. & S. Jolicoeur. 2000. Chemical weathering forests/managing/fmamap.aspx (date accessed: studies in relation to geomorphological research in 11/16/2007). southeastern Canada. Geomorphology 32: 213–238. Andelman, S.J. & W.F. Fagan. 2000. Umbrellas and flag- Bridge, S.R.J. & E.A. Johnson. 2000. Geomorphic princi- ships: Efficient conservation surrogates or expensive ples of terrain organization and vegetation gradients. mistakes? Proc.Natl.Acad.Sci.USA97: 5954–5959. J. Veg. Sci. 11: 57–70. Johnson & Miyanishi: Creating New Landscapes at Alberta Oil Sands 143

Bridgham, S.D., J.P. Megonigal, J.K. Keller, et al. 2006. In Appendices to the Guideline for Wetland Establishment on The carbon balance of North American wetlands. Reclaimed Oil Sands Leases Revised (2007) Edition.Cu- Wetlands 26: 889–916. mulative Environmental Management Association. Bruce,L.C.,D.Hamilton,J.Imberger,et al. 2006. A Edmonton, AB. http://www.cemaonline.ca/ (date numerical simulation of the role of zooplankton in accessed: 1/2/2008). C, N and P cycling in Lake Kinneret, Israel. Ecol. Devito, K.J., I.F. Creed & C.J.D. Fraser. 2005b. Controls Model. 193: 412–436. on runoff from a partially harvested aspen-forested Burton, P.J., C. Messier, D.W. Smith & W.L. Adamowicz headwater catchment, Boreal Plain, Canada. Hydrol. (Eds). 2003. Towards Sustainable Management of the Boreal Process. 19: 3–25. Forest. NRC Research Press. Ottawa, ON. Devito, K.J., J.M. Waddington & B.A. Branfireun. 1997. Camill, P. 2005. Discontinuous permafrost thaw accel- Flow reversals in peatlands influenced by local erates in boreal peatlands during late-20th cen- groundwater systems. Hydrol. Process. 11: 103–110. tury climate warming. Clim. Change 68: 135– Dufrene, M. & P. Legendre. 1997. Species assemblages 152. and indicator species: The need for a flexible asym- Camill, P. & J.S. Clark. 1998. Climate change disequilib- metrical approach. Ecol. Monogr. 67: 345–366. rium of boreal permafrost peatlands caused by local Elshorbagy, A., A. Jutla, L. Barbour & J. Kells. 2005. Sys- processes. Am. Nat. 151: 207–222. tem dynamics approach to assess the sustainability Campbell, I.D. & C.D. Campbell. 2000. Late Holocene of reclamation of disturbed watersheds. Can. J. Civil vegetation and fire history at the southern boreal for- Eng. 32: 144–158. est margin in Alberta, Canada. Palaeogeogr. Palaeoecol. Environment Canada. 2007. Ecozones of Canada. 164: 263–280. http://www.ec.gc.ca/soer-ree/English/vignettes/ Canadian Centre for Energy Information. 2007. Oil- default.cfm (date accessed: 11/16/2007). sands & Heavy Oil: Oil and Natural Gas Environment. Filippelli, G.M. & C. Souch. 1999. Effects of climate and Canadian Centre for Energy Information. Calga- landscape development on the terrestrial phosphorus ry, AB. http://www.centreforenergy.com/silos/ong/ cycle. Geology 27: 171–174. ongEnvironment/usOilAndGasIndEnv01.asp ?Post Franklin, J.F., K.K. Cromack, W. Denison, et al. 1981. ID= (date accessed: 10/12/2007). Ecological characteristics of old-growth Douglas-fir forests. Castendyk, D.N. & J.G. Webster-Brown. 2007. Sensitivity General technical report PNW-118. U.S. Dept. of Agri- analyses in pit lake prediction, Martha Mine, New culture, Forest Service, Pacific Northwest Forest and Zealand 1: Relationship between turnover and input Range Experiment Station. Portland, OR. water density. Chem. Geol. 244: 42–55. Geographic Dynamics Corp. 2002. Shrub Species Review CEMA-RWG Landscape Design Subgroup. 2005. Land- for Boreal Ecosite Re-establishment in the Oil Sands Re- scape Design Checklist, Revised RSDS Government Reg- gion (DOCMan/Public Documents/RWG). Cumulative ulator Version, May 2005. Cumulative Environ- Environmental Management Association. Fort Mc- mental Management Association. Fort McMur- Murray, AB. http://www.cemaonline.ca/ (date ac- ray, AB. http://www.cemaonline.ca/ (date accessed: cessed: 11/16/2007). 1/2/2008). Geographic Dynamics Corp. 2006. Development of Charron, I. & D.F. Greene. 2002. Post-wildfire seedbeds Site Types: Classification through the Groupings of and tree establishment in the southern mixedwood Ecosites and Interpretations for Reclamation (DOC- boreal forest. Can. J. For. Res. 32: 1607–1615. Man/Public Documents/RWG). Reclamation Work- Davis, M.B. 1976. Pleistocene biogeography of temperate ing Group of the Cumulative Environmen- deciduous forests. Geosci. Man 13: 13–26. tal Management Association. Fort McMurray, Denstedt, S. & J.P. Jamieson. 2007. New regulations AB. http://www.cemaonline.ca/ (date accessed: allow oil sands development on First Nation Re- 11/16/2007). serve. Osler, Hoskin & Harcourt LLP, http:// Golder Associates Ltd. 2003. Appendix I: Air emissions for www.osler.com/resources.aspx?id=12096 (date ac- the evaluation of historic and future acid deposition effects on cessed: 12/7/2007). soils in the Athabasca Oil Sands Region. Cumulative Envi- Devito, K., I. Creed, T. Gan, et al. 2005a. A framework ronmental Management Association. Fort McMur- for broad-scale classification of hydrologic response ray, AB. http://www.cemaonline.ca/ (date accessed: units on the Boreal Plain: Is topography the last thing 11/16/2007). to consider? Hydrol. Process. 19: 1705–1714. Golder Associates Ltd. 2005. Phase II 2005/2006 Devito, K. & C. Mendoza. 2006. Appendix C: Pit Lake Work Plan. Cumulative Environmen- Maintenance and dynamics of natural wetlands in tal Management Association. Fort McMurray, western boreal forests: Synthesis of current under- AB. http://www.cemaonline.ca/ (date accessed: standing from the Utikuma Research Study Area. 11/16/2007). 144 Annals of the New York Academy of Sciences

Halsey, L.A., D.H. Vitt & S.C. Zoltai. 1995. Disequilib- of lightning-caused forest fires in subalpine and bo- rium response of permafrost in boreal continental real forests. Can. J. For. Res. 26: 1859–1874. western Canada to climate change. Clim. Change 30: National Energy Board. 2007. Canada’s Oil Sands— 57–73. Opportunities and Challenges to 2015: An Hotchkiss, S., P.M. Vitousek, O.A. Chadwick & J. Price. Update Questions and Answers. http://www. 2000. Climate cycles, geomorphological change, and neb.gc.ca/clf-nsi/rnrgynfmtn/nrgyrprt/lsnd/ the interpretation of soil and ecosystem develop- pprtntsndchllngs20152006/qapprtntsndchllngs2015 ment. Ecosystems 3: 522–533. 2006-eng.html (date accessed: 11/16/2007). Johansson, T., N. Malmer, P.M. Crill, et al. 2006. Decadal National Wetlands Working Group (NWWG). 1998. Wet- vegetation changes in a northern peatland, green- lands of Canada. Ecological land Classification Series, house gas fluxes and net radiative forcing. Global No. 24, Environment Canada, Ottawa, ON and Change Biol. 12: 2352–2369. Polyscience Johnson, E.A., K. Miyanishi & J.M.H. Weir. 1995. Old- Nezat, C.A., J.D. Blum, A. Klaue, et al. 2004. Influence growth, disturbance, and ecosystem management. of landscape position and vegetation on long-term Can. J. Bot. 73: 918–926. weathering rates at the Hubbard Brook Experimen- Johnson, E.A., K. Miyanishi & J.M.H. Weir. 1998. Wild- tal Forest, New Hampshire, USA. Geochim. Cosmochim. fires in the western Canadian boreal forest: Land- Ac. 68: 3065–3078. scape patterns and ecosystem management. J. Veg. Oil Sands Consultations. 2007. Alberta’s Oil Sands. Alberta Sci. 9: 603–610. Department of Energy, Government of Alberta. Ed- Johnson, E.A., H. Morin, K. Miyanishi, R. Gagnon & D.F. monton, AB. www.oilsandsconsultations.gov.ab.ca Greene. 2003. A process approach to understand- (date accessed: 11/13/2007). ing disturbance and forest dynamics for sustainable Oil Sands Vegetation Reclamation Committee. 1998. forestry. In Towards Sustainable Management of the Boreal Guidelines for Reclamation to Forest Vegetation in the Forest. NRC Research Press. Ottawa, ON. Athabasca Oil Sands Region. Alberta Environmental Johnson, E.A. & D.R. Wowchuk. 1993. Wildfires in the Protection. Edmonton, AB. southern Canadian Rocky Mountains and their re- Pratt, L. & I. Urquhart. 1994. The Last Great Forest. NeWest lationship to mid-tropospheric anomalies. Can. J. For. Press. Edmonton, AB. Res. 23: 1213–1222. Price, J.S. 2003. The role and character of sea- Knighton, D. 1984. Fluvial Forms and Processes. E. Arnold. sonal peat soil deformation on the hydrology of London. undisturbed and cutover peatlands. Water Resour. Res. Koning, C.W. & S.E. Hrudey. 1992. Sensory and chem- 39: DOI:10.1029/2002WR001302. ical characterization of fish tainted by exposure Price, J.S., B.A. Branfireun, J.M. Waddington & K.J. De- to oil sand wastewaters. Water Sci. Technol. 25: vito. 2005. Advances in Canadian wetland hydrol- 27–34. ogy. Hydrol. Process. 19: 201–214. Larsen, C.P.S. 1997. Spatial and temporal variations in Ritchie, J.C. 1987. Postglacial Vegetation of Canada. Cam- boreal forest fire frequency in northern Alberta. J. bridge University Press. Cambridge, UK. Biogeogr. 24: 663–673. Rodriguez-Iturbe, I. 2000. Ecohydrology: a hydrologic Leskiw, L. 1999. Appendix C: Landscape design consid- perspective of climate-soil-vegetation dynamics. Wa- erations for wetland creation. In Guideline for Wetland ter Resour. Res. 36: 3–9. Establishment on Reclaimed Oil Sands Leases. Can-Ag En- RWDI AIR Inc. 2005a. NOx Dispersion and Chemistry As- terprises Ltd. for Oil Sands Wetlands Working Group.Cu- sumptions in the CALPUFF Model. Cumulative Envi- mulative Environmental Management Association. ronmental Management Association. Fort McMur- Fort McMurray, AB. ray, AB. http://www.cemaonline.ca (date accessed: Macias Fauria, M. & E.A. Johnson. 2006. Large-scale cli- 11/16/2007). matic patterns control large lightning fire occurrence RWDI Air Inc. 2005b. VOC Emissions Speciation Profile Devel- in Canada and Alaska forest regions. J. Geophys. Res. opment (DOCMan/Public Documents/NSMWG). Cumu- 111: G04008. lative Environmental Management Association. Fort Miyanishi, K. & E.A. Johnson. 2002. Process and patterns McMurray, AB. http://www.cemaonline.ca (date of duff consumption in the mixedwood boreal forest. accessed: 11/16/2007). Can. J. For. Res. 32: 1285–1295. Schneider, R.R. 2002. Alternative Futures: Alberta’s Boreal Montgomery, D.R., G.E. Grant & K. Sullivan. 1995. Wa- Forest at the Crossroads. The Federation of Alberta Nat- tershed analysis as a framework for implementing uralists and The Alberta Centre for Boreal Research. ecosystem management. Water Resour. Bull. 31: 369– Edmonton, AB. 386. Skinner, W.R., A. Shabbar, M.D. Flannigan & K. Logan. Nash, C.H. & E.A. Johnson. 1996. Synoptic climatology 2006. Large forest fires in Canada and the Johnson & Miyanishi: Creating New Landscapes at Alberta Oil Sands 145

relationship to global sea surface temperatures. Vitt, D.H., L.A. Halsey & S.C. Zoltai. 1994. The bog J. Geophys. Res. Atmos. 111: D14106. landforms of continental western Canada in relation Smerdon, B.D., K.J. Devito & C.A. Mendoza. 2005. Inter- to climate and permafrost patterns. Arct. Alp. Res. 26: action of groundwater and shallow lakes on outwash 1–13. sediments in the sub-humid Boreal Plains of Canada. Vitt, D.H., K. Wieder, L.A. Halsey & M.R. Turetsky. J. Hydrol. 314: 246–262. 2003. Response of Sphagnum fuscum to nitrogen de- Smith, J. 2006. Traditional Environmental Knowledge Research position: a case study of ombrogenous peatlands in Guidelines. Cumulative Environmental Management Alberta, Canada. Bryologist 106: 235–245. Association. Edmonton, AB. Waddington, J.M., T.J. Grifﬁs & W.R. Rouse. 1998. Smith, J. & K. Ottenbreit. 1998. Existing ecosites in the Northern Canadian wetlands: net ecosystem CO2 Oil Sands Region. In Guidelines for Reclamation to For- exchange and climatic change. Clim. Change 40: 267– est Vegetation in the Athabasca Oil Sands Region. Alberta 275. Environmental Protection. Edmonton, AB. Waddington, J.M. & N.T. Roulet. 2000. Carbon balance Suncor Energy. 2005. Suncor Energy Fact Sheet: Over of a boreal patterned peatland. Global Change Biol. 6: 30 Years of Reclaiming the Land. www.suncor.com 87–98. (date accessed: 12/5/2007). Walker, T.W. & J.K. Syers. 1976. Fate of phosphorus dur- Tarnocai, C., I.M. Kettles & B. Lacelle. 2005. Peatlands ing pedogenesis. Geoderma 15: 1–19. of Canada. Agriculture and Agri-Food Canada, Re- Weir, J.M.H., E.A. Johnson & K. Miyanishi. 2000. Fire search Branch. Ottawa, ON. frequency and the spatial age mosaic of the mixed- Tchir, T.L., E.A. Johnson & K. Miyanishi. 2004. A model wood boreal forest in western Canada. Ecol. Appl. of fragmentation in the Canadian boreal forest. Can. 10: 1162–1177. J. For. Res. 34: 2248–2262. White, P.S. 1979. Pattern, process and natural disturbance Turetsky, M.R., K. Wieder, L. Halsey & D. Vitt. 2002. in vegetation. Bot. Rev. 45: 229–299. Current disturbance and the diminishing peat- Williams, J.W., S.T. Jackson & J.E. Kutzbach. 2007. Pro- land carbon sink. Geophys. Res. Lett. 29: Art. No. jected distributions of novel and disappearing cli- 10.1029/2001GL014000. mates by 2100 AD. Proc.Natl.Acad.Sci.USA104: Turner, M.G. & V.H. Dale. 1998. Comparing large, infre- 5738–5742. quent disturbances: What have we learned? Ecosys- Winter, T.C. 2001a. The concept of hydrologic land- tems 1: 493–496. scapes. J. Am. Water Resour. As. 37: 335–349. Vitousek, P.M. & H. Farrington. 1997. Nutrient limita- Winter, T.C. 2001b. Ground water and surface water: tion and soil development: Experimental test of a the linkage tightens, but challenges remain. Hydrol. biogeochemical theory. Biogeochemistry 37: 63–75. Process. 15: 3605–3606. Vitousek, P.M., H.A. Mooney, J. Lubchenco & J.M. Wood, T., F.H. Bormann & G.K. Voight. 1984. Melillo. 1997. Human domination of Earth’s ecosys- Phosphorus cycling in a northern hardwood for- tems. Science 277: 494–499. est: Biological and chemical control. Science 223: Vitt, D.H., L.A. Halsey, M.N. Thormann & T. Martin. 391–393. 1996. Peatland Inventory of Alberta. Phase 1: Overview of Zedler, J.B. & J.C. Callaway. 1999. Tracking wetland Peatland Resources in the Natural Regions and Subregions of restoration: Do mitigation sites follow desired tra- Alberta. Alberta Peat Task Force. Edmonton, AB. jectories? Restor. Ecol. 7: 69–73. Management and Conservation Article Behavior of Moose Relative to a Road Network

CATHERINE LAURIAN, De´partement de biologie, chimie et geógraphie, Centre d’e´tudes nordiques, Universite´ du Que´bec a` Rimouski, 300 Alleé des Ursulines, Rimouski, PQ G5L 3A1, Canada CHRISTIAN DUSSAULT,1 Ministe`re des Ressources naturelles et de la Faune du Que´bec - Direction de la recherche sur la faune, 880 Chemin Ste-Foy, Que´bec, PQ G1S 4X4, Canada JEAN-PIERRE OUELLET, De´partement de biologie, chimie et geógraphie, Centre d’e´tudes nordiques, Universite´ du Que´bec a` Rimouski, 300 Alleé des Ursulines, Rimouski, PQ G5L 3A1, Canada RE´ HAUME COURTOIS, Ministe`re des Ressources naturelles et de la Faune du Que´bec - Direction du de´veloppement de la faune, 880 Chemin Ste-Foy, Que´bec, PQ G1S 4X4, Canada MARIUS POULIN, Ministe`re des Transports du Que´bec, Direction geńe´rale de Que´bec et de l’Est du Que´bec, 5353 Boulevard Pierre-Bertrand, Que´bec, PQ G2K 1M1, Canada LAURIER BRETON, Ministe`re des Ressources naturelles et de la Faune du Que´bec - Direction de la recherche sur la faune, 880 Chemin Ste-Foy, Que´bec, PQ G1S 4X4, Canada

ABSTRACT Roads often negatively affect terrestrial wildlife, via habitat loss or fragmentation, noise, and direct mortality. We studied moose (Alces alces) behavior relative to a road network, in an area with a history of moose–vehicle accidents, to determine when moose were crossing roadways or using areas near roads and to investigate if environmental factors were involved in this behavior. We tracked 47 adult moose with Global Positioning System collars in a study area crossed by highways and forest roads. We hypothesized that moose would avoid crossing roads but would make occasional visits to roadsides to feed on sodium-rich vegetation and avoid biting insects. Further, we expected moose avoidance to be greater for highways than forest roads. We recorded 196,710 movement segments but only observed 328 highway and 1,172 forest-road crossings (16 and 10 times lower than expected by chance). Moose usually avoided road proximity up to 500 m on each side but 20% of collared moose made visits to areas within 50 m of highways, which might have resulted from moose searching for sodium in vegetation and roadside salt pools. In fact, vegetation along highways had higher sodium concentrations and was browsed in similar proportions to vegetation in adjacent forest, despite moose avoidance of these zones. Moose, however, did not use areas near roads more during periods of biting insect abundance. Our results supported the hypothesis of scale-dependent selection by moose; avoidance of highways at a coarse scale may confer long-term benefits, whereas selection of highway corridors at finer scales may be part of a strategy to overcome short-term limiting factors such as sodium deficiency. We found a positive relationship between home-range size and the proportion of road axes they contained, suggesting that moose either compensated for habitat loss or made specific movements along highways to gather sodium. The presence of sodium along highways likely increases moose–vehicle accident risks. Removal of salt pools or use of a de-icing salt other than sodium chloride should render highway surroundings less attractive to moose. (JOURNAL OF WILDLIFE MANAGEMENT 72(7):1550–1557; 2008) DOI: 10.2193/2008-063

KEY WORDS Alces alces, behavior, biting insects harassment, feeding, highway crossing, moose, Quebec, road network, sodium availability.

Roads affect terrestrial and aquatic ecosystems in various Although they can successfully adapt to urban environments ways and cause mortality for many species (Forman and (Garrett and Conway 1999), the reaction of moose and Alexander 1998, Rondinini and Doncaster 2002). Many other ungulates to road disturbance may differ in rural or direct effects of roads are obvious during construction, but wild areas where individuals are not as habituated to more insidious effects are permanent and can extend anthropogenic activities. Moose were found to be more substantially beyond the actual road (road-effect zone; agitated and to adopt vigilance behavior more frequently Forman 2000). when feeding along highways (Singer 1978, Yost and In terrestrial ecosystems, road construction modiﬁes Wright 2001). habitat (e.g., fragmentation and loss, creation of new Roads and their surroundings may offer some beneﬁts to habitats, pollution, introduction of invasive species) and wildlife because roads can create highly desirable resources may increase human–wildlife interactions (e.g., hunting or microhabitats that are otherwise rare. Roads may create pressure, road accidents). For terrestrial wildlife, the most additional ecotone habitat and favor the growth of important road effects include habitat fragmentation (barrier deciduous shrubs preferred by ungulates (Child 1998) or to movement) and noise disturbance (Forman and funnel winds and deter biting-insect harassment of un- Alexander 1998, Spellerberg 1998). Roads have been shown gulates (Kelsall and Simpson 1987). In Nordic regions, the to hamper movement of mammals such as caribou (Rangifer use of road de-icers on highways may result in elevated tarandus; James and Stuart-Smith 2000, Dyer et al. 2002) sodium concentrations in vegetation bordering the roadway and wolf (Canis lupus, Whittington et al. 2004). The barrier and, thus, attract ungulates (Grenier 1974, 1980; Leblond et effect of roads and the noise disturbance created might also al. 2007; Laurian et al. 2008). Understanding animal result in avoidance of adjacent areas (Seiler and Eriksson behavior in relation to road networks is necessary to better 1997, Forman and Alexander 1998, Dyer et al. 2002). assess the impact of road development on wildlife and to implement appropriate mitigation measures. 1 E-mail: [email protected] We studied moose behavior relative to a road network in

1550 The Journal of Wildlife Management 72(7) an area with a history of moose–vehicle accident problems respectively, 9.08 C and 21.78 C in January and 9.58 C (60 to 70 accidents/yr during the past decade; Dussault et al. and 21.78 C in July (Environment Canada 2005). 2006). Specifically, we wanted to determine when moose Moose were the most abundant ungulate in the study area. were crossing roadways or visiting their environs and Moose density was estimated at 0.22 moose/km2 in the investigate if environmental factors were involved in this Laurentides Wildlife Reserve during the last aerial survey in behavior. To better understand the effect of disturbance on winter 1994 (St-Onge et al. 1995) but was found to be moose behavior, we considered both highways (paved higher in the northern part of the reserve where we surface, high vehicle speed and traffic level) and forest roads conducted this study (average of approx. 0.5 moose/km2 (unpaved, medium or low vehicle speed, low traffic level). and up to 0.8 moose/km2), because of favorable habitat. We hypothesized that moose would avoid crossing roads but Moose density has likely increased substantially since 1994 that they would make occasional visits along highways because of the implementation of more conservative hunting during spring and summer to feed on sodium-rich regulations (Lamontagne and Jean 1999). White-tailed deer vegetation and to avoid biting insects. We predicted 1) (Odocoileus virginianus) were also present in the study area moose would avoid crossing roads and using areas near but density was low (no survey available). roadways, 2) avoidance would be more pronounced for highways than forest roads, and greater during day than METHODS night, due to increased disturbance, and 3) avoidance would To locate moose, we conducted an aerial survey over a 2-km be lower during periods of biting insect abundance. Despite strip on each side of the highway prior to capture. Because the general avoidance of roads, we expected moose to eat a we were studying behavior of moose in relation to highways, higher proportion of available vegetation along highways we arbitrarily wanted 65% of captured moose to be within compared to forest roads in spring and summer because of the 2-km strip to obtain a sufficient sample of individuals the higher sodium content of roadside vegetation. Finally, likely to come into contact with highways. We captured because we consider roads and their surroundings to be some moose .2 km from highways to provide a sample of generally avoided by moose, leading to habitat loss (Seiler moose potentially less impacted by highways. We conducted and Eriksson 1997), we also expected moose home-range captures following methods approved by the Animal size to increase proportionally with increasing roadway area. Welfare Committee of Faune Que´bec (certificate no. 03- 00-01). We captured an initial sample of 30 adult moose STUDY AREA (.2.5 yr; 22 F and 8 M) in January–February 2003. In We conducted our study in the northern part of the January–March 2004, we recaptured 17 of these moose to Laurentides Wildlife Reserve, about 100 km north of download data and replace collar batteries. We also installed Que´bec City, Que´bec, Canada. Two highways (175 and Global Positioning System (GPS) collars on 12 new adults 169) and a well-developed forest-road network occurred in (6 M and 6 F) to replace mortalities or defective collars. the study area (1,800 km2; Fig. 1), totaling road densities of Similarly, during January–March 2005, we recaptured 18 0.06 km/km2 for highways and 0.16 km/km2 for forest moose (4 M and 14 F) and captured 13 new adults (6 M and roads. Highways were paved, with generally only one lane in 7 F). We recaptured all individuals during January–April each direction. In 2002, mean daily traffic was estimated to 2006 to recover collars. The recording capacity of the GPS be 1,460 vehicles on highway 169 and 2,800 vehicles on collars we used enabled us to schedule one location every 2 highway 175. The highway speed limit was 90 km/hour. hours. We estimated location error to be ,35 m 95% of the The study area was situated within the Canadian Shield time (Dussault et al. 2001). We conducted 3 telemetry where soils were mostly composed of glacial deposits, flights each year to identify dead moose and defective undifferentiated tills that were thinner on mountaintops and collars. thicker in valley bottoms. Forests in the study area were We analyzed moose behavior relative to road axes by typical of the boreal region: coniferous stands with balsam overlaying moose GPS locations on 1:20,000 digital maps of fir (Abies balsamea) and black spruce (Picea mariana) the study area, including road network, using ArcGIS 9.0. dominated high plateaus, whereas low-lying areas and We only considered roads suitable for motor vehicles in the valleys were dominated by mixed and deciduous stands. analyses because trails, restricted to access by foot, quad, or Common deciduous trees included white birch (Betula snowmobile, were less likely to disturb moose. We defined papyrifera), trembling aspen (Populus tremuloides), yellow the study area as the minimum convex polygon (MCP; birch (B. alleghaniensis), and maples (Acer rubrum and A. Mohr 1947) encompassing the locations of all moose saccharum). Topography was gently rolling with some 100– monitored during the study. We used the MCP method 350-m-deep river valleys. Winters were usually harsh in the to delineate moose annual home ranges. We selected the Laurentides Wildlife Reserve with among the highest snow MCP method because a preliminary examination of the data precipitation in the world (total annual snowfall .550 cm in indicated that visits near roads delimited the boundary of some areas), rendering winter driving hazardous (Jolicoeur moose home ranges and that, in most cases, other home- and Creˆte 1994). Each winter .100 metric tons of de-icing range estimation methods excluded those locations. We salts/km of highway were typically used (Jolicoeur and Creˆte calculated a different home range each year for individuals 1994). Mean maximal and minimal daily temperatures were, we followed .1 year. We conducted preliminary analyses

Laurian et al. Moose Behavior Relative to Roads 1551 Using locations spaced 2 hours apart may have under- estimated the true number of road crossings by moose but because the same bias also existed for our estimates of simulated road crossings, we could directly compare the crossing rate between real and simulated road networks. We created an index of road-crossing rate by dividing the number of crossings by highway (or forest road) length (i.e., no. crossings/km) within home ranges. We first used an analysis of variance (ANOVA) to determine if crossing rate differed between highways and forest roads. Second, we conducted 2 separate ANOVAs, one for highways and one for forest roads, to determine if road-crossing frequency differed between the existing and random road networks. We used year and moose identity as random factors in the preceding analyses. We used only individuals having both forest roads and highways within their home range in this analysis (n ¼ 45). We determined if frequency of highway crossings by moose differed by time of day (i.e., night, including dawn and dusk vs. day), as well as by month, using separate chi-square tests. We defined dawn and dusk as 2- hour periods centered on sunrise and sunset, respectively. We obtained times of sunrise and sunset on a daily basis at a weather station located near the study area, which allowed us to precisely categorize each movement segment relative to the time of day, given that sunrise and sunset varied according to time of year. Figure 1. Map of the study area where we studied the behaviour of moose We assessed moose preference towards roadways and their relative to a road network from 2003 to 2006. The study area was located in the north of the Laurentides Wildlife Reserve, Que´bec, Canada. The map surroundings by creating a series of buffer strips centered on shows the distribution of highways, forest roads, lakes and waterways, and highways and forest roads: 50 m, .50–250 m, .250–500 salt pools. m, .500–1,000 m, .1,000–1,500 m, .1,500–2,000 m, and .2,000 m. We considered the roadway of highways and that indicated behavior of males and females relative to road forest roads to have a width of 20 m and 6 m, respectively, axes did not differ. On that basis, we pooled all individuals which meant that the 0–50-m buffer strip began 10 m on for subsequent analyses. either side of the highway centerline and 3 m from the To determine whether roads were a barrier to moose middle of forest roads. movement, we successively simulated 100 road networks For each moose, we calculated a selection ratio for each within the home range of each moose and determined the buffer strip along highways and forest roads as the crossing frequency of real and simulated roads (Dyer et al. proportion of locations in a strip relative to proportion of 2002). To do so, we identified moose road crossings by that strip in the moose’s home range. We inferred avoidance linearly connecting successive locations. We assumed that a of a given strip when the selection ratio was significantly crossing occurred each time one of these interlocation lines ,1.0. We determined whether selection ratio varied with intersected a road. We generated simulated road networks respect to distance from the roadway and month using by moving the highways and forest roads existing within ANOVAs, with year and individual as random factors. We each moose home range into a new random position within conducted separate analyses for highways and forest roads. the home range, while keeping the same geographical We conducted biting insect (i.e., gadflies [Tabanidae orientation for random roads to take into account topo- spp.], mosquitoes [Culicidae spp.], and black flies [Simu- graphical constraints that might have influenced both road liidae spp.]) surveys to create an index of insect abundance layout and animal movement. We recognized that actual between late May and mid-August in 2004 and 2005. We roads were likely not distributed randomly with regard to conducted these surveys every 2–4 days using 18 permanent topography and expected simulated roads to be less likely to stations consisting of 9 pairs of stations, half of which were follow the bottom of valleys compared to actual roads. localized ,5 m from a highway and the others being in the Moose, however, were already found to travel preferentially nearby forest .100 m from the highway. For all stations, an in the bottom of valleys (Gundersen and Andreassen 1998; observer (always the same) noted the subjective level of Dussault et al. 2006, 2007). This potential bias would insect harassment for each group of biting insects as follows: overestimate moose utilization of actual compared to 0 ¼ no insect, 1 ¼ insects are present but no harassment, 2 ¼ random roads. We therefore considered our method to be low harassment, 3 ¼ medium harassment, 4 ¼ high a conservative approach to detect road avoidance by moose. harassment. We grouped insect data by 2-week periods a

1552 The Journal of Wildlife Management 72(7) posteriori: 1–15 June, 16–30 June, 1–15 July, 16–31 July, range. Moose movement in the study area was highest and 1–15 August. We used ANOVAS with year as a random during summer, and previous studies indicated summer factor to determine if insect abundance differed among 2- home-range area included .90% of annual home-range week periods and with distance from highways. For moose area (Dussault et al. 2005). As such, we did not include in with a highway in their home range, we calculated selection this analysis 4 moose that we did not monitor for a full ratios for each 2-week period to assess preference of moose summer. We conducted one analysis, grouping both high- towards highways and their immediate surroundings (i.e., ways and forest roads, because preliminary analyses the roadway plus a 50-m buffer strip) as well as for forested indicated that moose reactions towards these 2 road types sites located .50 m from highways. We calculated the were similar. selection ratios for each moose as proportion of radio- We performed all statistical analyses using SAS version 8.2 locations on the highway and in the 0–50-m strip (or (SAS Institute, Cary, NC) with an a priori significance level proportion of locations .50 m from a highway) divided by of 0.05. We log-transformed dependent variables when proportion of highway plus the 0–50-m strip (or proportion necessary to normalize residuals of the regression. of forested habitat .50 m from a highway) in home range. We used an ANOVA, with year and individual as random RESULTS factors, to determine if selection towards roadways and their We captured moose on average (6SE) 3.32 6 0.27 km immediate surroundings was higher during periods when (range ¼ 0.3–11 km) from a highway. Because moose home biting insects were most abundant. ranges averaged 53.9 6 4.3 km2, we believed that all moose Leaves of deciduous shrubs and trees were generally were likely to come into contact with highways and forest available from early June to mid-October in the Laurentides roads because both of these were widely distributed over the Wildlife Reserve. We collected vegetation samples along study area. highways to measure their sodium content on 3 occasions in We monitored individual moose for 2–36 months. We summer 2005: 29 June, 1 August, and 7 September. We monitored 47 moose; we followed 29, 11, and 7 moose collected leaves and annual shoots from 0.5 m to 3 m above during 1, 2, and 3 years, respectively (72 moose-yr). ground for 3 moose-preferred shrubs (white birch, trem- Telemetry data acquisition terminated prematurely because bling aspen, and willows [Salix spp.]) at 90 sites, half of of death for 22 moose and because of collar failure for 9 which were along highways (,10 m from roadway) where moose. We obtained 199,118 GPS telemetry locations (x¯ 6 de-icing salts were used and the other half of which were SE ¼ 4,187 6 349 locations/moose). along forest roads .200 m from a highway. We dried, Among the 47 moose monitored, 19 never crossed a crushed, ashed, and dissolved samples in a 20% chloride highway and 11 never crossed a forest road, even though 45 acid solution. We then prepared control Inductively home ranges included such features. Among the 196,710 Coupled Plasma solutions with 100 ppm of sodium. We movement segments recorded, we observed 328 highway analyzed 5 replicates for each site 3 period combination. We crossings (1.85 6 0.39 crossing/km of highway in the home used separate ANOVAs to determine the influence of range or 4.57 6 1.20 crossings/moose/yr) and 1,172 forest- sampling period and sampling location (i.e., close to road crossings (1.85 6 0.23 crossing/km of forest road or highway or in forest) on sodium concentrations by species. 16.28 6 2.29 crossings/moose/yr). Some moose crossed We estimated browsing rate of moose in various habitats road axes more often than others (75th percentile: highways by visually assessing the proportion of browsed twigs in a ¼ 8.0; forest roads ¼ 25.5). Highways and forest roads were 10-m radius (i.e., 0–5%, 6–20%, 21–40%, ..., 95–100%). crossed at a similar rate (F1,156 ¼ 0.29, P ¼ 0.592) and the We sampled 196 plots distributed in sites located 50 m interaction between road type and road status (real vs. from a highway, 50 m from a forest road, and in forested simulated) was not significant (F1,156 ¼ 1.76, P ¼ 0.186). areas .50 m from a road axis. Because browsed proportions Highway and forest-road crossing frequencies were, respec- were low, we had to collapse the 6–100% classes to avoid tively, 16 times and 10 times lower than expected, based on 0.0 values, leaving 2 classes remaining. We used log-linear simulated road networks (30.53 6 21.25 crossings/km for analyses to determine if browsing rate differed according to simulated highways and 18.33 6 14.19 crossings/km for distance from roadways for deciduous and coniferous simulated forest roads, F1,156 ¼ 70.16, P , 0.001). The (balsam fir only) species independently. In the same survey road-crossing frequency by moose differed by month (v2 ¼ plots, we assessed food availability in 10-m2 subplots by 655.9, df ¼ 11, P , 0.001, n ¼ 328 for highways and v2 ¼ counting the number of stems with 1 browsable twig (.5 1,893.9, df ¼ 11, P , 0.001, n ¼ 1,172 for forest roads). cm length and between 0.5 m and 3 m high). We used a Most highway and forest-road crossings occurred during Kruskal–Wallis test to determine if food availability differed May–July (54% of highway and 59% of forest-road among sampling plots close to highways, close to forest crossings; Fig. 2). A second moderate peak appeared in roads, and in the forest. October for highways and in September for forest roads, and We assessed the influence of road network on home-range crossings were less abundant during January–March. Finally, size with a linear regression between home-range area and crossings were more frequent at night for highways (v2 ¼ proportion of roadways (area of highways plus a 10-m buffer 245.1, df ¼ 1, P , 0.001) and forest roads (v2 ¼ 3.28, df ¼ 1, zone and forest roads plus a 3-m buffer zone) within home P ¼ 0.070; Fig. 2).

Laurian et al. Moose Behavior Relative to Roads 1553 Figure 2. Monthly proportion of crossings by day period (day vs. night crossings ) and monthly proportion of movement segments crossing a road axis (i.e., no. of crossings/total no. of locations ) by moose for highways and forest roads between 2003–2006, Laurentides Wildlife Reserve, Que´bec, Canada.

Moose did not use strips around highways (F7,4248 ¼ Figure 3. Box-plot diagram of selection ratios of moose toward roads and 171.17, P , 0.001) and forest roads (F7,4855 ¼ 196.45, P , their surroundings at the home-range scale between 2003–2006, Lauren- 0.001) randomly. Year-round moose avoidance was most tides Wildlife Reserve, Que´bec, Canada. * indicates a significant difference pronounced near the roadway itself and less pronounced for between use and availability; bars with the same letter do not differ the .2,000-m surrounding strip for both highways and significantly. forest roads (Fig. 3), and use was less than expected in strips up to 500 m from highways and up to 1,000 m near forest Sodium concentration in vegetation did not differ among roads. Moose avoided strips .1,000–2,000 m from high- the 3 sampling periods for both highways and forest roads ways (P , 0.010). The month 3 strip category interaction (white birch: F2,84 ¼ 0.27, P ¼ 0.760; trembling aspen: F2,84 was significant for highways (F77,4248 ¼ 1.68, P , 0.001) but ¼ 1.47, P ¼ 0.236; willows: F2,84 ¼ 1.13, P ¼ 0.329), but it not for forest roads (F77,4855 ¼ 0.93, P ¼ 0.654). Strips ,500 was higher along highways than along forest roads for all m from highways were mostly visited by moose in May and species (Table 2; white birch: F ¼ 19.72, P , 0.001; June but rarely during December–April, and the use of other 1,84 trembling aspen: F ¼ 16.78, P , 0.001; willows: F ¼ strips was similar year-round (Fig. 4). Not all individuals 1,84 1,84 23.07, P , 0.001). Differences between highways and forest had the same reaction to roads. Although the 0–50-m strip roads were greater at the end of June than in early August was on average avoided by moose (t ¼6.84, P , 0.001), 345 and September for trembling aspen (F ¼ 6.37, P ¼ 0.003) 4 individuals highly preferred it (selection ratio ¼ 7.8 to 2,84 and willows (F ¼ 3.43, P ¼ 0.037). 393.2), and it was moderately preferred by 4 others 2,84 (selection ratio ¼ 1.1 to 1.8; Fig. 3). The proportion of browsed twigs in sample plots was , % Abundance of biting insects varied among sampling usually 5 for both deciduous and coniferous stems and . , did not differ among highway sides, forest-roads’ sides, and periods for each species (F5,415 3.11, P 0.009), but 2 usually did not differ between highway and forest sampling forest, for both deciduous (v ¼ 1.08, P ¼ 0.578, df ¼ 2, n ¼ 180) and coniferous (v2 ¼ 1.88, P ¼ 0.392, df ¼ 2, n ¼ 184) sites (F1,32 . 1.64, P . 0.079), except for gadflies, which species. Availability of deciduous stems did not differ among were less abundant along highways (F1,32 ¼ 5.13, P ¼ 0.030). Gadflies and black flies were more abundant from mid-June highway, forest road, and forest sites (H ¼ 0.53, P ¼ 0.768, n to late July. Mosquitoes peaked in early July but were also ¼ 98 sites), but coniferous species were less abundant along abundant from mid-June to late July. Overall, biting insects highways than elsewhere (H ¼ 6.17, P ¼ 0.057, n ¼ 98). were most abundant in July. However, the relative Highways and forest roads occupied 0.54% of moose preference of moose for highways and their immediate home ranges (x¯ ¼ 0.22 6 0.02%), but home-range area surroundings did not increase during these periods of high increased with increasing roadway area (F1,66 ¼ 6.80, P ¼ biting-insect abundance (Table 1). 0.011; ln[home-range area] ¼ 0.119 3 ln[proportion of road

1554 The Journal of Wildlife Management 72(7) Figure 4. Monthly distribution of moose locations in a series of strips centered on highways between 2003–2006, Laurentides Wildlife Reserve, Que´bec, Canada. Numbers on top of bars are number of locations. axes in the home range þ 0.01] þ 4.046). Highway and contrast with previous generalizations that environmental forest-road area explained only 9% of the variation in factors may attract moose to road corridors (Thompson and home-range size. Moose home ranges containing a high Stewart 1998). We found little evidence that moose used proportion of roaded area were about 11% larger than those highways and forest roads as a refuge from biting insects as containing little roaded area. suggested by Kelsall and Simpson (1987). We did find moose more often in the 0–50-m strip along highways DISCUSSION during June and July, the 2 months with the highest biting- As expected, moose avoided crossing roads and frequenting insect abundance, but moose did not increase use of their surroundings. Moose may avoid proximity of road axes highways and their surroundings with increased insect for several reasons, including dislike of the roadway and abundance within these 2 months. Moreover, our data did associated forest opening, vehicle traffic (i.e., density or not lend support to the hypothesis that biting-insect noise), or predation (Forman and Alexander 1998, Dyer et abundance is lower in roadway corridors than in surrounding al. 2002). Similar to white-tailed deer, moose may adapt areas of vegetation. The low proportion of moose locations behaviorally to roads by avoiding nearby habitats where on or near highways overall also suggests that use of roads traffic noise inhibits predator detection and may prefer to for insect avoidance is unlikely. cross roads at specific sites and time periods, as we observed Higher sodium concentration in vegetation has been (Forman and Deblinger 2000, Dussault et al. 2007). previously suggested as a mechanism attracting moose close Interestingly, moose highway crossings were 4 times more to highways (Thompson and Stewart 1998). We found likely to occur at night, when traffic level was 33% of that sodium concentration to be higher in vegetation collected observed during the day (J. David, Ministe`re des Transports along roads where de-icing salt was used in winter. du Que´bec, unpublished data). These results were similar to However, the proportion of browsed twigs did not differ findings of Dyer et al. (2001, 2002), who found road among sites along highways, forest roads, or in the forest. avoidance by caribou to increase with traffic level and to The fact that moose browsed a similar proportion of Joyal et al. (1984) who found avoidance of moose toward available food along highways is noteworthy, considering power lines to increase with right-of-way width. avoidance of these areas. These results support the In agreement with Burson et al. (2000) and Yost and hypothesis of scale-dependent selection by moose (Rettie Wright (2001), our data suggest road avoidance by moose in and Messier 2000, Dussault et al. 2005). General avoidance

Laurian et al. Moose Behavior Relative to Roads 1555 Table 1. Results of analyses of variance we used to test if moose-selection Table 2. Mean sodium concentration (ppm 6 SE) in vegetation sampled ratio for the roadway of highways and their immediate surroundings (0–50- along highways and forest roads in summer 2005, Laurentides Wildlife m strip) varied among periods with differing biting insect abundance Reserve, Que´bec, Canada. between 2003–2006, Laurentides Wildlife Reserve, Que´bec, Canada. Along highways Along forest roads Yr Source df FP Vegetation type x¯ SE x¯ SE 2004 Perioda 4 1.01 0.406 Distance from roadwayb 1 87.67 ,0.001 White birch 147.9 12.8 89.7 9.1 Period 3 distance from roadway 4 1.16 0.336 Trembling aspen 170.3 25.3 98.6 7.8 2005 Period 4 0.36 0.838 Willows 281.3 64.3 114.6 8.4 Distance from roadway 1 52.83 ,0.001 Period 3 distance from roadway 4 0.49 0.743

a Period relates to biting-fly abundance periods: 1–15 Jun, 16–30 Jun, 1– by feeding on sodium-rich vegetation or drinking brackish 15 Jul, 16–31 Jul, 1–15 Aug. water in roadside salt pools. b Highway and 0–50-m strip vs. .50-m strips. MANAGEMENT IMPLICATIONS of highways at coarse scales may confer long-term benefits Our results demonstrate that moose globally perceive road to moose (e.g., better predator detection, avoidance of networks, including up to 500 m beyond the roadway, as vehicle-collision mortality), whereas selection of highway low-quality habitat. Moose crossing highways were more corridors at finer scales during some periods may be likely to do it at night, which is problematic because of the beneficial to overcome short-term limiting factors (e.g., reduction in motorists’ visual acuity in the dark (Joyce and sodium deficiency). We cannot reject the hypothesis that Mahoney 2001), making roadway lighting a potentially moose used areas along highways, at least partially, to feed important mitigation measure (Reed 1981). Because our on sodium-rich vegetation. We also found increased high- results suggest that moose frequented the vicinity of way crossing and use of the 0–50-m strip in spring and early highways to consume sodium from salt pools and vegeta- summer; vegetation is known to green-up faster in these tion, especially in spring and summer, presence of sodium open habitats (Rea 2003). We did not find deciduous-stems along highways may be a key risk factor for moose–vehicle density to be higher along roadways, as previously reported accidents (Dussault et al. 2006). Removal (e.g., drainage and (Child 1998, Finder et al. 1999). filling with rock boulders [Leblond et al. 2007]) of salt pools We found some moose to make frequent visits to the 0– should render highway surroundings less attractive to 50-m strip located on either side of highways. Nearly 20% moose. Use of a de-icing salt other than sodium chloride, of collared moose showed this behavior, which could be such as calcium chloride or calcium–magnesium acetate, is interpreted as a quest for sodium. Sodium was readily also likely to keep moose away from highways. Our results available in vegetation close to highways and also in roadside should guide managers in assessing the impact of further salt pools. Salt pools are formed in poorly drained sites road-development projects on moose and on moose–human following accumulation of de-icing salts (mainly composed interactions. of sodium chloride) and were found to have mean sodium ACKNOWLEDGMENTS concentrations of 500–600 ppm (Grenier 1980, Fraser 1979, We thank D. Be´rube´, G. Carl, J. Dufour-Gallant, J. Fortin, Fraser and Thomas 1982, Dussault et al. 2006, Leblond et A. He´bert, M. Lavoie, M. Leblond, and A. Le´vesque for al. 2007). Such sodium concentrations are even higher than their assistance in the field or in the laboratory. We also those found in aquatic vegetation (MacCracken et al. 1993). thank A. Caron for his help with Geographic Information Laurian et al. (2008) demonstrated that moose were making System and statistical analyses and T. Connor for improving directional movements to reach salt pools, which were often our English translation. This study was funded by the located at home-range boundaries, thereby increasing their Ministe`re des Transports du Que´bec, the Ministe`re des home-range area. Ressources naturelles et de la Faune du Que´bec, and the We hypothesize that moose visited highway surroundings Universite´ du Que´bec a` Rimouski.C.Laurianwas primarily to obtain sodium from salt pools and vegetation. supported by scholarships from the Conseil de recherches Sodium is essential to moose because it plays a major role in en sciences naturelles et geńie du Canada, the Fonds many vital functions (Church et al. 1971, Robbins 1993); que´bećois de recherche sur la nature et les technologies, and however, sodium is rare in northern ecosystems such as the the Centre d’e´tudes nordiques. Laurentides Wildlife Reserve, as outlined by Jordan et al. (1973). It has been suggested that, at the onset of spring, moose need more sodium than they can obtain from LITERATURE CITED terrestrial vegetation alone (Weeks and Kirkpatrick 1976, Belovsky, G. E., and P. A. Jordan. 1981. Sodium dynamics and adaptations Belovsky and Jordan 1981, Fraser et al. 1982, Jordan 1987, of a moose population. Journal of Mammalogy 62:613–621. Ohlson and Staaland 2001). In our study, use of sodium- Burson, S. L., J. L. Belant, K. A. Fortier, and W. C. Tomkiewicz. 2000. rich environments by moose was highest in spring and early The effect of vehicle traffic on wildlife in Denali National Park. Arctic 53:146–151. summer, a time when aquatic vegetation was less available. Child, K. N. 1998. Incidental mortality. Pages 275–301 in A. W. Some moose appear to have fulfilled their needs efficiently, Franzmann and C. C. Schwartz, editors. Ecology and management of

1556 The Journal of Wildlife Management 72(7) the North American moose. Smithsonian Institution Press, Washington, Joyal, R., P. Lamothe, and R. Fournier. 1984. L’utilisation des emprises de D.C., USA. lignes de transport d’e´nergie e´lectrique par l’orignal (Alces alces) en hiver. Church, D. C., G. E. Smith, J. P. Fontenot, and A. T. Ralston. 1971. Canadian Journal of Zoology 62:260–266. [In French.] Digestive physiology and nutrition of ruminants. Volume 2. Oregon State Joyce, T. L., and S. P. Mahoney. 2001. Spatial and temporal distributions University Bookstores, Corvallis, USA. of moose–vehicle collisions in Newfoundland. Wildlife Society Bulletin Dussault, C., R. Courtois, J.-P. Ouellet, and J. Huot. 2001. Influence of 29:281–291. satellite geometry and differential correction on GPS location accuracy. Kelsall, J. P., and K. Simpson. 1987. The impacts of highways on ungulates; Wildlife Society Bulletin 29:171–179. a review and selected bibliography. Prepared for Ministry of Environment Dussault, C., J.-P. Ouellet, R. Courtois, J. Huot, L. Breton, and H. and Parks, Kamloops, British Columbia, Canada. Jolicoeur. 2005. Linking moose habitat selection to limiting factors. Lamontagne, G., and D. Jean. 1999. Plan de gestion de l’orignal 1999– Ecography 28:619–628. 2003. Socie´te´ de la faune et des parcs du Que´bec, Que´bec, Canada. [In Dussault, C., J.-P. Ouellet, C. Laurian, R. Courtois, M. Poulin, and L. French.] Breton. 2007. Moose movement rates along highways and crossing Laurian, C., C. Dussault, J.-P. Ouellet, R. Courtois, M. Poulin, and L. probability models. Journal of Wildlife Management 71:2338–2345. Breton. 2008. Behavioral adaptations of moose to roadside salt pools. Dussault, C., M. Poulin, R. Courtois, and J.-P. Ouellet. 2006. Temporal Journal of Wildlife Management 72:1094–1100. and spatial distribution of moose–vehicle accidents in the Laurentides Leblond, M., C. Dussault, J.-P. Ouellet, M. Poulin, R. Courtois, and J. Wildlife Reserve, Quebec, Canada. Wildlife Biology 12:415–425. Fortin. 2007. Electric fencing as a measure to reduce moose–vehicle Dyer, S. J., J. P. O’Neill, S. M. Wasel, and S. Boutin. 2001. Avoidance of collisions. Journal of Wildlife Management 71:1695–1703. industrial development by woodland caribou. Journal of Wildlife MacCracken, J. G., V. Van Ballenberghe, and J. M. Peek. 1993. Use of Management 65:531–542. aquatic plants by moose: sodium hunger or foraging efficiency? Canadian Dyer, S. J., J. P. O’Neill, S. M. Wasel, and S. Boutin. 2002. Quantifying Journal of Zoology 71:2345–51. barrier effects of roads and seismic lines on movements of female Mohr, C. O. 1947. Table of equivalent populations of North American woodland caribou in north-eastern Alberta. Canadian Journal of Zoology small mammals. American Midland Naturalist 37:223–249. 80:839–845. Ohlson, M., and H. Staaland. 2001. Mineral diversity in wild plants: Environment Canada. 2005. National Climate Data and Information benefits and bane for moose. Oikos 94:442–454. , . Archive. http://www.climate.weatheroffice.ec.gc.ca . Accessed 12 Sep Rea, R. V. 2003. Modifying roadside vegetation management practices to 2005. reduce vehicular collisions with moose Alces alces. Wildlife Biology 9:81– Finder, R. A., J. L. Roseberry, and A. Woolf. 1999. Site and landscape 91. conditions at white-tailed deer vehicle collision locations in Illinois. Reed, D. F. 1981. Effectiveness of highway lighting in reducing deer– Landscape and Urban Planning 44:77–85. vehicle accidents. Journal of Wildlife Management 45:721–726. Forman, R. T. T. 2000. Estimate of the area affected ecologically by the Rettie, W. J., and F. Messier. 2000. Hierarchical habitat selection by road system in the United States. Conservation Biology 14:31–35. woodland caribou: its relationship to limiting factors. Ecography 23:466– Forman, R. T. T., and L. E. Alexander. 1998. Roads and their major 478. ecological effects. Annual Review of Ecology and Systematics 29:207– Robbins, C. T. 1993. Wildlife feeding and nutrition. Second edition. 231. Academic Press, San Diego, California, USA. Forman, R. T. T., and R. D. Deblinger. 2000. The ecological road-effect Rondinini, C., and C. P. Doncaster. 2002. Roads as barriers to movement zone of a Massachusetts (USA) suburban highway. Conservation Biology for hedgehogs. Functional Ecology 16:504–509. 14:36–46. Seiler, A., and I.-M. Eriksson. 1997. New approaches for ecological Fraser, D. 1979. Sightings of moose, deer, and bears on roads in northern consideration in Swedish road planning. Pages 253–264 in K. Canters, A. Ontario. Wildlife Society Bulletin 7:181–184. Piepers, and A. Hendriks-Heersma, editors. Proceedings of the Interna- Fraser, D., and E. R. Thomas. 1982. Moose–vehicle accidents in Ontario: relation to highway salt. Wildlife Society Bulletin 10:261–265. tional Conference on Habitat Fragmentation, Infrastructure and the Role Fraser, D., B. K. Thompson, and D. Arthur. 1982. Aquatic feeding by of Ecological Engineering. Ministry of Transport, Public Works and moose: seasonal variation in relation to plant chemical composition and Water Management, Road and Hydraulic Engineering Division, Delft, use of mineral licks. Canadian Journal of Zoology 60:3121–3126. the Netherlands. Garrett, L. C., and G. A. Conway. 1999. Characteristics of moose–vehicle Singer, F. J. 1978. Behavior of mountain goats in relation to U.S. highway collisions in Anchorage, Alaska, 1991–1995. Journal of Safety Research 2, Glacier National Park, Montana. Journal of Wildlife Management 42: 30:219–223. 591–597. Grenier, P. A. 1974. Orignaux tue´s sur la route dans le parc des Spellerberg, I. F. 1998. Ecological effects of roads and traffic: a literature Laurentides, Que´bec, de 1962 a` 1972. Le Naturaliste Canadien 101:737– review. Global Ecology and Biogeography Letters 7:317–33. 754. [In French.] St-Onge, S., R. Courtois, and D. Banville. 1995. Inventaires ae´riens de Grenier, P. A. 1980. Contribution a l’e´tude de moyens pre´ventifs pour l’orignal dans les re´serves fauniques du Que´bec. Ministe`re de l’Environ- re´duire le nombre d’accidents routiers impliquant des orignaux. Ministe`re nement et de la faune, Direction de la faune et des habitats, Que´bec, du Loisir, de la Chasse et de la Peˆche, Direction ge´ne´rale de la faune, Canada. [In French.] Que´bec, Canada. [In French.] Thompson, I. D., and R. W. Stewart. 1998. Management of moose habitat. Gundersen, H., and H. P. Andreassen. 1998. The risk of moose Alces alces Pages 377–401 in A. W. Franzmann and C. C. Schwartz, editors. collision: a predictive logistic model for moose–train accidents. Wildlife Ecology and management of the North American moose. Smithsonian Biology 4:103–110. Institution Press, Washington, D.C., USA. James, A. R. C., and A. K. Stuart-Smith. 2000. Distribution of caribou and Weeks, H. P., and J. R. Kirkpatrick. 1976. Adaptations of white-tailed deer wolves in relation to linear corridors. Journal of Wildlife Management 64: to naturally occurring sodium deficiencies. Journal of Wildlife Manage- 154–59. ment 40:610–625. Jolicœur, H., and M. Creˆte. 1994. Failure to reduce moose–vehicle Whittington, J., C. C. St Clair, and G. Mercer. 2004. Path tortuosity and accidents after a partial drainage of roadside salt pools in Quebec. Alces the permeability of roads and trails to wolf movement. Ecology and 30:81–89. Society 9:4. ,http://www.ecologyandsociety.org/vol9/iss1/art4.. Ac- Jordan, P. A. 1987. Aquatic foraging and the sodium ecology of moose: a cessed 15 May 2006. review. Swedish Wildlife Research Viltrevy Supplement 1:119–137. Yost, A. C., and R. G. Wright. 2001. Moose, caribou, and grizzly bear Jordan, P. A., D. B. Botkin, A. S. Dominski, H. S. Lowendorf, and G. E. distribution in relation to road traffic in Denali National Park, Alaska. Belovsky. 1973. Sodium as a critical nutrient for the moose of Isle Royale. Arctic 54:41–48. Proceedings of the North American Moose Conference Workshop 9:13– 42. Associate Editor: McCorquodale.

Laurian et al. Moose Behavior Relative to Roads 1557 -- -- NorthernRiuer Basins $tudy

NORTHERNRIVER BASINS STUDY PROJECT REPORT NO,63 EXECUTIVESUMMARY OFA WORKSHOPON THE IMPACTSOF LANDCLEARING ON THE HYDROLOGICAND AQUATIC RESOURCESOF BOREALFORESTS IN ALBERTA NOVEMBER18 AND 19, 1994 Preparedfor the NorthernRiver Basins Study underProject 5203-C1

ElizabethAlke E. Alke Consulting

NORTHERNRIVER BASINS STUDY PROJECT REPORT NO.63 EXECUTIVESUMMARY OF A WORKSHOPON THE IMPACTSOF LANDCLEARING ON THEHYDROLOGIC AND AQUATIC RESOURCESOF BOREALFORESTS IN ALBERTA NOVEMBER18 AND 19, 1994

Publishedby the NorthernRiver Basins Study Edmonton,Alberta December,1995 ffi

CANADIANCATALOGUING IN PUBLICATIONDATA ffi Alke,Elizabeth ffi Executivesummary of a workshopon the impacts of landclearing on the hydrologicand aquatic resourcesof borealforests in Alberta' B November18 and 19, 1994

(NorthemRiver Basins Study project report, ISSN1192-3571; no. 63) w lncludesbibliographical references. tsBN0€62-24035-9 Cat.no. R71-49/3-63E €

1. Hydrology,Forest - Alberta- Congresses' 2. Clearingof land- Alberta- Congresses- l. NorthernRiver Basins Study (Canada) E 11.Title. lll. Series. g GB7o8.A441995 551.48 C95-980313-0 T E B E E B B

@1995 by the NorthemRiver Basins Study' Copyright provided B nrrli'EirtJr"served. permissionis grantedto reproduceall or any portionof this p..ublication the reprJouctionincludesi properack--noruledgement of the Studyand a propercredit to the authors' The profit. expressedin rebroouctionmust be pr"""nteo withinits froper contextand mustnot be usedfor The views this publicationare solelythose of the authors. w B w FI fH. I *f. i fr PREFACE: "Canada-Alberta-Northwest ff"i The NorthernRiver Basins Study was initiatedthrough the TerritoriesAgreement ht Respectingthe peace-Athabasia-SlaveRiver Basin Study, Phase ll - TechnicalStudies" which was signed sepiemuei27 , 1gg1.The purposeof the Studyis to understandand characterizethe cumulativeeffects of Fr developmenton the waterand aquatic environment of the StudyArea by coordinatingwith existingprograms &l andundertaking appropriate new technical studies. Thispublication reports the methodand findings of particularwork conducted as partof the NorthernRiver BasinsStudy. As such,'Studythe work was govemed by a specificterms of referenceand is expectedto contribute s] the Rreawithin tne contextof the overallstudy as describedby the StudyFinal informationabout Report. This reporthas beenreviewed by the StudyScience Advisory Committee in regardsto scientific H contentand has beenapproved by the StudyBoard of Directorsfor publicrelease. It is explicitin the objectivesof the Studyto reportthe resultsof technicalwork regularly to the public'This objectiveis servedby distributingprojectreports to an extensivenetwork of libraries,agencies, organizations g an-dinterested individuals and by grantinguniversal permission to reproducethe material' B

tw ffi ffi

€ B B @ ffi ,F t.t.: ri {r NORTHERNRIVER BASINS STUDY {i gt FORM *l PROJECTREPORT RELEASE 8-I Thispublication maY be cited as: t Afke, E. 1995. lVorthernRiver Basins StudyProject ReportNo. 63, ExecutiveSummary of a Workshopon the Impactsof LandCtearing on the Hydrologic and Aquatic Resourcesof BorealForesfs in Alberta,November 18 and 19,1994.Northern River Basins Study, Edmonton, gl Alberta.

Whereasthe above publicationis the resultof a projectconducted under the NorthernRiver Basins Studyand the terms of referencefor that projectare deemedto be fulfilled, fi IT IS THEREFOREREQUESTED BY THE STUDYOFFICE THAT; this publicationbe subjectedto properand responsiblereview and be consideredfor releaseto the ffi tr "to Whereasit is an explicitterm of referenceof the ScienceAdvisory Committee review,for scientific content,material for publicationby the Board", il IT IS HEREADVISED BY THESCIENCE ADVISORY COMMITTEE THAT; thispublication has been reviewed for scientificcontent and that the scientificpractices represented in the ieportare acceptable given the specificpurposes of the projectand subject to the fieldconditions E]t-.r encountered. SUPPLEMENTALCOMMENTARY HAS BEEN ADDED TO THISPUBLICATION: I I VES 1VfruO Fl Ii €"i

ffi Whereasthe Study Board is satisfiedthat this publicationhas been reviewedfor scientificcontent and for immediatehealth implications, IT IS HERE APPROVEDBY THE BOARD OF DIRECTORSTHAT; ffi this pubticationbe reteasedto the public,and that this publicationbe designatedfor: tvf SfnNOAnO AVATLABILITY I I EXPANDEDAVAILABILITY

tlrl &,J fir"rrtt^, i/ / V4 ffi fl1 \I t! t*t // /r t f"? (RobertMcLeod, Co-chair) (Date) FI t.tw

ri&.J ffi FI $1 !-i €'I &.I EXECUTIVESUMMARY OF A WORKSHOPON THE IMPACTS OF LANDCLEARING ON THE YI HYDROLOGICAND EI AQUATICRESOURCES OF BOREALFORESTS IN ALBERTA, NOVEMBER18 AND 19. 1994 d'l g1 STUDYPERSPECTIVE

et A majorgoal of the NorthernRiver Basins Study is FI to understand and characterize the cumulative &'J effects of developmenton the water and aquatic Related Study Purpose environmentof the Peace,Athabasca and Slave fft purpose Frl rivers and their major tributaries. Through 2.2 The of the Study is to &J coordinationwith existing initiativesand initiating understand and characterize the appropriatenew technicalstudies the sponsoring cumulative effects of development on governmentsof Canada,Alberta and the Northwest the water and aquatic environment of Territorieshoped the Board would be able to make the study area by coordinating with H recommendationsto better predictand assess the existing programs and undertaking cumulativeeffects of development. appropriate new technical studies. FI Ii F., Under the terms of the Study agreementand in subsequentcorrespondence between the Board and the sponsoringgovernments, the Board has been advisedthat the emphasisof the Studyis to be on the water componentof the Peace,Athabasca and Slave rivers.The Board,public and ScienceAdvisory Committee expressed concern over the Study being limited t to an examinationof the mainstem Peace, Athabascaand Slave rivers and their major tributaries.As a consequence,the StudyBoard accepted a recommendationof its independentScience Advisory Committee sfl tst to sponsor a workshop dealingwith the effects of land-clearingwithin the boreal mixedwood forest *:l on the &.! aquaticenvironment. The Boardalso accepted the recommendationto contractthe work to Dr. BruceDancik, Universityof Alberta,in lightof Dr. Dancik'searlier work in the areaof forestrypractices and the environment.

6.1 The workshop was structuredto bring together a broad cross-sectionof invited representationfrom the &J variousdisciplines involved with ecologicalresearch and management of land and water applicableto the borealmixed wood forest.A combinationof plenaryand discussiongroup sessionsprovided the framework for the workshopto providefeedback on threeareas of Board/ ScienceAdvisory Committee interest. Those areas included:expert assessment and consensusof the significanceof land-clearingon preservingthe ffi ecological integrityof the Basins aquatic resources,consensus on the state of knowledgeand research priorities,and consensuson whatthe Studyand othersshould be doingon land-clearing. ffi The proceedingsof the workshopwere taped and transcribedand are availablethrough the StudyOffice. This report provides the text of the presentationsby the keynote speakersand a synthesisof the workshop discussions,with recommendations.Much of the emphasiscontained in the six key goals identifiedfor gt immediateaction and eightgoals for furtherresearch, focus on strengtheningthe levelof knowledge,building alliancesfor additionalresearch, and improvingthe levelof integratedland /water managementand decision making. The findingsof this workshopwill be used by the Board in formulatingrecommendations that tr;tr iti recognizethe inalienablerelationship of landand water. &.1 {i;; f:i &,i {-: Fr {} REPORT SUMMARY gj The Northern River BasinsStudy (NRBS) workshop on the impactsof largescale land clearingon the hydrologic andaquatic resources of the NorthernRiver Basinswas heldNovember 18 and 19, 1994in Ed-ooton, Alberta.Court reportersgenerated a transcriptof over548 pages,from which this executive EI summarywas derived.This summaryincludes precis of expertpresentations on the following topics:the historyof land-usepolicies in Atberta the hydrologicimpacts of forest land clearing,forest impactsand the extent of harvestingin NorthernAlberta, hydrologic impacts of agriculturalproduction" impacts of land fli clearingactivities on waterquality, approaches to managementat thewatershed scale, food chainsand large scaleland clearing,and social and humanissues associated with largescale land clearingand development g of borealecosystems. Thisreport alsoincludes a summaryof the discussionsof expertland managers, industry representatives, g andscientific researchers who took part in breakawayand plenary sessions that were designedto answer a series of relatedquestions. As a result,the body of this report reflectsthe expert knowledgeand experienceof participantsrelevant to 1) the uniquegeography, soils, climate, hydrology, water quality, vegetationand wildlife of the borealforest, 2) the impactsof land clearingactivities, agriculture, forestry, B hydrologyand roads on thesenatural processes and resources, 3) researchneeds interms ofagriculture, forestry,hydrolory, water quality, biology andgeneral land disturbanceactivities, 4) larger researchissues planning,Iand-use decisions, regulations, approaches to science,models and the ; suchas goals,strategic scaleof research,and 5) the challengesand advantages of interdisciplinarystudies. 6 The conclusioncontains comments about points of generalconsensus that emergedfrom the workshop as well as very specificrecommendations for further actionby the ScienceAdvisory Committeeof the g NorthernRiver Basins Study. g E €t 6] B ffi El il H e-1 E.t E,:i

F:l 9i f.ra1 TABLE OF CONTENTS

ii REPORT SUMMARY NngSSfUnYAREa """ ii rii I ... iii i:,t FOREWORD ti ACKNOWLEDGMENTS TABLE OF CONTENTS vl 1.0 INTRODUCTION 1 ii I i. ] I.I ABOUT THIS PUBLICATION 1 L.2 WORKSHOP DESIGN . . . ., fl 1.3 KEY QUESTIONS &J 2.0 PRECISOF PRESENTATIONSDELIVERED BY GUESTSPEAKERS J 2.I OPENING REMARKS ON THE BACKGROUND FOR THIS WORKSHOP . 3 FI 2.2 OVERVIEW OF THE HYDROLOGIC IMPACTS OF FORESTLAND &J CLEARING . .. 4 2.3 OVERVIEW OF FORESTIMPACTS AND THE EXTENT OF HARVESTING IN NORTHERN ALBERTA . 6 fll 2.4 OVERVIEW OF HYDROLOGIC IMPACTS OF AGRICULTURAL PRODUCTION 8 )< IMPACTS OF LARGE SCALE LAND CLEARING ON WATER QUALITY 10* ndlil 2.6 APPROACHESTO MANAGEMENT AT THE WATERSHED SCALE . . . t2 )1 FOOD CHAINS AND LARGE SCALE CLEARING . .. 15 ffi 2.8 SOCIAL AND HUMAN ISSUESASSOCIATED WITH LARGE SCALE LAND CLEARING AND DEVELOPMENT OF BOREAL ECOSYSTEMS 16 2.9 SUMMARY OF THE ROUND TABLE DISCUSSION. . l8 ?{ gl 3.0 SUMMARY OF BREAKAWAY SESSIONDISCUSSIONS ?{ 3.1 UNIQUE FEATURESOF THE BOREAL FOREST )1 3.2 IMPACTS OF LAND CLEARING ACTIVITIES ffi 3.3 STUDIESNEEDED 32 3.4 LARGER RESEARCHISSUES 37 ga B. l 3.5 INTERDISCPLINARY STUDIES 45 EJ 3.6 WHO SHOULD BE DOING THIS RESEARCH?. . 46 4.0 CONCLUSION 47 4.1 RATIONALE . . 47 fl 4.2 GOALS FOR IMMEDIATE ACTION 49 SPECIFICGOALS FOR RESEARCH 49 fl 4.3 Li 5.0 REFERBNCES 50 APPENDICES

il APPENDIXA TERMS OF REFERENCE 56 f'tr APPENDIXB WORKSHOPPROGRAM 3/ U APPENDIXC QUESTIONSFOR BREAKAWAY SESSIONS 62 APPENDIXD LIST OF INVITEES/ATTENDANCELIST 69

FItJ

$t :t vl f.i tt Ii

FI 3.1.6 Wildlife E1The boreal forest contains a variety of large mammals such as moose, white-tailed and mule deer,black bear, and wolves; small mammals on which the larger animals feed; subarctic birds such as the sandhill craneand caspiantern; and a number of neotopical migratory species.The study areahas a low diversity ti of fish, reptile and amphibian species.Little is known about some of these species.For example, the migratory routes of fish and how they adapt to the higttly turbid waters of some northem streamsare poorly understood.The importance of this adaptability and the potential impact of land use on wildlife are difficult gl to assessbecause of the lack of reliable population inventories of speciesin the boreal forest. gl 3.2 IMPACTS OF LAND CLEARING ACTIVITIES Generally,workshop participants agreedthat the causesand types of impacts from land clearing will be the samein the Northern River Basins as elsewhere.The consequencesof land clearing activities in the ffi study area will, however, manifest themselves in ways particular to the geography, soil, climate and hydrologic conditionsthat exist there.ln most cases,breakaway session participants articulated what they expectis happeningor will happenin the study area,based on their knowledgeand experiencewith similar H land clearing activities elsewhere. Everyone acknowledged that the impacts of forestry are better documentedthan impacts from agriculture, oil and gas extraction, mining activities, human settlement I activities and urban development. Ht H 3.2.1 Agricultural ImPacts Agricultural land clearing activities, are a major agent of change.Fifteen million hectaresof land within the study area are currently under cultivation compared with annual forest harvesting activities that affect H a much smaller area. Besidesaffecting alarger areathan forestry does, agticulture maintains the land in an un-treed state and introduces large amounts of pesticides into the environment. Furthernore, summer fallowing, a practice which makes the soil very susceptibleto erosion, is used on about 10 percent of g cultivated lands (C.F. Bentley, personalcomment, Land DisturbancesTranscript, p. 6).

The needfor more sustainableagricultural practicesand quantifiable evidenceto support them becomes R even clearer when pressureto expand agriculture norttrward is considered. Such pressureis likely to occur in responseto declining soil productivity, decreasedavailability of land, and predicted climate changesin H more southerly locations. A changein the micro-climate is anotheranticipated consequenceof agricultural land clearing activities. The accumulation of snow on the ground in larger, wide-open areas created through agriculture and E forestry land clearing, can result in a net loss of available moisture when compared with the moisture 'fl retainedwhere snow accumulatesin the undisturbedforest. In large, open areassnow is more exposedto sun and wind and disappearsat a faster rate during spring thaw than snow in the forest. In addition, winds havebetter accessto snow on theseopen areas,so that sublimation and evaporationincrease and the snow E] pack disappears into vaporous form. ln general cleared lands usually become drier. Snow pack fr 27 tit.! accumulationis maximized when the diameterof the clearedareas is about twice the height of surrounding trees.

Water quality is also likely to be affectedby agriculture.Phosphorus and other nutrient loading in streams and rivers, for example,tend to be high in cultivatedareas. Increased sediment in rivers also occursas a result of agriculture, and the high organic content of waters in the boreal forest predisposesthem to attachmentof metals. While there is no evidencethat this has happenedto the extent of being toxic, this combination of elements might cause lower levels of dissolved oxygen, which can have significant detrimental effects on fish and aquatic biota.

In terms of pesticides,those that are very water soluble don't bind to the soil and are less of a problem. Herbicides like Atrazine and Simazine can causeproblems (D. Grant, personal comment, Water Quality Transcript, p. 19). Even small concentrationsof pesticidesor other chemicals, dioxin for example, may endangerhuman or aquatic populations. Furthermore, the use of pesticides and fertilizers has increased phenomenally in the last 20 years and that has major implications for water quality. Agricultural land clearing or clearing on private land is not regulated in terms of environmental concerns (G. Hillman, written responseto noteson water Quality Transcript,p.5,23 February1995).

3.2.2 Forestry Impacts

The significant differencebetween foresty clearing and land clearing for agriculture is that lands cleared for forestry can potentially return to a more natural state (i.e., forest cover), whereas cultivated land remainsin continual use.However, this assertiondoesn't mean that forestry activities have a low impact. ln most casesthe impacts of forestry may not look the sarneas agricultural impacts, but they are felt. For example, while lauded by the timber industry and landowners, modem fire prevention programs have alteredthe age of the forest. More old-age classesof timber now exist than was the casewith natural fire cycles.Subsequently, natural processes of hydrology, tee regeneration,runoffand evapotranspirationhave been altered (M.T. Dick, personal comment, Hydrology Transcript, p. l5)

Given the poor hydraulic conductivity and poor drainage of soils in the boreal forest, removal of trees under certain conditions can exacerbatethe existing high water tables and contribute to flooding. Sustainableforesffy practices, such as regenerating spruce are often impossible where water tables are high becausetree seedlingscannot compete with phreatophyic speciessuch as grasses,willows and alders that take over theseharvested sites (R. Rothwell, written responseto draft report, p. 10, 23 May 1995).

In areasof the boreal forest presently conducive to peatlandformation (paludification), large scale forest clearing will produce unfavourable conditions for this process (D. Klym, personal comment, Land Disturbances Transcript, p. 25).It is expected that land clearing will cause an increasein runoff and a decreasein groundwater which rechargesthese wetland areas.Complicating this processis the fact that impactsare location specific.After logging, somepeatland areas have becomemarshy, other wetlandshave dried up (I.G. Coms, personalcornment, Land DisturbancesTranscript, p.24).It may take severalhundred yearsfor a site to reach its vegetation climax state.After harvesting,reseeding or replanting an areawith the sametree speciesthat were harvestedmay be appropriate.It may be necessaryecologically to ensure

28 t". {i

)itF-1

&iFl that the site go through the samenatural successionalstages before vegetation can reach the stagethat existedwhen the areawas harvested(lrl. Parker,personal comment, Land DisturbancesTranscript, p.l3). slri Li Finding low-impact forest practicesis a difficult and complex task. From a technicalperspective, the trend r1 toward larger clear-cutsoffers potential savingsin termsof reducingsoil impacts;but increasesthe impacts Li on hydrology. Those impacts will likely be more severebecause future harvesting activities may involve loggingon relatively steepslopes and adjacentto major rivers (i.e., the AthabascaRiver Valley, Wapiti, rjl Little Smoky, Clear fuver). It should be noted that the forestry industry FI is looking at alternativi methods t. I &l to clear-cuttingwhen harvestingon problem sites.Furthernore, ground rules set out by Alberta Land and Forest Services do not allow harvesting on riparian zueasor on slopes greater than 45 percent, with tl conventionalwheeled skidders (G. Hillman, written responseto draft report,p. 10, 8 May 1995). Besidestechnical considerations,it is difficult to define and quantifli sustainableuse of forest and other natural resources.Economic conditions influence decisionsmade by the forestry and agriculture industries $ as well as local communities.With a greaterunderstanding of technical and economicconcems among the generalpublic, clearly defined social values, a commitment to holistic, flexible long-rangeplanning and effective dialoguebetween representatives of commercialand public perspectives,it might be possiblefor H interestgoups to arrive at somesatisfactory, achievable conservation goals. Over the courseoith" NRBS workshop,participants acknowledged and placed increasinglyurgent emphasison achieving such a state of aflairs and idenrifying precedentswhere membersof scientific, industrial and local "o*rrr,roities have n.been able to learn from each other and plan cooperatively.

$l 3.2.3 Impacts on Hydrology $i: ..) tl Participants'viewson the impactsof land clearing LI activities on the hydrology ranged from predictions of hJ conservativechanges in hydrology to wamings about catastrophicincreases in water yield. Conservative estimatessuggested that there will probably be no changeor an increasein water yield, but that will be very scaledependent. At the other end of the continuum, someexperts predict that flood eventscould occur for tr the next 30 to 50 years (C. Hunt, personal comment, Biology Transcript, p.46)- There was agreemenr, however, that land clearing will likely include someor all of the following hydrologic impacts:

H o changesto water tables and water retention capacityof soil, a slow recovery of evapotranspirationprocesses, gi o changesin the capacity of peatlandsto store water, o reduction in the size and number of wetlands, . potential for increasedflow causing degradation(downcutting) of rivers and streamsat some locations and aggradationof river and streambeds (accumulation of sediment)at other tr locations, . decreasedstream gradients, gl . low nutrient soil environments,and . changesto sediment levels, water yield, water temperature,and aquatic biota. H] fl L.l 29 gif,: *rf-l fi E'! F] Ei 3.2.4 Impacts on Water QualitY

Because agriculture, forestry and hydrology have significant impacts on water quality and biological H organisms,-manyof the commentsabout impacts arising out of this discussion group have been included ,rria", those respective headings. The water quallty goup generateda considerable amount of information These on how to encourageinterdisciplinary studies and improve communication among stakeholders. tr suggestionsare containedin the Larger Research Issues sectionof this report'

goup. It was noted Defining the term water quality quite rightly received some attention in this breakaway H to abrupt that water quality can vary dependingupon one'sperspective. Aquatic organisms are sensitive levels changesin wateichemistry and must have certain temperatures,pH and specific dissolved oxygen water are more * *"lt as other water conditions to survive. Standardsfor human consumption of drinking ffi factors, stringent for factors like sediment than are standardsof safe water for aquatic organisms. For other standardsfor human use are less stringent. $ 3.2.5 Impacts on Wildlife

participants will have qualitative impacts on biology, but that quantitative w agleed that land clearing about animal assessmentof impacts will be difficult to do becausethere is little or no baseline information to populationsin the boreal forest. They also acknowledgedthat the make-up of speciesgroups is known like magpie t .f,*g. after logging. Local observationssupport this prediction, reporting that some species insectsfound have increasedandothers have disappeared.Dr. Robert Steedmanadded that the types of changes(R. near streams can be used as indicators of geomorphic characteristicsand environmental B Steedman,personal comment, Biology Transcript, p'79)'

from all types Habitat fragmentation, which affects wildlife migration among other behaviows, has resulted the rivers ffi of land clearing activities. For example, the larger trees grow where there's more moisture, along zones of and valleys, so that's where harvesting is often done (except on Crown lands where buffer are very undisturbld riparian vegetation are maintained). This is unfortunate because, riparian zones w Forests and important in terms of local and migrating wildtife that depend on them for a variety of foods- they offer the small openings that border on rivers and lakes are also important to wildlife because creeks protection from weather and predators. Besides directly reducing treed habitat, harvesting on rivers, B term, and islands affects the amount of water available to the vegetationon flood plains. Over the long Transcript, p. changesin vegetationinfluence ungulatebehaviour @. Wynes, personalcomment, Biology 42). B 3.2.6 Impact of Roads w participants agreed Discussionsabout the impacts of roadsbecame rather contentious at times. However, stream that impacts of the removal of vegetation combined with roads, drainage ditches, culverts and changes ffi crossingsconstructed for resource extraction have significant impacts on hydrologY, more so than p.21-22). to vegettive cover alone (F. Davies/A. Plamondon,personal comments, Hydrology Transcript, ffi

30 ffi ffi ri i. J i:'t til ti Roads,for example,can interferewith hydrologic systemsthrough a damming effect, and streamcrossings il are important sourcesof silt loading (G. Hillman, written responseto notes on Hydrology Transcript, p. E"l 6, l5 FebruNy 1995). With the network of peatlandsthat occur in the boreal forest, resourceextraction activities are frequentlylimited to winter when the ground is frozen. The building of snow bridgesand the impact of heary equipment used for crossing streams at this time of year may also be significant. ff Generally, the cumulative effect of sediment releasedby road crossing and accessactivities, rather than individual road constructionevents, were consideredto be continuing sourcesof sediment.As the sediment gtF.'l migrates downstream,it has different impacts at different levels (C. Hunt, personal comment, Biology t, l $,J Transcript,p.64).

Frf $l In the Pacific Northwest, 80 to 90 percent of sediment problems in streams are the result of roads. LI Solutions to the problem of impacts from road construction are quite complex due to the very strong culturaVsociaVeconomicvalues that various stakeholdersattach to them. For example, roads give access g to vacation homes,recreational activities such as fishing, and a spectrum of commercial activities (C.F. Bentley,personal cornment, Land DisturbancesTranscript, p.42). Participants acknowledgedthat roads, drainage,culverts and steam crossingscan have more impact on an ecosystemthan changesin vegetative cover. They also agreed that progress toward lower impact road and drainage activities needs to be ffi supportedby better community liaison, educationand government regulations rather than research. i 3.2.7 Impact Summary The impactsof forestry have receivedmore attention,in terms of researchand public awareness,than the E] impacts of agriculture, oil and gas extraction and mining activities. Similarly, impacts from human settlementsuch as road construction and sewagetreatnent are regulatedto some extent, but have a low fl profile in terms of researchand public awarenesswhen comparedto commercial activities. There is a perceptionthat the Northern River Basins are relativeiy untouchedby human activities, but in fact the whole landscapeis being managedeither directly or indirectly (D. McNabb, personalcomment, € Water Quality Transcript, p. 37).Even if the north continues to open up at a rate slower than that of the prairies, the following current and predicted impacts are bound to increase: g o alteredwater chemistry, o soil erosionand flooding, o inhospitable habitats for indigenous wildlife and aquatic biota, ffi o increasedwater yield from cleared areas, . decreasedgroundwater loading, f*l tl o changesin water table (lower or higher) dependentupon parent soil material, &*ttl o a dryer region due to increasedsnow sublimation and evaporation,and ffi o a hisher water table. fi LI tr 3l F1 LJ Given the number of different land clearing and resource extraction activities being pursued in the Northem River Basins,participants shared a common perceptionthat continuedlack of cooperationamong stakeholderscould have as yet unknown cumulative impacts on the region and that these impacts will be felt downstream.

3.3 STUDIESNEEDED

Throughout the four breakaway discussion groups, a few general and specific areasfor study received common emphasis.Some of the larger researchneeds that were mentionedincluded the need for research into the potential impacts of land use on global warming and climate changeand the effects of climate changeon the study area.Forhrnately, these impacts are currently being studied as part of the Mackenzie River Basin Impact Study (G. Hillman, written responseto Biology notes,p. 3). Researchersshould, however, be giving some thought to the type of timber and vegetationthat would thrive under an altered climate.

It was also noted that the amount of researchin the western part of the boreal forest is limited when compared to what has been done in the eastern portion. Another general area for future research is palalontology. The rationale for this suggestionis that climates of past agescould recur. Palaeontology tould help predict the potential effect of climate change on the natural resourcesin the Northern River Basins (D. Klym, personal comment, Land DisturbancesTranscript, p. 64).

Consensus also emerged from breakaway sessions, presenters, and plenary discussions about the proliferation of research questions, information needs, and the tendency for scientists to examine, ie-examine and expand upon each other's work without directly addressingthe land managers' urgent need for sustainableoptions. "Do ln responseto the implied question, we needfurther researchor a really good review of research?"(B- Swanson,personal comment, Biology Transcript, p. 51), participants suggestedthat a thorough literature searchbe -ompiled and reviewed by a multi-disciplinary panel. It was also suggestedthat some form of centralbulletin board on the lnternet, or even an existing agency should be used to centralize, update and distribute information on relevant studies. The Northwest Territories Science AcL which requires registrationof relevant studies,was cited as a meansof ensuring a central clearing house for research(K. Crutchfield,written responseto draft report,p.1,2 May 1995).

More specifically,the biology, hydrology, water quallty and land disturbancebreakaway sessions identified the need to better understand hydrologic processes,the sources and pathways of sediment, and the cumulative impacts of land clearing. Participants also felt strongly about the impact of roads, stream crossings,drainage and culvert construction on hydrology and suggestedthat these factors be looked at from a land managementor regulatory standpointrather than as topics for scientific research.

32 REPORT # ESD/LM/99-1

GUIDELINES FOR RECLAMATION TO FOREST VEGETATION IN THE ATHABASCA OIL SANDS REGION

ENVIRONMENTAL SERVICE

Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page i Prepared By: The Oils Sands Vegetation Reclamation Committee

EXECUTIVE SUMMARY

The Oil Sands Vegetation Reclamation Committee was formed in November 1996 with the mandate to prepare guidelines on the establishment of forest vegetation (ecosystems) for reclaiming oil sands leases in northeastern Alberta. The Committee focussed on starter vegetation and design criteria for ecosystems that would support primarily commercial forests and secondarily would provide wildlife habitat. In addition, biodiversity was considered an important aspect of reclaimed ecosystems. The guidelines have been based on successful reclamation techniques, and research and monitoring information that were available for the oil sands region at the time of document preparation. As research and monitoring programs continue in the region, new data will be used to update and refine the suggested approaches and techniques every 5 years.

The guidelines are intended to be used by government and industry staff. They provide detailed information on what terrestrial vegetation (ecosystems) can be re-established to support commercial forests and wildlife habitat, how to establish the ecosystems through reclamation techniques, and how to monitor whether the reclamation approach has been successful. Information gaps that exist and assumptions that have been made have been documented.

The Committee designed a seven step process to meet their mandate. The main conclusions are discussed below.

Step 1 – Identify Target Ecosites that can be Established on Reclaimed Landscapes

The first step was to identify which ecosites, of those that occur in the oil sands region, can be supported on reclaimed landscapes.

The ecosites that naturally occur in the oil sands region were identified based on the ecological classification system for northern Alberta produced by Beckingham and Archibald (1996) and are summarized in Section 3.1. The ecosites that can be re-established on reclaimed leases depend on the landscapes, drainage patterns and soils that can be re-established. The ecosites that are emphasized in this document are those that are most likely to develop on the range of landscape features (position, slope and aspect) and soil types that currently exist in reclaimed areas on the Syncrude Canada Ltd. (Syncrude) and Suncor Energy Inc. (Suncor) leases.

Step 2 – Identify Techniques to Establish Ecosites on Reclaimed Landscapes

Techniques that have been developed through research and monitoring by Syncrude and Suncor to reconstruct the landscape/drainage patterns and soils, and to initiate “starter vegetation” that should eventually succeed to the targeted ecosites on their oil sands leases were compiled. They are summarized in Section 3.2. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page ii Prepared By: The Oils Sands Vegetation Reclamation Committee

Step 3 – Identify Terrestrial Land Use Objectives for Reclaimed Landscapes

The Oil Sands Vegetation Reclamation Committee examined the design criteria for ecosystems and landscape components that would meet two land use objectives: the establishment of stands of commercial forest and the establishment of wildlife habitat.

Design criteria for other land uses that can be examined at a later date include recreation and traditional use. Ultimately, each oil sands lease will be designed to support several integrated land uses.

Step 4 – Identify the Design Criteria for Ecosystems and Landscape Patterns to Achieve the Selected Terrestrial Land Use Objectives

The fourth step in the process was to identify the design criteria required for the reclaimed landscape to support commercial forest and wildlife habitat. Design criteria included information on the preferred types of ecosites and preferred patterns of distribution of ecosites in the landscape. The design criteria for commercial forest are outlined in Sections 4.1 and 4.2. The design criteria to create habitat for ten species of wildlife are outlined in Sections 5.1.1 and 5.2, and Appendix J.

Step 5 – Integrate Design Criteria for All Land Use Objectives for Reclamation of Oil Sands Leases

It is intended that lease holders will use these guidelines to assist in designing their landscapes to achieve the required end land uses. Once the design criteria for each of the primary and associated land uses have been identified, the distribution pattern of the preferred ecosites for the primary land uses within the leases needs to be identified. Then the design criteria for the associated uses need to be integrated into the design criteria for the primary uses. This is an important planning exercise that should be completed by an operator of an oil sands development during the design of the closure plan. This step is not discussed further in this guideline manual.

Step 6 – Design Monitoring Programs to Verify that Land Use Objectives have been Successfully Met

The sixth step in the process was to design monitoring programs to determine if the land use objectives have been achieved. These have been outlined in Sections 6.1 and 6.2 of this document. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page iii Prepared By: The Oils Sands Vegetation Reclamation Committee

Recommendations for monitoring programs for the establishment of a commercial forest include:

· Confirmation that tree seedlings have been established (as benchmarked through compliance requirements under the Timber Management Regulations and Alberta Regeneration Survey Manual), and

· Confirmation that forest productivity has been re-established (as benchmarked through measurements of tree height, mean annual increment and site index, and soil capability).

Recommendations for monitoring the establishment of habitat capability for wildlife include:

· Measurement and modeling of biophysical characteristics of re-established ecosites and physical features of landscape components (fine filter approach), and/or

· Interpretation of biodiversity information on the plant community and landscape levels (coarse filter approach).

Recommendations for monitoring biodiversity on reclaimed landscapes includes the collection of information on the landscape, plant community and genetic levels of variation.

Step 7 – Recommend Research Programs to Address Information Gaps on Reclamation

Research programs to address information gaps identified in the reclamation information include:

· Vegetation productivity (site index) on reclaimed soils needs to be measured and compared to vegetation productivity on natural soils,

· The relationship between soil capability classes and vegetation productivity (site index) needs to be identified through monitoring programs,

· It is uncertain whether all plant species recommended as starter species can be propagated on reconstructed soils. Future propagation of additional plant species, such as those in Appendix H, needs to be studied,

· The survival and vitality of plant species moved through direct placement from various ecosystems to reclamation sites supporting different subsoils needs to be examined,

· The potential effects of elevated pH levels in reconstructed soils on plant growth are not known and need to be studied,

· The feasibility of using mineral soil through direct placement to develop upland ecosystems needs to be examined,

· The effectiveness of covering and reclaiming saline/sodic materials needs to be studied, Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page iv Prepared By: The Oils Sands Vegetation Reclamation Committee

· The feasibility of continuing the use of exotic plant species in reclamation needs to be examined,

· Methods to measure biodiversity need to be developed,

· Draft seed zone maps for the oil sands area were developed with the intention of covering both woody and other vascular plants but may need to be further refined to determine their suitability for other vascular plants,

· The feasibility of re-establishing ecosystems on consolidated/composite tailings (CT) needs to be researched,

· The feasibility of applying sulphur to reduce soil pH and sodicity needs to be examined,

· Methods to enhance the establishment of native understorey species to achieve greater biodiversity than is possible through seeding/planting need to be developed,

· The biology and productivity of reclaimed soils should be examined. Mycorrhizae, nutrient cycling and sustainability of peat-mineral mix amendments to ensure that the “living” components of the soil system are functioning effectively and in balance should be the focus of the research,

· Seed sources for species that provide good timber/fibre production need to be improved,

· Essential ecosystem functions and plant species required to accomplish these functions need to be determined, and

· The ability to create ecosites d (low-bush cranberry) and e (dogwood) without adding clay is uncertain. The sustainability of the ecosites needs to be studied. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page v Prepared By: The Oils Sands Vegetation Reclamation Committee

ACKNOWLEDGEMENTS

The Oil Sands Vegetation Reclamation Committee has been comprised of the following representatives:

· Earl Anderson Syncrude Canada Ltd.

· Harry Archibald Alberta Environmental Protection Land and Forest Service

· Bert Ciesielski Alberta Environmental Protection Land and Forest Service

· Tom Coolen Syncrude Canada Ltd.

· Garry Ehrentraut Northland Forest Products Ltd.

· David Fox Alberta Pacific Forest Industries

· Chris Hale Alberta Environmental Protection Land and Forest Service

· Leonard Leskiw Can-Ag Enterprises Ltd.

· Jeff Sansom Alberta Environmental Protection Land Reclamation Division

· Judith Smith Shell Canada Limited

· Steve Tuttle Suncor Energy Inc.

· Carl Warner AGRA Earth & Environmental Limited

Other technical information was supplied by:

· Narinder Dhir Alberta Environmental Protection Land and Forest Service

· Grant Klappstein Alberta Environmental Protection Land and Forest Service

· Calvin Duane AGRA Earth & Environmental Limited Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page vi Prepared By: The Oils Sands Vegetation Reclamation Committee

TABLE OF CONTENTS

PAGE

EXECUTIVE SUMMARY...... i ACKNOWLEDGEMENTS...... v TABLE OF CONTENTS...... vi LIST OF TABLES AND FIGURES ...... viii LIST OF APPENDICES ...... ix

1. INTRODUCTION...... 1 1.1 MANDATE AND OBJECTIVES OF THE COMMITTEE...... 3 1.2 USE OF THE GUIDELINES BY GOVERNMENT AND INDUSTRY ...... 4 2. APPROACH USED TO PREPARE GUIDELINES...... 5 2.1 GUIDING PRINCIPLES...... 5 2.2 PROCESS FOR ESTABLISHING VEGETATION (ECOSYSTEMS) TO MEET VARIOUS LAND USE OBJECTIVES ...... 8 3. ECOSITE ESTABLISHMENT ...... 15 3.1 ECOSITES TO BE ESTABLISHED ON RECLAIMED LAND ...... 15 3.2 ESTABLISHMENT OF ECOSITES ON RECLAIMED LANDSCAPES ...... 19 3.2.1 Reclamation Goals ...... 19 3.2.2 Landscape and Soil Construction ...... 20 3.2.2.1 Terrain Development ...... 20 3.2.2.2 Terrain Stability Objectives...... 21 3.2.3 Reclamation Approach...... 23 3.2.3.1 Soil Reconstruction...... 23 3.2.3.2 Reclaimed Soils...... 24 3.2.3.3 Vegetation Establishment ...... 27 Revegetation Objectives ...... 27 3.3 SUMMARY OF UNCERTAINTIES AND DATA GAPS IN ECOSITE RECLAMATION AND FUTURE RESEARCH...... 31 4. DESIGN CRITERIA FOR COMMERCIAL FORESTS...... 33 4.1 DEFINITION OF, AND CONDITIONS TO SUPPORT COMMERCIAL FOREST...... 33 4.1.1 Operating Restriction Guidelines ...... 34 4.1.1.1 Terrain Considerations ...... 34 4.1.1.2 Stand Characteristics...... 34 4.2 ECOSITES THAT MEET THE DEFINITION OF A COMMERCIAL FOREST...... 35 4.3 METHODS TO MAINTAIN GUIDING PRINCIPLES ...... 35 Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page vii Prepared By: The Oils Sands Vegetation Reclamation Committee

5. DESIGN CRITERIA FOR ASSOCIATED LAND USE - WILDLIFE HABITAT...... 37 5.1 DEFINITION OF AND CONDITIONS TO SUPPORT WILDLIFE HABITAT ...... 38 5.1.1 Target Wildlife Species and Habitat Requirements ...... 38 5.2 ECOSITES AND LANDSCAPES THAT MEET THE DEFINITION OF WILDLIFE HABITAT ...... 39 5.3 MAINTAINING GUIDING PRINCIPLES AND RECLAIMING DISTURBED AREAS AS WILDLIFE HABITAT ...... 39 5.3.1 Landscape Diversity...... 40 5.3.2 Plant Community Biodiversity ...... 41 5.3.3 Starter Plant Species ...... 42 5.3.4 Slash and Deadfall ...... 42 6. MONITORING PROGRAMS ...... 43 6.1 MONITORING PROGRAM TO VERIFY ACHIEVEMENT OF COMMERCIAL FOREST OBJECTIVES ...... 43 6.1.1 Compliance - Establishment of Seedlings ...... 43 6.1.1.1 Benchmark - Reforestation Regulations...... 43 6.1.1.2 Forest Establishment Program Design...... 45 6.1.2 Long-term Forest Productivity...... 46 6.1.2.1 Benchmark - Mean Annual Increment and Site Indices ...... 46 6.1.2.2 Benchmark - Measurement of Soil Capability...... 47 6.1.2.3 Long-term Forest Productivity Program Design ...... 47 6.2 MONITORING PROGRAM TO VERIFY ACHIEVEMENT OF WILDLIFE HABITAT OBJECTIVES...... 49 6.3 DIVERSITY/BIODIVERSITY MEASUREMENT OPTIONS ...... 50 7. ONGOING PROCESS REFINEMENT BASED ON MONITORING AND RESEARCH...... 51

8. GLOSSARY OF TERMS...... 51

9. LITERATURE CITED...... 56 Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page viii Prepared By: The Oils Sands Vegetation Reclamation Committee

LIST OF TABLES

Table 3.1 Regional Mean Site Index by Species for each Ecosite for the Boreal Mixedwood, Northern Alberta...... 16 Table 3.2 Summary of Ecosites and Related Site Productivity and Soil Capability for the Boreal Mixedwood, Northern Alberta...... 17 Table 3.3 Reclaimed Soils, Soil Capability and Target Ecosites...... 18 Table 3.4 Planting Prescription by Ecosite Phase...... 29 Table 4.1 Stand Characteristics ...... 35 Table 5.1 Factors Used to Select Target Wildlife Species ...... 39 Table 6.1 Yield Curve for Commercial Forest Ecosites ...... 44 Table 6.2 Forest and Soil Monitoring of Benchmark Sites...... 48

LIST OF FIGURES

Figure 2.1 Types of Diversity/Biodiversity to be Addressed During Reclamation ...... 6 Figure 2.2 Flow Diagram of Process for Establishing Vegetation (Ecosites) for Different Land Use Objectives ...... 9 Figure 2.3 End Land Use Options Recommended by Oil Sands End Land-Use Committee. The Shaded Boxes Represent the Land Use Objectives Addressed in this Document...... 12 Figure 3.1 The Effects of Aspect on the Moisture Regime...... 22 Figure 3.2 Soil Handling Options and Related Soil Capability Classification for Overburden Reclamation...... 25 Figure 3.3 Soil Handling Options and Related Soil Capability Classification for Tailings Sand Reclamation ...... 26 Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page ix Prepared By: The Oils Sands Vegetation Reclamation Committee

LIST OF APPENDICES

APPENDIX A Existing Ecosites in the Oil Sands Region

APPENDIX B Capabilities and Diversity of Existing Soils Within Some Oil Sands Leases

APPENDIX C Acts, Regulations, Policies and Guidelines Relevant to Revegetation and Reclamation in the Oil Sands Region, Dated July 1997

APPENDIX D Research on Plant Varieties

APPENDIX E Ecological Diversity Monitoring Framework, Draft #6, Prepared for the Biodiversity Monitoring Working Group (August 1997)

APPENDIX F History of Terrestrial Reclamation in the Oil Sands Region

APPENDIX G Seed Zones and Sources

APPENDIX H Native Plants Suitable for Reclamation in the Central Mixedwood Subregion of the Boreal Forest Region

APPENDIX I Standards and Guidelines for Operating Beside Water Courses. From: Timber Harvest Planning and Operating Ground Rules

APPENDIX J Wildlife Populations and Habitat Capability in the Oil Sands Region Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 1 Prepared By: The Oils Sands Vegetation Reclamation Committee

1. INTRODUCTION

The Oil Sands Vegetation Reclamation Committee (herein referred to as the Committee) was formed in November 1996 to prepare guidelines on the establishment of forest vegetation (ecosystems) with emphasis on providing appropriate starter vegetation to use for reclaiming oil sands leases in northeastern Alberta. Vegetation in wetlands, commercial forest and environmental reserves is governed by the Alberta Forest Act and Government policy documents such as the Regeneration Survey Manual, Timber Harvesting Ground Rules, Recommended Native Grasses and Legumes for Revegetating Disturbed Lands in the Green Area, Forest Conservation Strategy and/or the Fort McMurray-Athabasca Oil Sands Subregional Integrated Resource Plan.

An integrated approach to oil sands lease development involves three reclamation steps: facility design and layout, land use planning, and developing reclamation strategies and techniques. The objective of reclamation is to establish sustainable ecosystems that replace equivalent levels of predisturbance land use capabilities. The ecosystems that will become established will be determined by the design of the reclaimed landscape, the types of reconstructed soils, revegetation techniques, and natural processes over time. In most instances, all ecosystems re-established after mining will contribute to more than one land use. However, some ecosystems and their landscape patterns may be designed to enhance their suitability for one particular land use.

Stakeholders in northeastern Alberta have requested that the oil sands leases be reclaimed to some of the predisturbance ecosystems and land uses. For example, commercial timber harvesting companies have requested that an equivalent predisturbance landbase and productivity of white spruce, jack pine, mixedwood and aspen ecosystems be returned. Aboriginal groups, the public, other industries and Government have asked that re-established ecosystems produce equivalent levels of productivity for wildlife habitat. Other important predisturbance values and land uses that could be re-established include watershed functions, wetlands, sources of traditional foods and medicinal plants, and recreation.

The guidelines in this document have been designed to provide its users with information on:

· Terrestrial ecosystems comprised of forst vegetation that can be supported by reclaimed landscapes and soils (i.e., nutrient and moisture regimes) established on oil sands leases (Section 3.1),

· Practical techniques to re-establish forest ecosites (Section 3.2),

· Information gaps on reclamation and research recommendations (Sections 3.4 and 5.3.3),

· Terrestrial ecosystems and landscape patterns that support two land use objectives for the oil sands leases region, i.e., commercial forest (Section 4.0) and wildlife habitat (Section 5.0),

· Practical approaches to monitor the re-establishment of ecosystems and to verify that the selected land use objectives have been met (Section 6.0), and Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 2 Prepared By: The Oils Sands Vegetation Reclamation Committee Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 3 Prepared By: The Oils Sands Vegetation Reclamation Committee

· Natural landscapes, soils, vegetation and ecosystems that occur in the oil sands region (Appendices A and B).

To date, several plant communities have been developed ranging from grassland complexes to a variety of forest cover types. The guidelines will help define and streamline the reclamation planning process.

The guidelines have been based on current reclamation objectives, successful reclamation techniques, and research and monitoring information available for the oil sands region at the time of document preparation. As research and monitoring programs continue in the region, new data will be used to update and refine the suggested approaches and techniques (Section 7.0).

A glossary of terms used in this document is included in Section 8.

1.1 MANDATE AND OBJECTIVES OF THE COMMITTEE

The Committee was formed following a November 26, 1996 workshop sponsored by the Alberta Land and Forest Service, and involving Syncrude Canada Ltd. (Syncrude), Suncor Energy Inc. (Suncor), and other industry representatives. The Committee was comprised of representatives from the oil sands and forest industries, government and consultants. These representatives provided a broad array of expertise on policy and regulatory requirements, reclamation approaches and techniques, soils, vegetation, forestry and wildlife for northeastern Alberta. The Committee’s mandate was to prepare recommendations that will provide guidance on the establishment of appropriate terrestrial vegetation (ecosystems) to meet two land use objectives in the oil sands region: commercial forest and wildlife habitat.

The objectives of the Committee were to identify:

· Approaches to oil sands lease reclamation that will meet regulatory guidelines and revegetation requirements,

· The forest ecosites that can be established on oil sands leases,

· The existing reclamation methods available to develop these ecosystems,

· The principles of landscape and ecosystem design to meet the land use objectives of commercial forest and wildlife habitat,

· The biodiversity objectives that should be considered,

· How the success of reclamation should be monitored, and

· The additional research that is required to address present information gaps and future improvements in reclamation research.

The approach and process used by the Committee to address these objectives and develop the guidelines are outlined in Section 2.0. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 4 Prepared By: The Oils Sands Vegetation Reclamation Committee

Recommendations of the Committee will be forwarded to the participating industry and government divisions. The recommendations are consistent with Provincial Acts, Regulations, Policy Statements, Manuals and Approvals that are applicable to vegetation, and reclamation of oil sands leases (Appendix C). For example, the “Recommended Native Grasses and Legumes for Revegetating Disturbed Lands in the Green Area” (Alberta Environmental Protection 1996b) states that vegetation used for ecosystem development needs to be native to eco-subregions. The ecosystems developed need to be sustainable and provide a sustainable range of uses.

1.2 USE OF THE GUIDELINES BY GOVERNMENT AND INDUSTRY

The guidelines are intended to be used by government and industry staff. They provide detailed information:

· What terrestrial vegetation (ecosystems) can be re-established to achieve two land use objectives (commercial forest and wildlife habitat) in the oil sands region,

· How to establish these ecosystems through reclamation techniques, and

· How to monitor whether the reclamation approach has been successful. Information gaps that exist and assumptions that have been made are documented. Research may need to be designed and conducted to fill some of the knowledge gaps.

These guidelines represent a working document that will be reviewed every five years and revised with new information. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 5 Prepared By: The Oils Sands Vegetation Reclamation Committee

2. APPROACH USED TO PREPARE GUIDELINES

This section outlines the guiding principles and process that were used by the Committee to prepare the guidelines.

2.1 GUIDING PRINCIPLES

Guiding principles that were followed in developing the approach to reclamation presented in this document are:

· Diversity/Biodiversity – The maintenance of ecological diversity has become a central goal of forest management in Alberta (AFCSSC 1997) and across North America (Hunter 1991; Everett 1993; Grumbine 1994). Consequently, government and industry must monitor forest ecosystems to ascertain if ecological diversity is being maintained as resources are extracted (Noss 1990; Christiensen 1996; AFCSSC 1997). Maintenance of biodiversity is required to meet Canada’s and Alberta’s commitments made in the National Forest Strategy (CCFM 1992), Canadian Biodiversity Strategy (EC 1995), and Rio Convention (UNEP 1992). In response to these commitments the Canadian Council of Forest Ministers (CCFM 1995) has agreed to a set of criteria and indicators for the sustainable management of Canada’s forests.

Biodiversity is a measure of ecosystem sustainability as represented by the range in variety of biotic and abiotic components at landscape, plant community and within species (genetic) levels. To successfully reclaim lands to a sustainable ecosystem, the vegetation systems should have a range of biodiversity similar to predisturbance values. Figure 2.1 outlines the types of biodiversity to be addressed during reclamation.

The creation of a variety of macro and micro landform components and soils on the reclaimed areas is required to create a range of ecosystems (ecosite phases) and hence biodiversity at the landscape level (Figure 2.1). By combining the landform components with biotic components, and further enhancing with spatial variability, the opportunity to approach predisturbance biodiversity exists.

Time is another key element in creating biodiversity levels similar to the predisturbance levels. Creation of a diverse landscape will certainly enhance the opportunity for species variation but this will not happen in the first year after the area has been reclaimed. The objective for industry will not be to try and create the level of biodiversity in year one that is characteristic of a mature ecosystem, but to provide the key elements to each site that will create an environment that over time will reach acceptable levels of biodiversity. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 6 Prepared By: The Oils Sands Vegetation Reclamation Committee

Figure 2.1 Types of Diversity/Biodiversity to be Addressed During Reclamation

Landscape Level of Diversity

Terrain Soils · Reclamation Substrates · Number of Soil Types · Slopes · Moisture Regimes · Aspects · Capabilities · Drainage Patterns · Chemistry · Topographical Diversity · Distribution Patterns (size, shape, · Distribution Patterns (size, shape, dispersion) dispersion)

Ecosystems (Ecosite Phases) · Number of Ecosite Phases · Number of Seral Stages · Distribution Patterns (size, shape, dispersion)

Plant Community Level of Biodiversity

· Number and Abundance of Plant Species · Vertical Structure Diversity · Productivity · Functional Diversity · Range of Site Indices · Abundance of Exotic or Hybrid Species

Within Species (Genetic) Level of Biodiversity

· Population Diversity (adaptive, productive and edaphic variation) · Genotype Diversity Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 7 Prepared By: The Oils Sands Vegetation Reclamation Committee

The plant community level of biodiversity addresses the number and abundance of plant species in ecosystems, the diversity of vegetation forms (vertical structure), productivity and functional diversity (Figure 2.1).

Conserving genetic diversity, and particularly population diversity, is the key element to ensuring that species retain their capacity to evolve and adapt to change within their environment (Figure 2.1). As genetic diversity is reduced, the risk of genetic maladaptation increases logarithmically. Genetic diversity is essential for sustaining the productive capability and resilience of forest ecosystems. Therefore, genetic diversity is the fundamental basis of the diversity of all species and the sustainability of the ecosystems.

To re-establish equivalent land use capabilities, careful consideration should be given to the component choices required to satisfy the needs of biodiversity, the types of commercial forest, wildlife habitat and other uses.

When planning mine and facility development, and during operational phases, careful consideration should be given to the preservation of refugia. Insitu conservation includes leaving original ecosystems intact from which diverse genetic plant material can emanate and provide for a variety of native species. In addition, consideration should be given to selecting vegetation species for reclamation that do not adversely impact surrounding ecosystems by preventing long-term restoration of biodiversity to natural levels.

Documents that support and give direction for biodiversity in forested landscapes include:

- The Canadian Framework of Criteria and Indicators for Sustainable Management and Defining Sustainable Forest Management, A Canadian Approach to Criteria and Indicators, published by the Canadian Council of Forest Ministers (CCFM 1995),

- The proposed Forest Conservation Strategy, Alberta Environmental Protection, Land and Forest Service, Alberta Environmental Protection (AFCSSC 1997), and

- Ecological Diversity Monitoring Framework, Draft #6, Prepared for the Biodiversity Monitoring Working Group (Schneider 1997) (Appendix E).

· Sustainability - Sustainable ecosystems are able to adapt and evolve over time.

· Adaptive Management Approach – Reclamation approaches and techniques should be regularly improved based on the results of ongoing research and routine monitoring.

· Productivity - Forest productivity across the landscape should be re-established to levels that existed before the mining operations were initiated. Productivity is defined as the site index at 50 years. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 8 Prepared By: The Oils Sands Vegetation Reclamation Committee

· Ecosite - The ecosite classification system for northern Alberta (Beckingham and Archibald 1996) was used as a guide to select plant species for reclamation. The definitions of ecosite, ecosite phase and plant communities are presented in the Glossary of Terms (Section 8) and in Appendix A.

· Native and Exotic Species – Plant species native to the Central Mixedwood Natural Subregion should be used for reclamation. Exotic species should not be used in operatonal practices, but can be used in research testing. Appendix D provides an overview of research on plant varieties.

· Starter and Early Successional Plant Species – Plant species used to initiate the establishment of plant communities should be suitable to site conditions and to early successional stages for each ecosite. Other species representing later stages of succession may invade reclaimed areas naturally over time.

· Cost-effective Implementation - Reclamation should be conducted to re-establish viable ecosystems in the most efficient and cost-effective manner possible. Reclamation should take advantage of direct placement of soil materials and other cost- effective opportunities as they become available.

2.2 PROCESS FOR ESTABLISHING VEGETATION (ECOSYSTEMS) TO MEET VARIOUS LAND USE OBJECTIVES

The Committee designed a seven step process to evaluate which ecosystems could be established to support terrestrial land use objectives on oil sands leases, and what techniques should be used to establish the preferred ecosystems. The term “ecosite” or “ecosite phase” will be used in place of ecosystem throughout the rest of this document. These terms are defined in the Glossary of Terms (Section 8) and in Appendix A. The seven steps are illustrated in Figure 2.2, and are described in the following text. The steps include:

1. Identify Target Ecosites that can be Established on Reclaimed Landscape.

2. Identify Techniques to Establish Ecosites on Reclaimed Landscape.

3. Identify Terrestrial Land Use Objectives for Reclaimed Landscape.

4. Identify Design Criteria for Ecosystems and Landscape Patterns to Achieve the Selected Terrestrial Land Use Objectives (i.e., commercial forest, wildlife habitat and other). Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 9 Prepared By: The Oils Sands Vegetation Reclamation Committee

Figure 2.2 Flow Diagram of Process for Establishing Vegetation (Ecosites) for Different Land Use Objectives

STEP 1 Identify Target Ecosites that can be Established on Reclaimed Landscape (Section 3.1)

STEP 2 Identify Techniques A to Establish Ecosites D On Reclaimed Landscape A (Section 3.2) P T STEP 3 Identify Terrestrial Land I Use Objectives for V Reclaimed Landscape E

Commercial Wildlife Other Forest Habitat Land Uses M A STEP 4 N STEP 4 STEP 4 Identify Design Criteria Identify Design Criteria Identify Design Criteria A for Commercial Forest for Wildlife Habitat for Other Land Uses G (Section 4.1); (Section 5.1.1); Identify Target Ecosites E Identify Target Ecosites Identify Target Ecosites and Landscape Patterns M and Landscape Patterns for and Landscape Patterns for Commercial Forest E Wildlife Habitat (Section 4.1 and 4.2) (Section 5.2 and Appendix J) N T STEP 5 Integrate Design Criteria for All Land-Use Objectives for Reclamation of Oil Sands Lease(s)

STEP 6 STEP 7 Design Monitoring Program Design Research to Verify Success of Meeting Commercial Forest Program (Section 7) Objectives (Section 6.1), Wildlife Habitat Objectives (Section 6.2), Biodiversity (Section 6.3) and/or Other Land Use Objectives Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 10 Prepared By: The Oils Sands Vegetation Reclamation Committee

5. Integrate Design Criteria for all Land Use Objectives for Reclamation of Oil Sands Leases. 6. Design Monitoring Programs to Verify that Land Use Objectives have been Successfully Met. 7. Recommend Research Programs to Address Information Gaps on Reclamation.

Step 1 - Identify Target Ecosites to be Established on Reclaimed Landscape

The first step was to identify which ecosites, of those that occur in the oil sands region, can be supported on reclaimed landscapes. The ecosites that naturally occur in the oil sands region were identified based on the ecological classification system for the Boreal Mixedwood produced by Beckingham and Archibald (1996) entitled “Field Guide to Ecosites of Northern Alberta”.

Of the ecosites that exist in the region, those that can be re-established on reclaimed landscapes were identified. The ecosites that can be re-established will depend on the landscape, drainage and soils that are re-established within the reclaimed landscapes, and hence on the available moisture and nutrient regimes. Therefore, to determine which ecosites can be established on reclaimed oil sands leases, two tasks were completed.

· Landscape features including slope, aspect, drainage pattern (moisture regime) and reclamation substrate, and the types of soils that can be established on oil sands leases were identified. Information was based on current mining and reclamation techniques that are being used on the Syncrude and Suncor oil sands leases. Several soil types can be reconstructed on tailings sand, overburden and consolidated/composite tailings (CT). For each soil type, the soil capability for forest ecosystems (based on Leskiw 1998), potential moisture regimes, potential nutrient regimes and salinity were identified. Three landscape features were classified: slope, aspect and reclamation substrate (overburden, tailings sand and CT), and

· Based on the known ecological requirements for the various ecosites (broad nutrient and moisture regimes) and for plant species in northeastern Alberta (Beckingham and Archibald 1996) and “Forest Sites Interpretation and Silivicultural Prescription Guide for Alberta” (Alberta Environmental Protection 1996a), the types of ecosites and plant species that could be supported by the reconstructed landscapes and soils were identified. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 11 Prepared By: The Oils Sands Vegetation Reclamation Committee

Step 2 - Identify Techniques to Establish Ecosites on Reclaimed Landscape

Using information from the extensive research and monitoring programs conducted by Suncor and Syncrude over the past several years, and the list of plant species associated with various ecosite phases (Beckingham and Archibald 1996), techniques were identified that could be used to reconstruct the landscape and soils, and to initiate “starter vegetation” that should eventually succeed to the targeted ecosites. Techniques outlined include: preferred landscape and soil reconstruction methods to establish Class 2 and 3 soil capabilities for forest ecosystems (these are based on presently used construction techniques), preferred revegetation methods such as planting densities, fertilizers, and preferred tree and shrub species. In addition, information has been provided on preferred sources of seeds and potential herbaceous and native species that may invade reclaimed sites.

Step 3 - Identify Terrestrial Land Use Objectives for the Reclaimed Landscape

The third step was to identify land use objectives in the oil sands region. The Oil Sands End Land-Use Committee has worked with key stakeholders in the region to define allowable land use options (Figure 2.3). The Oil Sands Vegetation Reclamation Committee chose to look at two acceptable objectives within a forest ecosystem: the primary land use of commercial forest for timber production, and the associated use of wildlife habitat. Other land uses that could be examined at a later date include recreation, traditional use and others as shown in Figure 2.3.

Each oil sands lease will be designed for several integrated land uses. The oil sands developer will negotiate what portion of their oil sands lease will be allocated to each land use based on the guidelines recommended by the Oil Sands End Land-Use Committee. The expectation is that each re-established ecosystem will support several land uses. For example, a commercial forest designed to produce timber will also provide habitat for wildlife and recreational opportunities.

Step 4 - Identify Design Criteria for Ecosystems and Landscape Patterns to Achieve the Selected Terrestrial Land Use Objectives

The fourth step in the process was to identify the design criteria required for the reclaimed landscape to support commercial forest and wildlife habitat. Design criteria included information on the preferred types of ecosites and preferred patterns of distribution of ecosites in the landscape. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 12 Prepared By: The Oils Sands Vegetation Reclamation Committee

Figure 2.3 End Land Use Options Recommended by Oil Sands End Land-Use Committee. Shaded Boxes Represent the Land Use Objectives Addressed in this Document. End Land Use Options for Lease Reclamation

Terrestrial Aquatic Ecosites Ecosites

Forestry Natural / Conservation Human Development Areas

Primary Use: Primary Use: Primary Use: · Reclamation research Commercial forest · Wildlife habitat · Biodiversity · Infrastructure for · Aesthetics transportation · Traditional aboriginal · Intensive food crop land uses production · General community · Grazing hunting, trapping, · Industrial development fishing, and gathering of sites plants · Intensive recreation

Associated Uses: Associated Uses: · Wildlife habitat · Wildlife sanctuaries · Extensive recreation and management areas · Biodiversity · Extensive recreation · Traditional land uses · Aesthetics · General community hunting, fishing, trapping and gathering of plants · Forest research sites · Orchards for tree seed production Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 13 Prepared By: The Oils Sands Vegetation Reclamation Committee

Commercial Forest: “The Forest Site Interpretation and Silvicultural Prescription Guide” (Alberta Environmental Protection 1996a), and the “Provincial Timber Harvesting Planning and Operations Ground Rules” (Alberta Environmental Protection 1994a), were used to identify important landscape and vegetation design criteria to be used for commercial forest. These criteria identify the forests that may be considered commercial within normal operating standards. Only forest stands of acceptable tree species that provide adequate volumes of timber at maturity (i.e., based on tree age, diameter at breast height, height and tree density) are considered merchantable. Additionally, only a subset of merchantable forests are accessible for the harvesting of timber and, hence are considered to be commercial. Merchantable stands that are located on steep slopes or within protective buffers around watercourses and wetlands, or are too small and not economical to retrieve (ground rule constraints), are considered non-commercial. The productivity (volume and growth) on the reclaimed landscape should meet or exceed predisturbance productivity levels.

Wildlife Habitat: Ecological (biophysical) factors that directly influence the suitability of habitat for wildlife include vegetation cover type, slope, aspect and other topographic features. Good habitat provides a combination of high quality food and cover for species of wildlife. Although all vegetation types provide some potential value or suitability for most species of wildlife, each group of species prefers certain ecological factors that better meet their requirements.

Habitat can be re-established based on either a coarse or fine filter approach. The objective of the coarse filter approach is to re-establish the same types of vegetation communities, in the same abundance and dispersion patterns, as existed prior to disturbance. This is pursued with the expectation that wildlife populations will return to the reclaimed sites in the same composition and abundance as existed prior to disturbance. This approach is meaningful and can be achieved by meeting biodiversity objectives as outlined in Sections 2.1 and 6.3. The objective of the fine filter approach is to reclaim an area to provide the specific biophysical habitat requirements of indicator wildlife species. This approach is outlined for ten wildlife indicator species in Appendix J.

Step 5 - Integrate Design Criteria for all Land Use Objectives for Reclamation of Oil Sands Leases

It is intended that lease holders will design their landscapes to achieve the required land uses using this document to assist with planning. Once the design criteria for each of the primary and associated land uses have been identified, the distribution pattern of the preferred ecosites for the primary land use over the development leases needs to be identified. Then the design criteria for the associated uses need to be integrated into the design criteria for the primary use. This is an important planning exercise that should be completed by an operator of an oil sands development during the design of the closure plan. This step is not discussed further in this manual. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 14 Prepared By: The Oils Sands Vegetation Reclamation Committee

Step 6 - Design Monitoring Program to Verify That Land Use Objectives Have Been Successfully Met

The sixth step in the process was to design monitoring programs to determine if the land use objectives have been achieved. For commercial forest and wildlife habitat, the benchmark to assess whether the reclamation approach has been successful has been identified and potential monitoring programs have been described.

For example, to ensure the objectives of establishing a commercial forest are being achieved, two monitoring programs are recommended:

· A program to confirm the establishment of tree seedlings, as benchmarked through compliance requirements under the “Timber Management Regulations” and the “Regeneration Survey Manual” (Alberta Environmental Protection 1994b), and

· A program to verify the establishment of forest productivity as benchmarked through measures of tree height growth, mean annual increment and site index, and its linkage to the measurement of soil capability.

Monitoring programs have also been presented for biodiversity. Specifically, the establishment of biodiversity is benchmarked through the variety of plant communities, plant species and diversity within species that become established, as compared to the pre-disturbance situation.

If results from the monitoring programs indicate that the land use objectives are not being achieved, the guidelines in this document for each of the steps in the process will be revised.

Step 7 – Recommended Research Programs to Address Information Gaps on Reclamation

The seventh step was to identify research needed to address information gaps identified in the reclamation information. In addition, research programs to pursue potential areas of improvement in reclamation techniques will be developed on an ongoing basis as monitoring data identifies these needs. New research and monitoring information will then be used to understand what additional ecosites can be established, and to improve the techniques used to establish these ecosites. This latter step follows the principle of Adaptive Management. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 15 Prepared By: The Oils Sands Vegetation Reclamation Committee

3. ECOSITE ESTABLISHMENT

The first step in the process was to identify which ecosystems of those that occur in the region, can be supported on reclaimed landscapes. The ecosystems that naturally occur in the oil sands region were identified based on the ecological classification system produced by Beckingham and Archibald (1996) for the Boreal Mixedwood. There are 12 ecosites (based on moisture and nutrient regimes), 25 ecosites phases (based on similar dominant plant species) and 73 plant community types (based on similar understorey species composition and abundance). The ecological classification system is detailed in Appendix A. Of those ecosites available in the region, the ecosites that could be supported on reclaimed landscapes, based on the re-established topography and soils were identified.

3.1 ECOSITES TO BE ESTABLISHED ON RECLAIMED LAND

The “Land Capability Classification for Forest Ecosystems” (Leskiw 1998) as it is applied to land reclamation has two main components: soils and landscape. Each component is evaluated separately, then the overall rating is determined by the most limiting of the two. The rating system has 5 classes, with Class 1 lands having the highest capability for forest ecosystems, and Class 5 lands having no potential for forest ecosystems. Climate also affects vegetation but it is assumed to remain unchanged so it does not affect the ratings before and after land disturbance.

In the soils component of the capability classification, the emphasis on soils is at a series level, and capability is closely related to productivity. The focus is on soil chemical and physical properties (reaction, salinity, nutrients, texture, structure, consistence, water retention, and moisture regime), and the resultant quality of the root zone.

A review of the ecosite classification of the boreal forest (Beckingham and Archibald 1996) reveals a close relationship among ecosite phase, site index and soil capability (Table 3.1). Table 3.2 provides a summary of the relationship among ecosite, moisture regime, site index, soil subgroup and soil capability for forest ecosystems across the boreal forest.

A synopsis of principal soils (soil series) as reclaimed to date on the Syncrude and Suncor oil sands leases is given in Table 3.3. These are also targeted soils for future reclamation along with corresponding targeted ecosite types. There will also be soils developed on consolidated/composite tailings (CT) in the future, but these soil types are not addressed in this document. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 16 Prepared By: The Oils Sands Vegetation Reclamation Committee

Table 3.1 Regional Mean Site Index by Species for each Ecosite for the Boreal Mixedwood, Northern Alberta

MEAN SITE INDEX AT 50 YEARS Ecosite Moisture Indicator White Jack Black Balsam Regime Species Aspen Spruce Pine Spruce Poplar a xeric 2 lichen 13.4 subxeric 3 b submesic 4 blueberry 15.8 17.5 14.3 c mesic 5 Labrador tea-mesic 14.3 11.5 d mesic 5 low-bush cranberry 18.2 16.8 15.2 15.7 17.3 e subhygric 6 dogwood 21.4 17.8 19.7 f subhygric 6 & horsetail 19.8 16.4 17.8 hygric 7 g subhygric 6 & Labrador tea-subhygric 11.7 9.9 hygric 7 h hygric 7 Labrador tea/horsetail 12.9 9.5 i subhydric 8 bog 9.8 j subhydric 8 poor fen 10.4 k subhydric 8 rich fen 7.2 l hydric marsh Not Applicable

Source: Beckingham and Archibald (1996). Also see Appendix B for soil and site index relationships found in other applicable studies. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 17 Prepared By: The Oils Sands Vegetation Reclamation Committee

Table 3.2 Summary of Ecosites and Related Site Productivity and Soil Capability for the Boreal Mixedwood, Northern Alberta

Soil Moisture Site Productivity Capabilit Ecosite Regime (Site Index At 50 Soil Subgroup y Years) Classa Lichen (a) Xeric 2 10-14 Orthic Brunisols 4 Lichen (a) Subxeric 3 10-14 Orthic Brunisols & Luvisols 4 Blueberry (b) Submesic 4 14-18 Orthic Luvisols & Brunisols 3 Labrador tea (c) low-bush cranberry (d) Mesic 5 14-18 Orthic Luvisols 2 Dogwood (e) Subhygric 6 18-22+ Gleyed Luvisols 1 Horsetail (f) Horsetail (f) Subhygric 6 & Hygric 7 14-18 Gleysols (aerated) 3 Labrador tea-subhygric (g) Labrador tea/horsetail (h) Labrador tea-subhygric (g) Hygric 7 10-14 Gleysols (reduced) 4 Labrador tea/horsetail (h) Bog (i) Subhydric 8 <10 Organic 5 poor fen (j) rich fen (k) Marsh (l) Hydric 9 Not Applicable

Source: Beckingham and Archibald 1996.

Note: Indices commonly ± 1 m. a Soil Capability Class (from Leskiw 1998):

Class 1 – High Capability (Index 81 to 100): Land having no significant limitations to supporting productive forestry, or only minor limitations that will be overcome with normal management practices.

Class 2 – Moderate Capability (Index 61 to 80): Land having limitations which in aggregate are moderately limiting for forest production. The limitations will reduce productivity or benefits, or increase inputs to the extent that the overall advantage to be gained from the use will be still attractive, but appreciably inferior to that expected on Class 1 land.

Class 3 – Low Capability (Index 41 to 60): Land having limitations which in aggregate are moderately severe for forest production. The limitations will reduce productivity or benefits, or increase inputs to the extent that the overall advantage to be gained from the use will be low.

Class 4 – Conditionally Productive (Index 21 to 40): Land having severe limitations; some of which may be surmountable through management, but which cannot be corrected with existing knowledge.

Class 5 – Non-Productive (Index 0 to 20): Land having limitations which appear so severe as to preclude any possibility of successful forest production. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 18 Prepared By: The Oils Sands Vegetation Reclamation Committee

Table 3.3 Reclaimed Soils, Soil Capability and Target Ecosites

Soil Reclaimed Soils Soil Drainage Moisture Nutrient Electrical Target ( Series (Depth ± 5 cm) Capability Class(c) Regime(d) Regime Conductivity Ecosite (e a) (dS/m) )

A 20 cm peat-mineral mix/ 2 W-MW mesic medium <1 D 30 cm clay/TSS(b) 1 I subhygric medium <1 d-e

B 50 cm clay/TSS 3 W-MW mesic low <1 c

2 I subhygric <1 g

C 50 cm sandy loam/TSS 3 R-W submesic low <1 a,b

D 20 cm peat-mineral mix/ 3 W mesic medium <1 d 30 cm sandy loam/TSS 2 I subhygric medium <1 d-e-f

E 20 cm peat-mineral mix/ 2 W-MW mesic medium 3 (2-4) d 30 cm clay/OB(b) 1 I subhygric medium 3 (2-4) d-e-f

F 50 cm clay/OB 3 W-MW mesic low 3 (2-4) c

2 I subhygric low 3 (2-4) g

G Overlay of 4 R-W subxeric low < 1 a > 50 cm TSS

H 20 cm peat-mineral 3 R-W submesic medium <1 b mix/TSS <1 2 I subhygric medium d-e-f

I 20 cm peat-mineral 2 W-MW mesic medium 3 (2-4) d mix/OB 1 I subhygric medium 3 (2-4) d-e-f

J 100 cm peat-mineral mix 2 W-MW mesic medium <1 d-e

4 VP subhydric medium <1 g,h,i,j,k

K 100 cm mineral soil 4 P hygric (r) low <1 g, h

5 VP subhydric low <1 l,j,k

L 20 cm peat-mineral mix/ 4 P hygric (r) medium <1 g, h 80 cm mineral soil 5 VP subhydric medium <1 l,j,k

(a) Soil series are designated with capital letters and these do not correspond to ecosite types which are shown in lower case letters. (b) Reclaimed soils substrate materials: TSS = tailings sand; OB = overburden, usually oily; pH levels of all reclaimed materials about 7.5. (c) Drainage Class: R-rapid; W-well; MW-moderately well; I-imperfect; P-Poor; VP-very poor. (d) This is the expected upland moisture regime. Lower slopes and level areas may become wetter. Also, on slopes >20%, north aspects (270° - 135°) will be one moisture regime class wetter, and south aspects (135° - 270°) will be one moisture regime class drier. See Section 8 – Glossary of Terms for definitions of moisture classes. (e) Names of ecosites are listed in Table 3.2. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 19 Prepared By: The Oils Sands Vegetation Reclamation Committee

The resultant soil and landscape capability classes are used for comparison with baseline conditions. Appendix B provides descriptions of natural baseline soils in the region and corresponding site indices for principal tree species. To augment the “equivalent capability approach”, it is recommended that forest productivity be confirmed on the principal soils, both before and after reclamation. This can be accomplished by determining site index of principal tree species, in suitable stands (considering age, density and health), at the same location where soil descriptions are made on reclaimed areas.

The assumption made in predicting ecosites is that much of the “upland” landscape will be rapid to moderately well-drained, corresponding to submesic to mesic moisture regimes. Areas with a peat-mineral mix application will tend to have a medium nutrient regime. The reclaimed soils in each ecosite will differ from natural soils in the matching natural ecosite. Reclaimed soils will tend to have higher pH levels (6.5 to 8.0), and there will likely be higher salinities ranging from electrical conductivity (EC) of 0.5 to 4.0 dS/m in the root zone. Nevertheless, ecosites should develop with productivity levels similar to those in natural ecosites. The above soils can be established on a variety of landscapes, including dyke slopes and level to undulating areas.

The target ecosites that may establish on each of the principal soils (soil series) that have been reconstructed by Syncrude and Suncor are listed on Table 3.3. The ecosites range from those that will establish on low nutrient, drier sites (subxeric, submesic) such as the lichen (a) and blueberry (b) ecosites to those that will establish on wetter (mesic, subhygric), higher nutrient sites such as low-bush cranberry (d), dogwood (e) and horsetail (f).

In time, lower areas, especially on overburden landscapes, may become imperfectly drained (subhygric or hygric). Depending on the degree of wetness, tree growth may be enhanced or retarded. The enhanced growth areas should be suitable for e-type (dogwood) ecosites, whereas wetter areas are destined for g (Labrador tea-subhygric), h (Labrador tea/horsetail) and I (bog) ecosites.

3.2 ESTABLISHMENT OF ECOSITES ON RECLAIMED LANDSCAPES

Reclamation techniques used currently and historically by Suncor and Syncrude are outlined in Appendix F. This section presents a summary of current starter reclamation practices and methods to establish landscape and soils for ecosite development.

3.2.1 Reclamation Goals

The goal of reclamation is to achieve maintenance-free, self-sustaining ecosystems with capabilities equivalent to or better than pre-disturbance conditions. Maintenance-free reclamation means that human maintenance activities are not required, except for circumstances where future human activities lead to re-disturbance of areas. This does not imply a changeless state, as landforms will experience gradual reshaping of the landscape through normal geologic processes typical of the region and vegetation will evolve through various seral stages to more mature ecosystems over time. Self-sustaining ecosystems, typical of those in the region, will evolve on revegetated terrains, from new plantings toward mature systems typical of those in the region, with little management input from man following the initial plant establishment. Equivalent land capability refers to the capability of post-reclamation land to Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 20 Prepared By: The Oils Sands Vegetation Reclamation Committee support various land uses similar to those that existed prior to an activity being conducted on the land; however, individual predisturbance and post-reclamation land uses may not be identical.

The vision for reclamation includes the construction of stable landforms and re-establishment of productive, self-sustaining ecosystems which will provide land use capabilities equivalent to those of the pre-mining environment. The following general operational and reclamation criteria form the basis for reclamation program design:

· Structures will be geotechnically stable,

· Discharge of earth materials through surface erosional processes will be controlled to rates which are acceptable to the environment,

· Discharge of surface and seepage waters will be managed to ensure an acceptable level of impact on watercourses, and

· The ecosystems re-established on disturbed lands will be self-sustaining and will mature naturally without presenting significant risk to plants, or resident and migratory wildlife species.

3.2.2 Landscape and Soil Construction

Landscape evaluation focusses on general tree growth (aspen, white spruce, jack pine) as determined primarily by steepness of slope and modified by effects of slope position and aspect, stoniness, and actual erosion. As a guide, slopes for each landscape capability class are: Class 1 is <30%, Class 2 is 31-45%, Class 3 is 46-60%, Class 4 is 61-75%, and Class 5 is >75%. Position effects are considered to impact a rating by one-half a class: moisture shedding positions are downgraded, moisture receiving positions are upgraded, level areas remain unchanged, and depressional areas are downgraded. South-easterly to westerly exposures (Figure 3.2) on slopes >20% are also downgraded by about one-half class. Northerly aspects have cooler temperatures offset by improved moisture balance, resulting in no change to the class. Stoniness reduces rooting volume and can downgrade landscapes from Class 1 to Class 5 as volume of stones increases from <20% to >80%, respectively. Erosion (visible rills and gullies) can downgrade the landscape as much as two classes (for example, Class 2 soil on a 35% slope downgraded to Class 4 if erosion is extreme). Refer to the “Land Capability Classification for Forest Ecosystems in the Oil Sands Region” (Leskiw 1998) to rate specific landscape conditions.

3.2.2.1 Terrain Development

Existing landforms are altered as a result of mining. Flat and depressional areas will be replaced by more upland landforms with slopes up to 3:1 (horizontal:vertical). These new areas will be better drained than the pre-existing landforms. Tailings dyke structures are designed to provide secure impoundment of tailings (fines in suspension, tailings water and consolidated/composite tailings (CT)). Tailings materials will be reclaimed in two ways. Fluid or mature fine tails will be placed below a cap of clean water, while CT is expected to develop into a dryland/wetland system and be revegetated. Overburden dumps are created to store overburden materials comprised of recent fluvial deposits, a variety of glacial deposits, and bedrock formations Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 21 Prepared By: The Oils Sands Vegetation Reclamation Committee including hydrocarbon enriched materials. The topography of overburden dumps ranges from level to steeply sloping. Slope stability of overburden materials and chemical composition (salinity and sodicity) are important considerations in final design and resultant soil and landscape capability.

Peat-mineral mix materials may also be stockpiled, temporarily and possibly permanently. The intent is to salvage peat deposits (organic soils) for future reclamation purposes. Findings to date suggest that pure peat stockpiles may be droughty in mesic or drier moisture regimes, and they may develop permafrost in wetter moisture regimes. Peat-mineral mixes (4:1 peat: mineral ratio or less peat) appear to behave as a mineral soil without drought or frost problems. Therefore, in stockpiling peat, it is recommended that peat be overstripped to incorporate at least 20% (by volume) mineral material, as is currently practiced for placement as a surface soil.

3.2.2.2 Terrain Stability Objectives

Dyke and dump structures are created with the primary objective of geotechnical stability. This does not preclude making alterations to the final design of these structures while maintaining the stability objective. Long-term post-reclamation landform stability (of all retention structures) is evaluated through ongoing stability monitoring and analyses. These structures have been designed and are operated to accepted Canadian standards. Their design, construction and performance is supported by extensive monitoring programs reviewed by independent review boards and regulatory agencies. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 22 Prepared By: The Oils Sands Vegetation Reclamation Committee

Figure 3.1 The Effects of Aspect on the Moisture Regime

North

Wetter

West East

Drier

South Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 23 Prepared By: The Oils Sands Vegetation Reclamation Committee

3.2.3 Reclamation Approach

Soil capability is a principal determinant for many terrestrial environmental characteristics. Both vegetation and wildlife habitat, as well as their sustainability and biodiversity are dependent on soil capability. The reclamation objective is to provide equivalent (or better) soil capability (with consequent and equivalent plant and animal ecosystems) as the mine is developed and reclaimed. Annual detailing of soil salvage and reclamation area establishment allows quantification of losses/gains in capability. This is accomplished through use of the land capability classification system for the oil sands area (Leskiw 1998).

The history of reclamation success through vegetation establishment is provided in Appendix H.

3.2.3.1 Soil Reconstruction

Reconstruction of soil for reclamation areas is a critical component of a reclamation plan, since the ultimate capability of the reclaimed area is determined largely by the quality of reconstructed soil. Surfaces of reconstructed landforms are covered with a layer of soil amendment, primarily a peat-mineral mix that has been salvaged from areas to be mined, and then either stockpiled or (preferably) transported directly to reclamation areas (i.e., direct placement). Stockpiling is employed where surface disturbance has just begun on a site and where there are no areas available for reclamation.

Soils are reconstructed using either “one-lift” or “two-lift” soil replacement. In one lift reclamation, sites are enhanced with quality soil-building material, using a technique which involves overstripping muskeg (peat) to include 25 to 50% (by volume) of mineral overburden (usually 1 m of peat and 0.4 m of mineral overburden). This material becomes the cover soil described as peat-mineral mix amendment. It is hauled, placed on prepared subsoil and then spread to an average depth of 15 to 50 cm over the underlying materials. The subsoil base materials are either tailings sand, suitable overburden or possibly (in the future) CT. Where peat is mixed with fine-textured till, clay or silt (fines), the mixture is primarily used as an amendment for tailings sand areas. Where peat is mixed with coarse textured material (sand and gravel), the mixture is primarily used to amend overburden dumps or dykes.

In two lift operations, about 50 cm of either sandy or clayey suitable subsoil material is placed over tailings sand, suitable overburden, or possibly (in the future) CT. The mineral soils are carefully selected with respect to texture and salinity/sodicity to ensure they result in a suitable root zone medium. In the final lift, 15 to 25 cm of cover soil (peat-mineral mix amendment) is placed on top of the subsoil layer.

Modifications of the above procedures occur. In the event that peat-mineral mix material is unavailable as a cover soil amendment, 50 to 70 cm of sandy or clayey soil material that is slightly enriched with organic matter may be placed over tailings sand or suitable overburden. If Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 24 Prepared By: The Oils Sands Vegetation Reclamation Committee there is a surplus of peat-mineral mix material, a 1 m layer can be placed over tailings sand, overburden or CT, resulting in a high soil capability rating.

Overburden dumps created from overburden materials consisting of the “Clearwater Formation” and the “McMurray Formation” require additional thickness of cover soil due to their high sodicity and salinity. Depending on the percentage of these types of materials in the dumps, the cover soil cap requires an additional 50 cm of non-sodic overburden, overlain by 50 cm of peat-mineral mix for a total depth of 1 m.

All of the above reclaimed soils normally have a mesic moisture regime. Drier conditions occur on dyke crests and steep south slopes. Wetter conditions commonly develop at the toe of slopes and in depressions within overburden dump areas.

This system of soil salvage has been integrated with the land capability classification (Leskiw 1998) to ensure that the desired land capabilities are achieved. Therefore, peat-mineral mix amendment will be salvaged and directed to the reclamation sites with forest capability development as the primary consideration. This change in focus is not expected to drastically alter soil salvage criteria, but will assist in managing the placement of the better-suited reclamation amendments. As a result, peat-mineral mix soil amendments having a granular subsoil will be directed to sites with a higher proportion of fines in the pre-reclamation soil mix. Peat-mineral mix soil amendment with a higher fines ratio will be directed to areas with a tailings sand or CT substrate. This strategy will result in better water infiltration for the overburden reclamation sites and improved water holding capability for the reclaimed soils on tailings sand and CT.

3.2.3.2 Reclaimed Soils

Several reclamation profiles are illustrated in Figures 3.2 and 3.3 based on materials available and conditions encountered during an inventory of reclaimed soils at Syncrude and Suncor. These figures indicate some common reclamation options. A complete rating of specific soils is required to confirm the ratings and to ensure all soil factors are evaluated. Important assumptions made in rating these selected profiles are that the pH of all materials is 7.5 and there are no salinity or physical limitations in topsoils and subsoils except that overburden materials are saline with electrical conductivity levels of 3.5 to 4.5 ds/m. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 25 Prepared By: The Oils Sands Vegetation Reclamation Committee

Figure 3.2 Soil Handling Options and Related Soil Capability Classification for Overburden Reclamation OVERBURDEN

ONE-LAYER TWO-LAYER

Peat-Mineral Mix Peat-Mineral Mix Organic Enriched over Sandy Loam Mineral Material or Finer Subsoil

Class 2 Class 3 Class 2 Class 3 Class 2 Class 3

>20 cm of >10 cm of > 20 cm of peat-mineral peat-mineral > 50 cm of > 20 cm of >50 cm of sandy loam or mix over >30 mix over >10 peat-mineral peat-mineral loam or finer finer material cm of sandy cm of sandy mix over mix over material over over loam or finer loam or finer overburden overburden overburben overburden material over material over overburden overburden

Assumptions: All materials have a pH of 7.5. EC levels are <2 dS/m in all materials except for the overburden material, which has an EC of 3.5 to 4.5 dS/m. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 26 Prepared By: The Oils Sands Vegetation Reclamation Committee

Figure 3.3 Soil Handling Options and Related Soil Capability Classification for Tailings Sand Reclamation TAILINGS SAND

ONE-LAYER TWO-LAYER

Organic Enriched Peat-Mineral Mix Peat-Mineral Mix Peat-Mineral Mix Mineral Material over Sandy Loam over Loamy Sand or Finer Subsoil Subsoil

Class 2 Class 3 Class 2 Class 3 Class 2 Class 3 Class 2 Class 3

20 cm of peat-mineral 10 cm of 30 cm of 10 cm of > 20 cm of mix over 20 peat-mineral peat-mineral peat-mineral > 50 cm of > 20 cm of >50 cm of sandy loam or cm of sandy mix over 10 mix over 20 mix over 10 peat-mineral peat-mineral loam or finer finer material loam or finer cm of sandy cm of loamy cm of loamy mix over mix over material over over tailings subsoil over loam or finer sand subsoil sand subsoil tailings sand tailings sand tailings sand sand tailings sand subsoil over over tailings over tailings or 10 cm over tailings sand sand sand 30 cm

Assumptions: All materials have a pH of 7.5, and an EC of <2 dS/m. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 27 Prepared By: The Oils Sands Vegetation Reclamation Committee

The most important soil properties considered are depths of materials, along with their respective texture, structure/consistence, salinity and organic matter content. The loamy and clayey textures are superior in retaining water. However, if compaction occurs during placement, these finer textured soils are limiting to root growth. Any direct placement material or overburden should not be saline or sodic. If such limitations occur, the soils are downgraded at least one class. Organic matter in the upper profile is fundamental to vegetation development and should be given special attention in materials handling.

It is important to note that the profiles illustrated in Figures 3.2 and 3.3 are not prescriptions or endorsements for constructing all classes of soils. The intention is: to select the desirable level(s) of reclaimed soil capability, examine alternatives to obtain that level, and choose the best option based on source materials hauling distances and timing. The Class 2 soils in Figures 3.2 and 3.3 respectively, represent the best treatments. The Class 3 soils are created using materials of lower organic matter content, or lower water holding capacity, or by placing shallower depths of better quality materials. In locations where a subhygric moisture regime develops, Class 2 soils are upgraded one class in rating to Class 1 because of improved water availability for plant growth.

3.2.3.3 Vegetation Establishment

Revegetation Objectives

The primary objectives of revegetation programs are to:

· Provide an erosion-resistant plant cover on tailings dyke slopes and overburden dump slopes,

· Focus on utilization of native woody-stemmed reclamation species common to the region,

· Strive to establish a diverse range of plant species to re-create the level of biodiversity common to the pre-disturbed site, and

· Establish a viable plant community capable of developing into a self-sustaining cover of species suitable for commercial forest, wildlife habitat, traditional land uses, and with possibilities for recreation and other end uses. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 28 Prepared By: The Oil Sands Vegetation Reclamation Committee

Current Revegetation Practices

The revegetation of reclaimed landform surfaces is determined by the nature and type of landform structures (dykes, overburden dumps, tailings sand basins, CT deposits), slope, aspect, soil type (capability class) and soil moisture regime. The types of vegetation communities that will successfully establish and develop under various combinations of these factors have been the subject of Suncor and Syncrude research programs for more than 20 years. Typically, the revegetation process begins with excavation and hauling of undisturbed peat- mineral mix soils to the reclamation area. This method (which is completed in the winter whenever possible) enhances site revegetation because dormant, insitu native seeds and root fragments are transferred with the soil. Spreading of the amendment on the reclamation site is completed in early spring with the usual result being the emergence of a variety of native, woody-stemmed species, forbs and grasses beginning in the first growing season. Tree Planting Prescription Establishment of woody plants on reclamation areas is integral to the reclamation process. Selection of species and the proportion of each species in the supplemental planting mix are based on the woody-stemmed species common to the ecosites within the region, existing field conditions, the vegetation type expected to develop on the site (based on landscape terrain features), and the expected growth of woody-stemmed species from seeds and root fragments in the soil amendment layer. The species composition is designed to accelerate the process of natural succession towards desired vegetation types (ecosites). The micro-environments in the area change as woody cover develops on a reclamation area, providing favorable conditions for later successional species. The planting program is designed to ensure that plant species that are capable of taking advantage of these changes in condition are present. Therefore, four to six species are typically planted to supplement the natural processes of woody plant establishment. Table 3.4 outlines the starter woody stemmed planting prescriptions to establish each of the ecosite phases. Fertilizer Application

Fertilizer is applied to both the reclaimed overburden and tailings sand areas following soil placement. Syncrude does not apply maintenance fertilizer to reforested areas, whereas Suncor applies fertilizer annually for two to four years after planting. Maintenance fertilizer rates are determined from criteria such as soil tests, cover performance and cover objectives. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 29 Prepared By: The Oil Sands Vegetation Reclamation Committee

Table 3.4 Planting Prescription by Ecosite Phase

Tree Speciesa Shrub Speciesa Landscape Soil Capability And Target (Total Density of 1800 - (Total Density of 500 - 700 Features Moisture Regime Ecosite Phase 2200 Stems / ha) Stems /ha) Tailings Sand, Soil Class 4, a1 Lichen, jack pine Jack Pine Blueberry, Bearberry, Crests Xeric, Subxeric Green Alder Tailings Sand Soil Class 4-3, Subxeric, b1 Blueberry, jack pine- Jack Pine Blueberry, Bearberry, Slope, Submesic aspen Aspen Labrador tea, South Aspect White Spruce Green Alder Tailings Sand Soil Class 3-2, Subxeric, b2 Blueberry, aspen Aspen Blueberry, Bearberry, Slope, Submesic (white birch) White Birch Labrador tea, North Aspect White Spruce Green Alder b3 Blueberry, aspen- Aspen Blueberry, Bearberry, white spruce White Spruce Labrador tea, Green Alder White Birch b4 Blueberry, white White Spruce Blueberry, Bearberry, spruce-jack pine Jack Pine Labrador tea, Green Alder Overburden, Soil Class 3, c1 Labrador tea Jack Pine Labrador tea, Green Alder, Low Organic Mesic, Submesic (mesic), jack pine-black Black Spruce Bog cranberry, Blueberry spruce Overburden, Soil Class 3-2, d1 Low-bush Aspen Low-bush Cranberry, South Aspect Mesic cranberry, aspen White Spruce Canada Buffalo-berry, Balsam Poplar Saskatoon, Green Alder, White Birch Rose, Raspberry Overburden, Soil Class 3-2, d2 Low-bush Aspen Low-bush Cranberry, North Aspect Mesic cranberry, aspen-white White Spruce Canada Buffalo-berry, spruce Balsam Poplar Saskatoon, Green Alder, White Birch Rose, Raspberry Overburden, Soil Class 3-2, Mesic, d3 Low-bush White Spruce Low-bush Cranberry, North Aspect Subhygric cranberry, white Aspen Canada Buffalo-berry, spruce Balsam Poplar Saskatoon, Green Alder, White Birch Rose, Raspberry Near Level Soil Class 3-2, Subhygric, e1 Dogwood, balsam- Aspen Dogwood, Low-bush Overburden or Mesic aspen Balsam Poplar Cranberry, Raspberry, Tailings Sand White Spruce Green Alder, Rose White Birch Near Level Soil Class 3-2-1, e2 Dogwood, balsam- White Spruce Dogwood, Low-bush Overburden or Subhygric, Mesic white spruce Aspen Cranberry, Raspberry, Tailings Sand Balsam Poplar Green Alder, Rose White Birch e3 Dogwood, white White Spruce Dogwood, Low-bush spruce Aspen Cranberry, Raspberry, Balsam Poplar Green Alder, Rose White Birch Near Level Soil Class 2-1, Subhygric f1 Horsetail, balsam- Balsam Poplar Rose, Green Alder, Dogwood, Overburden or aspen Aspen Raspberry, Low-bush Tailings Sand, Birch Cranberry Lower Slope White Spruce Position f2 Horsetail, balsam- White Spruce Rose, Dogwood, Low-bush white spruce Aspen Cranberry Balsam Poplar Birch f3 Horsetail white White Spruce Rose, Low-bush Cranberry spruce a In general, species are listed in order of dominance to be planted in the target ecosite phase. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 30 Prepared By: The Oil Sands Vegetation Reclamation Committee

Seeding and Seed Source Considerations

A focus of the reclamation program is to encourage the invasion by native vegetation and establish woody seedlings. Selection of appropriate seed mixtures, application methods and application rates can impact plant species diversity and seedling survival on the reclamation areas. To enhance seedling survival and native species invasion, a barley nurse crop is used on the reclamation sites. Sites are not seeded with grasses, as grasses compete with tree seedlings for nutrients and moisture.

The seed source or tree seed variety used in seeding or planting programs must be registered with the Alberta Tree Improvement and Seed Centre and follow seed provenance rules, as the seed source or variety must be specified on the Silviculture Records Management System (SRMS) or an equivalent system for long-term documentation.

The use of plant varieties with local seed provenance has two major advantages:

· Reduces the risk of maladaptation in the regenerated forest, and

· Maintains genetic integrity and evolutionary potential of local forest tree populations.

See Appendix I for further discussion of seed sourcing and seed zones.

Plant Variety Use Consideration

Specifically developed, locally adapted genetic varieties of plants may provide some advantage in land reclamation depending on traits selected. These traits may include superior growth and survival, climatic and pest hardiness, soil salinity tolerance, and others. Deployment of selected varieties requires further consideration of the impact on genetic diversity and ecosystem adaptations.

Exotic tree species, non-native hybrid varieties and clonal materials are not to be deployed in operational practice (see Appendix D), but research testing using these species can be conducted to obtain scientific information and experience. Research testing should be defined and research trials catalogued. Certain types of research trials may require referral, review and approval by Alberta Environmental Protection, Land and Forest Service. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 31 Prepared By: The Oil Sands Vegetation Reclamation Committee

3.3 SUMMARY OF UNCERTAINTIES AND DATA GAPS IN ECOSITE RECLAMATION AND FUTURE RESEARCH

There are several areas of reclamation that still need to be researched. These include:

· Vegetation productivity (site index) on reclaimed soils needs to be measured and compared to vegetation productivity on natural soils,

· The relationship between soil capability classes and vegetation productivity (site index) needs to be identified through monitoring programs,

· It is uncertain whether all plant species recommended as starter species can be propagated on reconstructed soils (Table 3.4). Future propagation of additional plant species, such as those in Appendix H, needs to be studied,

· The survival and vitality of plant species moved through direct placement from various ecosystems to reclamation sites supporting different subsoils needs to be examined,

· The potential effects of elevated pH levels in reconstructed soils on plant growth are not known and need to be studied,

· The feasibility of using mineral soil through direct placement to develop upland ecosystems needs to be examined,

· The effectiveness of covering and reclaiming saline/sodic materials needs to be studied,

· The feasibility of continuing the use of exotic plant species in reclamation needs to be examined,

· Methods to measure biodiversity need to be developed,

· The feasibility of re-establishing ecosystems on CT needs to be researched,

· The feasibility of applying sulphur to reduce soil pH and sodicity needs to be examined,

· Methods to enhance the establishment of native understorey species to achieve greater biodiversity than is possible through seeding/planting need to be developed,

· The biology and productivity of reclaimed soils should be examined. Mycorrhizae, nutrient cycling and sustainability of peat-mineral mix amendments to ensure that the Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 32 Prepared By: The Oil Sands Vegetation Reclamation Committee

“living” components of the soil system are functioning effectively and in balance should be the focus of the research,

· Seed sources for species that provide good timber/fibre production need to be improved,

· Essential ecosystem functions and plant species required to accomplish these functions need to be determined, and

· The ability to create ecosites d (low-bush cranberry) and e (dogwood) without adding clay is uncertain. The sustainability of the ecosites needs to be studied. Further research requirements will be identified based on results of the recommended monitoring programs discussed in Section 6.0. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 33 Prepared By: The Oil Sands Vegetation Reclamation Committee

4. DESIGN CRITERIA FOR COMMERCIAL FORESTS

4.1 DEFINITION OF, AND CONDITIONS TO SUPPORT COMMERCIAL FOREST

In this document commercial forest is defined as:

· Land with a soil capability class of 1 to 3 which can support sustainable forest growth. Forests can be grown on class 4 soils, however these are not considered commercial,

· Lands stocked with native tree species which may include white spruce, black spruce, jack pine, aspen poplar, balsam poplar, white birch and/or larch, and

· Forest stands not limited by operating restrictions such as stream buffers, potential recreation lake buffers, stand size, arrangement or accessibility. These operating restrictions are discussed in detail in the following section.

In the context of reclamation, a commercial forest also includes established seedlings that can be reasonably expected to become an operable stand of acceptable trees. Operability is controlled by the following factors: · Location must not be within a watercourse or wildlife protection buffer, · Maximum permissible slopes are 45% (20% on tailings sand) and preferred slopes £ 30%, · The minimum stand size must be greater than 4 ha, and · Volume at maturity must provide greater than 50 m3 per ha of acceptable trees (i.e., trees that provide adequate volumes of timber at maturity).

Acceptable trees are defined to include:

· Species as defined in the “Alberta Regeneration Survey Manual” (Alberta Environmental Protection 1994b):

- established seedlings for an area harvested primarily for coniferous timber, and acceptable established seedlings for an area harvested primarily for deciduous timber,

· Minimum size at maturity as defined in the “Timber Harvest Planning and Operating Ground Rules” (Alberta Environmental Protection 1994a), is 15 cm diameter at the butt and with a useable length of 3.66 m to a 10 cm top diameter, and

· Suitable quality meaning less than 50% cull as defined in the Provincial Scaling Regulation. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 34 Prepared By: The Oil Sands Vegetation Reclamation Committee

4.1.1 Operating Restriction Guidelines

A number of characteristics limit the operability of a forest stand. Although these limitations do not preclude the development of forest stands, the resulting forest should be designated non- commercial for timber production. For complete design guidelines, refer to the “Timber Harvest Planning and Operating Ground Rules” (Alberta Environmental Protection 1994a), that are updated periodically.

4.1.1.1 Terrain Considerations

Slopes up to 45% can support commercial forest, although harvesting is normally conducted on slopes of 30% or less. Slopes over 30% and more than 50 m long are considered severe logging areas (Alberta Environmental Protection 1996a) and therefore reclaimed landscape on slopes between 30 and 45% that will be considered for commercial forest will be limited to an area equivalent to what existed prior to disturbance.

Furthermore, because of the erosive nature of the tailings sands, slopes up to only 20% are considered suitable for commercial forest. On overburden, slopes greater than 20% for distances greater than 200 m require special design considerations to ensure access and log decking capability. This includes requirements for a 40 m wide area of less than 5% cross slope and less than 8% adverse grade or 12% favorable grade for road and deck placement.

In addition to the terrain considerations outlined above, reclamation planning must also consider stream and watershed protection requirements, access limitations and harvesting constraints. Vegetated watercourse buffers are required on all streams and lakes. The characteristics of the vegetated buffers are based on the size and uses of the watershed. These buffers are not considered part of the commercial forest in the reclaimed landscape. Current watershed guidelines are provided in Appendix I. When slopes greater than 20% are adjacent to, or overlap with the vegetated buffer, the slopes becomes inoperable and therefore are not considered as part of the commercial forest land base.

Landscape capability classes 1, 2 and 3 are suitable for commercial forestry, provided slopes comply with regulatory requirements. To replace an equivalent commercial forest land base, landscape capability classes (on an area basis) must be maintained at pre-disturbance levels. However, improvements in capability classes are acceptable.

4.1.1.2 Stand Characteristics

The minimum stand size for operability is 4 ha. Stands must be at least 100 m wide, with a preferred width of 400 m.

Seventy percent of the spruce stands should be over 10 ha, with a minimum average size of 16 ha. Eighty percent of the pine and aspen stands should be over 20 ha, with a minimum average size of 40 ha. These characteristics are similar to those of natural occurring spruce, pine and aspen stands (Table 4.1). Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 35 Prepared By: The Oil Sands Vegetation Reclamation Committee

Table 4.1 Stand Characteristics

Tree Stand Width (Meters) Stand Sizes (Hectares)

Species Minimum Natural State Minimum

Minimum Preferred % Greater than Average

White Spruce 100 400 4 70% 10 16

Jack Pine 100 400 4 80% 20 40

Aspen 100 400 4 80% 20 40

For calculation of operability standards, stands that are adjacent and have common merchantability characteristics may be considered as one unit.

Natural forests have irregular shapes. However, to be able to harvest a finger, the skidder must be able to turn the trees which requires a width of 40 m. Narrower extensions of forests into non-commercial stands are currently not operable.

4.2 ECOSITES THAT MEET THE DEFINITION OF A COMMERCIAL FOREST

The target reclaimed ecosites considered to be commercially suitable in the oil sands region are: b (blueberry), c (Labrador tea-mesic), d (low-bush cranberry), e (dogwood) and f (horsetail). Ecosite a (lichen) is too dry, and ecosites g (Labrador tea-subhydric) to l (marsh) are too wet for commercial forest purposes. In natural areas, ecosites b to f are considered to have commercial potential.

4.3 METHODS TO MAINTAIN GUIDING PRINCIPLES

The maintenance of the guiding principles (Section 2.1) will not be difficult in reclamation areas that are designated to provide commercial forest.

The prescriptions should include a range of stand types and sizes characteristic to the region. This is required to adequately address the site requirements of the commercial tree species. Black spruce should be prescribed surrounding and in wet imperfectly drained areas, white spruce on lower slope and hygric sites, mixedwood on mid-slope moisture regimes and jack pine stands on xeric sites. In the mixedwoods, the conifer component should trend from white spruce to jack pine as the site trends to drier moisture regimes.

The establishment of these prescriptions must be based on the ecosite classification for northern Alberta (the ecosite principle). During establishment, the planting of herbaceous and shrub layer species, similar to those found in natural ecosites, will promote stand biodiversity. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 36 Prepared By: The Oil Sands Vegetation Reclamation Committee

The sustainability and productivity of the commercial forests is a function of the forest and soils development and are expected to be met by natural processes. Should results of the forest and soil productivity monitoring program (Section 6.1.2) show declining productivity, the causes will need to be investigated. Remediation methods that will be implemented to improve productivity will depend on the capability levels and the current technology.

The commercial forest monitoring system (Section 6.1) will provide ongoing information for periodic revision of this manual and the practices being used to establish commercial forests on reclaimed oil sands leases. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 37 Prepared By: The Oil Sands Vegetation Reclamation Committee

5. DESIGN CRITERIA FOR ASSOCIATED LAND USE - WILDLIFE HABITAT

Ecological factors that directly influence the suitability of terrestrial habitat for wildlife include vegetation types and landscape factors, such as topography, slope, aspect, and the size and interspersion patterns of ecosite phases/plant communities.

Good habitat provides a combination of high quality food and cover for species of wildlife. Landscape diversity and ecosite phase or plant community diversity are fundamental factors required to support a sustainable forest and wildlife community. Factors of importance to an animal’s survival may be specific and narrow, such as a reliance on a particular type of plant for food, or the presence of minimum diameter escape trees. However, for most wildlife species, habitat suitability is governed by broader biological and physical characteristics such as shrub or tree canopy cover, slope and topographic characteristics. Although food and cover requirements can be provided to a wildlife species by one vegetation community, many animals are dependent on a variety of communities to meet their annual or seasonal life history requirements. Therefore, the interspersion pattern of different community types is an important component of habitat quality for a variety of wildlife species.

Habitat can be re-established based on either a coarse or fine filter approach. The objective of the coarse filter approach is to re-establish the same types of vegetation communities in the same abundance and dispersion patterns, as existed prior to disturbance. This is pursued with the expectation that wildlife populations will return to the reclaimed sites in the same composition and abundance as existed prior to disturbance. This approach is meaningful and can be achieved by meeting biodiversity objectives as outlined in Sections 2.1 and 6.3. Therefore, this approach is not discussed further in this section.

The objective of the fine filter approach is to reclaim an area to provide the specific biophysical habitat requirements of indicator wildlife species. Although all vegetation types provide some value or suitability for several species of wildlife, each group of species prefers certain ecological factors that better meet their requirements. Therefore, the groups of wildlife species for which habitat establishment is to be targeted need to be identified before lands being reclaimed can be designed to support wildlife species. Then, the ecological factors that provide the most suitable habitat for the targeted wildlife can be documented in the design criteria.

The fine filter approach to preparing design criteria for wildlife habitat to be established on reclaimed landscapes, was broken into four steps:

· Identify the target wildlife species (Section 5.1.1),

· Identify the habitat requirements of these representative target wildlife species (Appendix J),

· Identify the ecosites and landscape components that meet the habitat needs of the targeted wildlife species (Section 5.2 and Appendix J), and Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 38 Prepared By: The Oil Sands Vegetation Reclamation Committee

· Identify the methods that could be used to reclaim disturbed areas to establish habitat for these species (Section 5.3).

5.1 DEFINITION OF AND CONDITIONS TO SUPPORT WILDLIFE HABITAT

5.1.1 Target Wildlife Species and Habitat Requirements

Over 41 species of mammals, 188 species of birds, 4 species of amphibians and 1 species of reptile potentially inhabit the oil sands region, either on a seasonal or year-round basis. A summary of wildlife populations that occur in the oil sands region is presented in Appendix J. Several wildlife species that represent niche associations (species that have similar habitat requirements) were selected in order to define design criteria for reclaimed landscapes for wildlife. They were selected based on ecological importance, niche representativeness and abundance, and resource use value (Appendix J). Generally these target wildlife species are of ecological or socio-economic importance, should be herbivores or omnivores rather than carnivores (as the abundance and distribution of most carnivores are influenced to a greater degree by prey availability than by habitat parameters), should have relatively well understood habitat requirements, and should adequately represent the habitat requirements of a variety of other species.

The representative target species for terrestrial habitats are: · Moose, · Black Bear, · Snowshoe Hare, · Red Squirrel, · Ruffed Grouse, · Fisher, · Great Gray Owl, · Passerines (Cape May warbler, ovenbird, warbling vireo), and · Microtines (red-backed vole and deer mouse).

Justification for selecting these wildlife species is presented in Appendix J. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 39 Prepared By: The Oil Sands Vegetation Reclamation Committee

Table 5.1 Factors Used to Select Target Wildlife Species

Criteria For Selecting Wildlife Species Description

Ecological Importance The degree to which wildlife species is involved in or contributes to nutrient cycling, the food chain or physical maintenance of ecological communities.

Niche Representativeness The degree to which the species represent the habitat requirements of a variety and Abundance of other wildlife species, and that they are relatively abundant in the region.

Resource Value The value of species for subsistence and recreational hunting and trapping, and non-consumption use (wildlife viewing, photography or aesthetic value).

The general habitat requirements of the target wildlife species are presented in Appendix J.

5.2 ECOSITES AND LANDSCAPES THAT MEET THE DEFINITION OF WILDLIFE HABITAT

The ecosites that can be supported by the landscape, drainage and soil characteristics of the reclaimed oil sands leases have been discussed in Section 3.1 and include the following ecosites: a (lichen), b (blueberry), c (Labrador tea-mesic), d (low bush cranberry), e (dogwood), f (horsetail), g (Labrador tea - submesic), and h (Labrador tea/horsetail), i (bog), j (poor fen), k (rich fen), and l (marsh).

The ecosites and landscape patterns on reclaimed oil sands leases that would meet the habitat requirements of the target wildlife species are outlined in Appendix J.

5.3 MAINTAINING GUIDING PRINCIPLES AND RECLAIMING DISTURBED AREAS AS WILDLIFE HABITAT

Several approaches to the material handling, water management and revegetation would increase the topographic and habitat diversity, water availability and shelter and forage quality available for wildlife species. An undulating landscape with a diversity of slopes, aspects, elevations and moisture holding capabilities, slightly more rolling than that designed for reforestation, and with a mosaic of vegetation types with a variety of lakes and ponds is recommended for wildlife habitat. Both landscape diversity and plant community biodiversity need to be addressed when designing habitat for wildlife. Palatable plant species, both herbaceous and woody-stemmed should be included in starter plant species mixes (Table J.1, Appendix J) outlines the palatability of plant species that occur in the oil sands region). The addition of slash and deadfall will improve the quality of habitat of the reclaimed landscape for some wildlife species. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 40 Prepared By: The Oil Sands Vegetation Reclamation Committee

Each plant community will provide habitat with varying degrees of wildlife capability for different species. Wildlife utilization of the reclaimed areas will increase as the food and shelter becomes available with the maturing vegetation on the reclamation sites. Some mammalian species are routinely observed on the reclamation sites, while others frequently utilize the fringe areas adjacent to undisturbed stands. Stands will increasingly be utilized by these latter animals as they mature. For example, lynx, fisher and red squirrels prefer climax coniferous forest with a high degree of crown closure, numerous deadfalls and a few openings where shrubs can develop. Wildlife utilization of reclaimed areas will also vary by season.

5.3.1 Landscape Diversity

Wildlife species that exist in forest ecosystems vary widely in their requirements for space. Territory or home range sizes vary from less than one hectare for small animals such as red-backed voles and many songbird species to hundreds of square kilometers for wide ranging species such as wolves and bears. A forest must be spatially diverse to support a high diversity of wildlife, although the degree of diversity required varies from species to species. Generally, most species that occupy extensive ranges tend to be habitat generalists; that is, they use a diversity of preferred habitat types. The most difficult species to manage in a reclamation plan are habitat specialists, those species requiring relatively uniform environments or specific habitat conditions, or single habitat types. Most habitat specialists have relatively small home ranges, although some forest interior species require larger blocks of habitat.

Depending on the wildlife target species selected, a mosaic of plant community types will need to be established within the reclaimed landscape to meet the habitat requirements of wildlife. The size, shape and dispersion patterns of the plant communities are referred to as the landscape diversity. Using the coarse filter approach, the mosaic of plant communities (number of communities, size, shape and dispersion) established should be similar to those in the predisturbed landscape. Using the fine filter approach, the size, shape, and dispersion of plant communities will be determined by the landscape requirements of the target wildlife species. Various sizes of habitat blocks required by the potential target wildlife species have been presented in Appendix J.

The landscape and resulting habitats can be diversified in two ways:

· First, overburden dump and dyke walls can be recontoured to provide a complex topography of slopes and aspects. Recontouring can include micro-scale modifications or macro-scale modifications. Micro-scale modifications can create microsites. Differences in aspect, soil moisture regime and water or snow accumulation between micro-sites will result in improved vegetation diversity. This will in turn benefit wildlife by providing a greater diversity of browse and forage species. An example of a recontoured dump (macro-scale modifications) designed to provide good ungulate habitat is described by Proctor et al. (1983). It is crescent shaped with the outer circumference consisting of south-facing slopes with grass for winter range development, and with trees on the north side of the structure. Snow fences on south- and west-facing slopes can help retain moisture and facilitate shrub establishment which provide shelter on winter range. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 41 Prepared By: The Oil Sands Vegetation Reclamation Committee

· Second, water impoundments can be developed throughout the reclaimed site. These features would provide water sources and escape cover with additional tree and shrub growth along the edges.

The maintenance of diversity is also dependent on the maintenance of wildlife movement corridors between habitat patches. These corridors meet the needs of wildlife in two ways. The first is for species that make periodic movements between different habitat types required for breeding, birthing, feeding, cover and roosting. Second, such movements may range from annual migrations of ungulates between winter and summer ranges, to daily movements of birds between feeding and roosting sites. Such movements allow for gene flow between populations and recolonization of areas following local extinctions.

It is difficult to specify criteria for the design of movement corridors for wildlife. Requirements will vary between species and because of factors such as vegetation density, topography and proximity to disturbance factors. Furthermore, there has only been limited research conducted to determine the design and utility of movement corridors. Based on the literature that does exist, forested corridors 100-500 m wide are probably required to effectively support seasonal movements of large ungulates such as moose. The establishment of effective movement corridors in the oil sands region may, however, be dependent on the collection of baseline information to identify key movement corridors and their characteristics. Such studies should also examine the role natural corridors, such as river valleys, play in supporting local and regional wildlife movement patterns.

5.3.2 Plant Community Biodiversity

Plant community biodiversity refers to the structural diversity (number of vegetation layers e.g., herbaceous, shrubs, trees) and plant diversity (number of species and their abundance) within plant communities that provide habitat.

The establishment of more structurally complex and productive wildlife habitats on reclaimed areas can be assisted by planting a diverse mixture of native plant species of different life forms (e.g., grasses, forbs, shrubs and trees). Selected species should include some of the more important wildlife food plants (Appendix J). The structure and composition of the initially established communities will be simplistic in comparison with the natural undisturbed ecosystems. The newly reclaimed communities will lack the “within habitat” diversity that characterizes natural ecosystems. Over the long-term, however, additional native species should recolonize reclaimed areas, resulting in an increase in plant and animal diversity. As time passes, structural diversity will also be created through natural succession and through the succession of forest communities after natural disturbance such as fire. For relatively complex ecosystems it may take hundreds of years before recolonization is complete and the full complement of native species are restored. The rate of recolonization of reclaimed landscapes by native plant species can be increased by retaining refugia or “islands” of intact natural ecosystems within the larger development area. This practice can be achieved by leaving intact, areas within the development footprint that are not required for excavation or facilities construction. Where possible, native habitat corridors Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 42 Prepared By: The Oil Sands Vegetation Reclamation Committee should be maintained to connect extensive reclaimed areas with undisturbed habitat. These connections would accelerate recolonization of reclaimed areas by wildlife and would enhance habitat interspersion. These refugia would serve as sources of seed for native plant establishment and would assist in speeding recolonization of reclaimed areas by amphibians, birds, small mammals and the hundreds of species of invertebrates that exist in forest soils. Another method of increasing the rate of recolonization is to transplant patches of soil and vegetation from natural ecosystems to reclamation areas. A mixture of peat and underlying surficial materials, spread over reclamation areas as a soil amendment, contains seeds and roots of many native plant species, some of which become established on the reclaimed site. Placing this material in “islands” across extensive reclamation areas is expected to facilitate the recovery of natural biodiversity.

5.3.3 Starter Plant Species

Two approaches to establishing plants are currently used on oil sands leases:

· A layer of peat-mineral mix is applied to the reclamation sites and root fragments and seed within this amendment become established. The result of this volunteer growth is that reclamation sites are soon covered with a mix of vegetation, more closely related to a cut-over area than a mining disturbance, and

· Specific woody stemmed plant species are selected for out-planting to establish plant communities during the reclamation phase (Table 3.4). Many of the starter species are quite palatable for wildlife (Appendix J). For example, favoured browse species for moose include saskatoon, willow, low-bush cranberry, aspen, red-osier dogwood, white (paper) birch, balsam poplar, pincherry and chokecherry. Most of these species are included as part of the revegetation program, while the remainder are expected to invade naturally.

5.3.4 Slash and Deadfall

Habitat in boreal forests is provided not just by living vegetation but also by the dead and decaying vegetation components. Many species depend on snags and fallen logs for cover, as nesting or denning sites, drumming sites and as feeding substrates. Some of these wildlife benefits could be achieved by distributing logs and slash across areas undergoing reclamation. This practice would also result in nutrient enrichment of these reclamation areas. Decomposing slash provides a moist seedbed, and serves as sources of mycorrhizal fungi. Deadfall is also an important habitat element for small mammals, such as red-backed voles, which consume the fruiting bodies of the fungi and serve as agents of dispersion for spores. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 43 Prepared By: The Oil Sands Vegetation Reclamation Committee

6. MONITORING PROGRAMS

Monitoring programs are necessary to determine if the land use objectives have been achieved. For each of the two land use objectives, commercial forest and wildlife habitat, the benchmarks to assess whether the reclamation approach has been successful has been identified, and the detailed monitoring program has been described. Monitoring options have also been presented for diversity/biodiversity.

6.1 MONITORING PROGRAM TO VERIFY ACHIEVEMENT OF COMMERCIAL FOREST OBJECTIVES

The commercial forest monitoring program has two purposes. The first purpose is to demonstrate compliance with the conditions of the approval to operate the oil sands facility. The second purpose is to provide information to improve continued oil sands reclamation activities.

Two monitoring programs are recommended to ensure that a commercial forest is being established: a program to verify the establishment of tree seedlings to meet compliance regulations and a program to verify the establishment of a productivity equivalent to or better than a natural forest. The latter long-term productivity of the forest is dependent on the soils and other site characteristics.

6.1.1 Compliance - Establishment of Seedlings

The establishment of a forest is considered complete when sufficient acceptable growing stock is on the site at 14 years after planting. Acceptable growing stock is described in the Alberta Regeneration Survey Manual (Alberta Environmental Protection 1994b).

6.1.1.1 Benchmark - Reforestation Regulations

The Province of Alberta administers the forests within the Green Zone of Alberta under the authority of the Forests Act. The Timber Management Regulation (passed under the authority of the Forests Act) specifies the Alberta Regeneration Survey Manual (Alberta Environmental Protection 1994b) as the document describing reforestation standards. This manual is modified periodically to ensure currency with reforestation technology. The process generally involves negotiations between the province and representatives of the forest industry.

It is important to note that an equivalent amount of conifer, mixedwood and deciduous commercial forest is required in the reclaimed landscape as the predisturbance condition. The productive capability of these forests (appropriate seedling height at a given age) should meet or exceed predisturbance conditions as outlined in Table 6.1 and Appendix B. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 44 Prepared By: The Oil Sands Vegetation Reclamation Committee

Table 6.1 Yield Curve for Commercial Forest Ecosites

Ecosite Ecosite Phase Yield Class Site

a – lichen (a1) jack pine 24F - Coniferous jack pine Fair

b – blueberry (b1) jack pine-aspen 18M - Coniferous/Deciduous jack pine Medium

(b2) aspen (white birch) 2M - Deciduous Medium

(b3) aspen-white spruce 9M - Deciduous/Coniferous white spruce Medium

(b4) white spruce-jack pine 21M - Coniferous white spruce Medium

c – Labrador tea mesic (c1) jack pine-black spruce 25F - Coniferous jack pine Fair

d - low-bush cranberry (d1) aspen 2M - Deciduous Medium

(d2) aspen-white spruce 9M - Deciduous/Coniferous white spruce Medium

(d3) white spruce 21M - Coniferous white spruce Medium

e – dogwood (e1) balsam poplar 2G - Deciduous Good

(e2) balsam poplar-white spruce 9G - Deciduous/Coniferous white spruce Good

(e3) white spruce 21G - Coniferous white spruce Good

f – horsetail (f1) balsam poplar-aspen 2G - Deciduous Good

(f2) balsam poplar-white spruce 9G - Deciduous/Coniferous white spruce Good

(f3) white spruce 21G - Coniferous white spruce Good

g - Labrador tea- (g1) black spruce-jack pine 23F - Coniferous black spruce Fair subhygric

h - Labrador tea/horsetail (h1) white spruce-black spruce 21M - Coniferous white spruce Medium Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 45 Prepared By: The Oil Sands Vegetation Reclamation Committee

The benchmark to be applied will be the standards defined in the Alberta Regeneration Survey Manual (Alberta Environmental Protection 1994b) in place at the time of planting the reclaimed site.

Currently, there are three acceptable standards, all of which require 80% of the plots in a standard survey to be stocked with an acceptable tree(s) to a minimum height standard. The three standards are conifer, mixedwood or deciduous. The conifer standard requires a minimum 70% stocking with conifer (white spruce, jack pine or black spruce) seedlings and 10% conditional seedlings (aspen, balsam poplar, birch or fir). The conifer seedling may not have another tree or shrub seedling within 1 meter that is taller than the crop trees. The mixedwood standard reduces the conifer requirements to 50% and the deciduous standard does not require a conifer component.

The choice of standard in the Alberta Regeneration Survey Manual is based on the stand occupying the site prior to harvest. This provides a silviculture prescription that is appropriate for the ecosite. On reclaimed oil sands sites the important characteristics that define the choice of standard are slope position, aspect and moisture regime. Drier sites (top of slope and south facing) tend to be more suitable for jack pine with a trend to aspen, mixedwood, white spruce and eventually black spruce as the moisture regime increases to flat sites with impaired drainage. It is important to note that an equivalent amount of conifer, mixedwood, and deciduous commercial forest, at an equivalent productive capability (fiber volumes as per Table 6.1 and Appendix B) is required in the reclaimed landscape as the predisturbance condition.

6.1.1.2 Forest Establishment Program Design

The forest establishment program is to follow the reforestation standards for all sites designated for commercial forest production. Each unit with a uniform silviculture prescription should be considered a stand. The program requires each stand to meet establishment standards by five to eight years after the reclamation date and meet the performance standard prior to fourteen years after the reclamation date. The reclamation date is the first day of May immediately following the tree planting phase of reclamation.

The Alberta Regeneration Survey Manual, Section 3, details the field survey procedures to be used including number of plots, layout, data collection and mapping. Sections 4 and 5 of the manual outline office compilation and survey submission procedures. Tally cards can be obtained from the local Land and Forest Service office. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 46 Prepared By: The Oil Sands Vegetation Reclamation Committee

The standard that is to be applied to an individual stand will be the legislated standard at the time of reclamation and will not change after the start of reclamation activities, regardless of changes in the Alberta Regeneration Survey Manual that may occur from time to time.

The deciduous regeneration standards, subject to amendment from time to time, will apply to reclaimed oil sands sites reforested for deciduous landbase. The 1997 standard requires an average density of 7000 deciduous stems at the establishment survey (3 to 5 years), which is not readily achievable on most reclaimed oil sands sites. Therefore, it is recognized that most of the deciduous landbase will be subject to a performance survey (10 to 14 years) where there is no density requirement but subject to the 80% stocking and the associated height requirements.

6.1.2 Long-term Forest Productivity

The forest establishment program includes surveys up to 14 years on every stand but does not continue past the 14 year performance survey unless the request for a Reclamation Certificate is delayed. The long-term measurement of productivity will be completed by measuring forest growth over time and the changes in the soil capability. The forest establishment program will be based on target ecosites (not individual stands). Due to uncertainties related to the productivity of forests on reconstructed soils, additional information is required including:

· Assessment of crop tree health describing any symptoms of nutrient deficiency, and

· Measurement of the last 3 years annual height increments to determine if growth is comparable to that predicted by the stand yield curves or growth intercept curves on corresponding soil capability classes. In order to confirm or refine the soil capability and forest productivity relationship, as modeled by the soil capability classification approach, it is necessary to have ongoing monitoring of soils and forest growth at the same plots. Such comparisons are required to properly characterize baseline conditions as well. Superimposing forest inventory polygons on soil polygons to establish forest-soil relationships provides a less accurate comparison.

Polygons to sample forest productivity will be randomly selected permanent sampling plots measured every 5 years.

6.1.2.1 Benchmark - Mean Annual Increment and Site Indices

The forest productivity program benchmarks will be the yield curves developed by Alberta-Pacific Forest Industries (Timberline Forest Inventory Consultants 1997). These yield curves are reviewed periodically (approximately every 10 years) and adjusted according to additional data collected during the period. Table 6.1 summarizes the appropriate yield curves for each phase of the commercial forest ecosites. The yield curves provide a variable Mean Annual Increment Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 47 Prepared By: The Oil Sands Vegetation Reclamation Committee

(MAI) and an expected volume per hectare based on the age of the stand. These quantities provide the baseline for stand productivity measurement.

6.1.2.2 Benchmark - Measurement of Soil Capability

The soil monitoring program is intended to be integrated with the forest productivity monitoring, hence soil sampling sites should be in the immediate vicinity of the permanent forest plots and should be sampled at the same frequency. When plots are being established, it is recommended that the soils be inspected to ensure they match the design profile and are representative of the reclaimed area. Table 6.2 displays the forest and soil monitoring requirements at various stages of the reclamation process.

6.1.2.3 Long-term Forest Productivity Program Design

The Northern Interior Vegetation Management (NIVMA) Sampling Protocol (Szauer 1995) will provide the basic plot design for measurement of forest productivity. This protocol may require slight modification for oil sands reclamation. The design will include a permanent plot with remeasurement every 5 years for the first 15 years after reclamation and every 10 years thereafter. This survey system will be able, or could be modified to, provide information for other monitoring programs such as wildlife habitat and biodiversity.

There should be three plots created for the first 300 ha of each reclamation prescription. Every additional 100 ha of the same reclamation prescription should have another plot created. Continued use of the same reclamation prescription for subsequent years should continue to add to the data collection. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 48 Prepared By: The Oil Sands Vegetation Reclamation Committee

Table 6.2 Forest and Soil Monitoring of Benchmark Sites

MAI and Site Index Soil Capability(a) Nutrient Levels(b) Other(c)

Baseline - Year Zero ü ü ü ü

5 year ü ü ü

10 year ü ü ü ü

15 year ü ü ü

25 year v ü ü ü

10 year increments ü ü ü

Sampling Protocol a Standard profile and site description at representative site outside edge of vegetation plot. b,c For LFH take minimum of 10 subsamples to make 1 sample, along outside edge of vegetation plot; for TS (topsoil) and US (upper subsoil) take minimum of 5 subsamples each to make 1 sample each, along outside edge of vegetation plot; TS should be about 10-20 cm thick and US should be below TS to about 50 cm. Sample by horizon rather than by depth, and record depths at each subsample point. Laboratory (a) pH (H20), Electrical Conductivity, SAR, Organic Carbon (OC) content, bulk density, (texture only first time) in TS, US, LS. (b) Total N, P, K. and S; OC, pH (CaCl2) in TS and US, and LFH when one develops. (c) Bitumen content in TS and US on overburden materials. Optional mychorrizal studies in TS and US (to be determined). Frequency Match forest productivity monitoring; that is, 3 plots for first 300 ha of any reclamation prescription, plus an additional plot for every additional 100 ha of the same reclamation prescription. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 49 Prepared By: The Oil Sands Vegetation Reclamation Committee

6.2 MONITORING PROGRAM TO VERIFY ACHIEVEMENT OF WILDLIFE HABITAT OBJECTIVES

The wildlife habitat monitoring program will provide ongoing information for periodic revisions of this document and the practices being used to establish wildlife habitat on reclaimed oil sands leases.

As noted in Section 5, habitat can be re-established based on either a coarse filter or fine filter approach, and both approaches have been discussed in this document.

A monitoring program to verify the achievement of wildlife habitat objectives using the coarse filter approach would need to document whether the same types, abundance, sizes (patches) and distribution patterns of plant communities as existed in the predisturbance landscape have been re-established in the reclaimed landscape. This is the benchmark for the monitoring program based on the coarse filter approach. The landscape level monitoring program for biodiversity outlined in Section 6.3, documents changes in ecosite phase communities. Therefore, data from this program could be evaluated to assess whether habitat capability has been generally replaced for all wildlife species. The monitoring program should also be designed to assess whether the design criteria for movement corridors in specific areas of the lease have been achieved.

A monitoring program to verify the achievement of wildlife habitat objectives using the fine filter approach would need to document whether the biophysical habitat requirements of several wildlife species have been provided in the reclaimed landscape. The food, cover, landscape and special habitat requirements of twelve indicator target wildlife species that represent the habitat requirements of a broad range of species in the region are outlined in Appendix J.3.0. These habitat requirements are the benchmarks for the monitoring program based on the fine filter approach. Appendix J.4.0 indicates which ecosite phases provide high and moderate suitability habitat for these species.

Habitat Suitability Index (HSI) models and Habitat Evaluation Procedures (HEP) developed by the US Fish and Wildlife Service have been widely applied to assess wildlife habitat quality and could be readily applied to evaluate the suitability of reclaimed landscapes for wildlife. An HSI model uses the physical and biological attributes of a particular habitat to calculate an index of habitat suitability that is assumed to be proportional to the habitat’s carrying capacity for a species. HSI values for a particular habitat range between 0 and 1; a habitat with an HSI of 1 is considered to have optimal habitat, while an HSI of 0 indicates that the habitat has no value for the wildlife species in question. The Habitat Evaluation Procedure combines measures of habitat quality in terms of numbers of Habitat Units (HU) available in an area, where HU=HSIxArea.

The monitoring program should be designed to collect data on the biophysical parameters in the models. Initially, these models could be run in the planning stages to determine whether planned landscapes would contain suitable habitat for wildlife species over time. As vegetation Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 50 Prepared By: The Oil Sands Vegetation Reclamation Committee communities are established, the models should be run using actual field measurements to determine whether there is a need for further refinement of reclamation plans.

6.3 DIVERSITY/BIODIVERSITY MEASUREMENT OPTIONS

The objective of the diversity/biodiversity monitoring program will be to compare diversity on reclaimed lands to the range of natural variability. It is important that diversity in reclaimed lands approach the natural range of variation in the predisturbed landscape. Long-term control sites in natural vegetation types are needed to provide baseline data for comparison with reclaimed lands because change to the environment due to climate, industrial development, air pollution and wildfire can influence development of natural vegetative communities. The range of diversity should be determined across the scope of landscape, plant community and within species levels (Figure 2.1).

Program design features for diversity/biodiversity monitoring include:

· The use of a detailed statistical power analysis to determine the overall number of sample sites required, and

· Sampling the sites beginning at year zero and then sampling every 5 to 10 years.

Parameters that should be measured during the monitoring programs are listed below:

· Landscape Level:

- Number, range of sizes and shapes, spatial distribution of ecosite phases, communities, and number of seral stages,

- Variability and distribution pattern (size, shape and dispersion) of slope classes, aspect, and reclamation substrates,

- Variability and distribution of drainage patterns,

- Topographical diversity, and

- Number and dispersion pattern (size, shape and dispersion) of soil types, capabilities, and moisture and chemical parameters. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 51 Prepared By: The Oil Sands Vegetation Reclamation Committee

· Plant Community Level:

- The range in number and abundance of plant species common to natural ecosystems as compared to the number and abundance of plant species found in the similar reclaimed ecosystem established, noting any endangered or vulnerable species in the predisturbance and reclaimed landscape,

- The diversity of vertical structures and productivities compared to natural systems,

- The range of site indexes for each plant community type,

- The abundance of indicator species and composition of indicator assemblages, and

- The abundance of exotic or hybrid species.

· Within Species (Genetic) Level:

For each species the following information should be documented or measured:

- The source location and elevation of seed or revegetative material,

- The numbers of seed sources and hence genetic diversity of plant material,

- Genetic composition of seed sources,

- Where applicable: variability in growth rates, nutrient cycling and successional trajectory,

- The types and abundance of insects and disease, and

- Variability in tree mortality rates and gap initiation.

Biodiversity monitoring methods are diverse and some potential methods are referenced in Schneider (1997, Appendix E).

7. ONGOING PROCESS REFINEMENT BASED ON MONITORING AND RESEARCH

These guidelines will be updated approximately every five years with new information derived from monitoring programs (described in Section 6.0) and research programs (described in Section 3.3). Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 52 Prepared By: The Oil Sands Vegetation Reclamation Committee

8. GLOSSARY OF TERMS

BIODIVERSITY: Biodiversity is the variety of living components of ecosystems. Biodiversity within plant communities is composed of the range of genetic, species, structural, and functional diversity. Structural diversity being the structure of plant communities both horizontally and vertically, while functional diversity encompasses the physiological differences between vegetation.

CAPABILITY See Land, Landscape and Soil Capability.

COMPOSITE/CONSOLIDATED Composite (syncrude) and consolidated (suncor) tailings is formed by TAILINGS: injecting mature fine tailings from the tailings ponds into the regular tailings sand stream, with a floculant such as gypsum. This mixture is sent to the tailings ponds to form a non-segregating soil mixture which will result in a trafficable surface in the reclaimed landscape.

ECOSITE: Ecological units that develop under similar environmental influences (climate, moisture, and nutrient regime). Ecosites are groups of one or more ecosite phases that occur within the same portion of the edatope (e.g., Lichen ecosite). Ecosite, in this classification system, is a functional unit defined by moisture and nutrient regime. It is not tied to specific landforms or plant communities as in other systems (lacate 1969), but is based on the combined interaction of biophysical factors that together dictate the availability of moisture and nutrients for plant growth. Thus, ecosites are different in their moisture regime and/or nutrient regime (Beckingham and Archibald, 1996).

ECOSITE PHASE: A subdivision of the ecosite based on the dominant tree species in the canopy. On some sites where a tree canopy is lacking, the tallest structural vegetation layer determines the ecosite phase (e.g., Shrubby and graminoid phases). Some variation in humus form or plant species abundance may be observed between ecosite phases (Beckingham and Archibald, 1996).

ECOSYSTEM: A system of living organisms interacting with each other and their environment, linked together by energy flows and material cycling.

EDATOPE: Moisture/nutrient grid that displays the potential ranges of relative moisture (very dry to wet) and nutrient (very poor to very rich) conditions and outlines relationships between each of the ecosites. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 53 Prepared By: The Oil Sands Vegetation Reclamation Committee

EQUIVALENT CAPABILITY: The ability of the land to support various land uses after conservation and reclamation is similar to the ability that existed prior to an activity being conducted on the land, but that the individual land uses will not necessarily be identical.

EXOTIC SPECIES: In this document, exotic species refer to plant species that are not native to the province and which are not native within the natural region of interest.

GRAMINOID PHASE: In the ecosite classification, graminoid phases are those ecosite phases dominated by grass or sedge species.

HABITAT SUITABILITY Habitat Suitability Index Models estimate the value of habitat for wildlife INDEX: species by relating a species’ need for food and cover to structural and spatial attributes of vegetation types within a defined area. The Habitat Suitability Index (HSI) refers to the quality or suitability for a species or species group, and ranges in value from 1.0 (optimal value) to 0.0 (no value).

HYDRIC: A soil moisture regime used to describe sites where the water table is at or above the soil surface all year.

HYGRIC: A soil moisture regime used to describe sites where water is removed slowly enough to keep the soil act for most of the growing season.

LAND CAPABILITY: Ability of land to support a given land use, based on an evaluation of the physical, chemical and biological characteristics of the land, including topography, drainage, hydrology, soils and vegetation.

LANDSCAPE CAPABILITY: The evaluation of landscape factors as they affect general tree growth including: slopes, position, aspect, stoniness and erosion.

LICHENS: A group of organisms consisting of fungi and algae growing together symbiotically.

MAP UNIT: A mappable portion of the soil landscape with attributes varying within narrow limits that are determined by the intensity of survey and its objectives such as land use planning and management requirements.

MESIC: A soil moisture regime used to describe sites where water is removed somewhat slowly in relation to supply and where soil may remain moist for significant but sometimes short periods of the growing season.

MONTANE: A climatic region in alberta defined in the natural regions and subregions of alberta. The montane subregion is characterized by forests of lodgepole pine, douglas fir, white spruce and aspen. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 54 Prepared By: The Oil Sands Vegetation Reclamation Committee

NATIVE SPECIES: Species native to the province and to the natural sub-region of interest.

OVERSTOREY SPECIES: An overstorey is the tallest vegetation layer within a plant community and most often consists of trees in the oil sands area. An overstorey species is a species that occurs within the overstorey vegetation layer.

PASSERINES: A group of perching birds belonging to the zoological order passeriformes.

PEAT-MINERAL MIX: A mixture of peat and mineral material resulting in a “mineral” soil. Peat- mineral mixes typically contain a ratio of peat:mineral ranging from about 1:1 to 1:4 (volume basis).

PLANT COMMUNITY TYPE: A subdivision of the ecosite phase and the lowest taxonomic unit in the classification system. While plant community types of the same ecosite phase share vegetational similarities, they differ in their understorey species composition and abundance. These differences may not be mappable from aerial photography but may be important to wildlife, recreation, and other resource sectors (Beckingham and Archibald 1996).

PRODUCTIVITY: Evaluation of tree growth by site index which is a measurement of tree growth expressed as height (m) at 50 years breast height and/or by mean annual increment expressed as volume m3/ha.

RECLAIMED SOIL: Soil created by the selective placement of suitable topsoil and subsoil material on reshaped spoil, or parent geological material.

REFUGIA: A stand of undisturbed natural vegetation retained within a mine development area that serves as a source of native species for revegetation.

RIPARIAN: Areas or species associated with river or creek systems or other wetlands.

SOIL: Unconsolidated, mineral or organic material at the surface of the earth that serves as a medium for plant growth. SOIL CAPABILITY: The nature and degree of limitations imposed by the physical, chemical and biological characteristics of a soil for forest productivity.

SOIL SERIES: A soil series is a conceptual class that has defined limits of relatively detailed soil properties including horizon depth and expression, color, texture, structure, consistence, stoniness, salinity, ph and soil drainage. In soil mapping, the names of soil series are often used to name the map units.

SOIL TYPE: In ecosite classification soil types are functional taxonomic units used to stratify soils based on soil moisture regime, effective soil texture, Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 55 Prepared By: The Oil Sands Vegetation Reclamation Committee

organic matter, thickness, and soil depth. The concept of the soil type is more general than that of a soil series in this context.

SUBHYDRIC: A soil moisture regime used to describe sites where the water table is at or near the surface for most of the year.

SUBHYGRIC: A soil moisture regime used to describe sites where water is removed slowly enough to keep the soil wet for a significant part of the growing season.

SUBMESIC: A soil moisture regime used to describe sites where water is removed readily in relation to supply and where water is available for moderately short periods following precipitation.

SUBXERIC: A soil moisture regime used to describe sites where water is removed rapidly in relation to supply and where soil is moist for only short periods following precipitation.

TAXONOMIC UNIT: A classified unit within a hierarchical classification system. In this case the hierarchical classification system is the ecological classification system (Beckingham and Archibald 1996).

UNDERSTOREY SPECIES: A vegetation species found in one of the lower vegetation layers within a plant community; lower layers are commonly shrub, grass or moss layers.

VEGETATION: The species which comprise a plant community including trees, shrubs, forbs, graminoids, mosses, and lichens.

VERY XERIC: A soil moisture regime used to describe sites where water is removed extremely rapidly in relation to supply and soil is moist for a negligible period following precipitation.

XERIC: A soil moisture regime used to describe sites where water is removed very rapidly in relation to supply and soil is moist only for brief periods following precipitation. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 56 Prepared By: The Oil Sands Vegetation Reclamation Committee

9. LITERATURE CITED

Alberta Environmental Protection. 1996a. Forest Site Interpretation and Silvicultural Prescription Guide for Alberta, Version 2.0. Published by the Environmental Training Center. Publication No: 105 ISBN 0-7732-5030-1.

Alberta Environmental Protection. 1996b. Recommended Native Grasses and Legumes for Revegetating Disturbed Lands in the Green Area. Land and Forest Service Recommendation and Policy Document.

Alberta Environmental Protection. 1996c. Fort McMurray-Athabasca Oil Sands Subregional Integrated Resource Plan.

Alberta Environmental Protection. 1994a. Timber Harvest Planning and Operating Ground Rules. Publication No.: Ref. 71.

Alberta Environmental Protection. 1994b. The Alberta Regeneration Survey Manual (Addendum April 1997). Publication No.: Ref. 70.

Alberta Forest Conservation Strategy Steering Committee (AFCSSC). 1997. Alberta Forest Conservation Strategy, Alberta Environmental Protection, Land and Forest Service. Edmonton, Alberta.

Beckingham, J.D. and J.H. Archibald. 1996. Field Guide to Ecosites of Northern Alberta. Natural Resoures Canada, Canadian Forest Service, Northwest Region, North. For. Cent. Spec. Rep. 5. Edmonton, Alberta.

Beckingham, J.D., I.G.W. Corns, and J.H. Archibald. 1996. Field Guide to Ecosites of West- Central Alberta. Natural Resources Canada, Canadian Forest Service, Northwest Region, North. For. Cent. Spec. Rep. 9. Edmonton, Alberta.

Beckingham, J.D., D.G. Neilsen, and V.A. Futoransky. 1996. Field Guide to Ecosites of the Mid- Boreal Ecoregions of Saskatchewan. Nat. Resoures Canada, Canadian Forest Service, Northwest Region, North. For. Cent. Spec. Rep. 6. Edmonton, Alberta.

Canadian Council of Forest Ministers (CCFM). 1995. Defining Sustainable Forest Management: A Canadian Approach to Criteria and Indicators. The Canadian Forest Service. Ottawa, Ontario.

Canadian Council of Forest Ministers (CCFM). 1992. National Forest Strategy. Sustainable Forests: A Canadian Commitment. Canadian Forest Service. Ottawa, Ontario. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region Page 57 Prepared By: The Oil Sands Vegetation Reclamation Committee

Christiensen, N.L. 1996. The Report of the Ecological Society of America Committee on the Scientific Basis for Ecosystem Management. Ecol. Applic. 6:665-691.

Environment Canada (EC). 1995. Canadian Biodiversity Strategy. Environment Canada. Hull, Quebec.

Everett, R. 1993. Eastside Forest Ecosystem Health Assessment: Executive Summary. U.S. Department of Agriculture. Washington. D.C.

Gerling, H.S., M.G. Willoghby, A. Schoepf, K.E. Tannas and C.A. Tannas. 1996. A Guide to Using Native Plants on Disturbed Lands. Alberta Agriculture, Food and Rural Development and Alberta Environmental Protection. 247 pp.

Grumbine, E. 1994. What Is Ecosystem Management? Cons. Biol. 8:27-38.

Hunter, M.L. 1991. Natural Fire Regimes as Spatial Models for Managing Boreal Forests. Biol. Conser. 72:115-120.

Leskiw, L.A. 1998. Land Capability Classification for Forest Ecosystems in the Oil Sands Region. Revised Edition. Prepared by Tailings Sands Reclamation Practices Working Group.

Leskiw, L.A., A. Laycock, and J. Pluth. 1996. Baseline Soil Survey for the Proposed Suncor Steepbank Mine. Prepared for Suncor Inc. Oil Sands Group.

Noss, R.F. 1990. Indicators for Monitoring Biodiversity: A Hierarchical Approach. Cons. Biol. 4:355-364.

Schneider, R. 1997. Ecological Diversity Monitoring Framework, Draft #6. Prepared for the Biodiversity Monitoring Working Group.

Soil Quality Criteria Working Group. 1987. Soil Quality Criteria Relative to Disturbance and Reclamation. Revised. Edmonton, AB. 56 pp.

Szauer, T. 1995. Introduction to the Revised NIVMA Silvicultural Monitoring Protocol. In: Proceedings of the NIVMA 1995 Annual General Meeting.

Timberline Forest Inventory Consultants Ltd. 1997. Alberta Pacific Forest Industries Inc. Round 8 Timber Supply Analysis. Yield Curve Development.

United Nation Environment Program (UNEP). 1992. Convention on Biological Diversity. UNEP Publication No. 92-7807. Geneva, Switzerland.

Wildlife Management Division. 1996. The Status of Alberta Wildlife. Wildlife Management Division, Natural Resources Service, Alberta Environmental Protection. Publication Number I/620. 44 pp. APPENDIX A

EXISTING ECOSITES IN THE OIL SANDS REGION Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX A Page A-1 Prepared By: The Oil Sands Vegetation Reclamation Committee

EXISTING ECOSITES IN THE OIL SANDS REGION

Prepared by Judith Smith and Kimberly Ottenbreit, BOVAR Environmental

A1.0 OVERVIEW

This overview of local ecosystems in the oil sands region describes the most common terrestrial ecosystems, their dominant plant species composition, successional status, and the ecological factors that explain their position in the landscape. The description is focused on terrestrial ecosystems, although the aquatic ecosystems are outlined briefly.

An Ecosystem is an interacting, dynamic system of living organisms (plants, animals, fungi, bacteria) and the physical environment (soil, air, water, bedrock). Ecosystems are determined by climate, landform, topography, soils, animals and vegetation.

An Ecological Classification System (ECS) is an attempt to organize these complex systems and functions into understandable and workable units. The ecosystems described for the oil sands region are based on the Ecological Classification System prepared by Beckingham and Archibald (1996) for the Boreal Forest and Canadian Shield Natural Areas of Alberta.

· The ECS consists of an integrated hierarchical ecological system with three levels: ecosite, ecosite phase and plant community type, and a separate soil classification.

· The ECS is nested within Alberta geographically based on the Natural Region and Subregion Classification System prepared by Alberta Environmental Protection (1994).

A1.1 ECOLOGICAL UNITS

Table A.1 illustrates the Ecological Units in the oil sands region, using the two classification systems outlined. The units are defined through an analysis of climate, vegetation, soils and sites i.e., topography, slope, aspect, etc. The highest Ecological Unit is the Natural Region and the lowest unit is the Plant Community Type.

Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX A Page A-2 Prepared By: The Oil Sands Vegetation Reclamation Committee

Table A.1 Ecological Classification System for the Oil Sands Region

Classification Level Ecological Unit Definition Oil Sands Classification

Natural Region — Boreal Forest

Subregion Similar climate, natural vegetation and to Central Mixedwood lesser extent soils

Ecosite Similar moisture and nutrient regimes 12 units (e.g., lichen)

Ecosite Phase Similar dominant species in the canopy 25 units (i.e. trees, shrubs) (e.g., jack pine lichen)

Plant Community Type Similar understory species composition 73 units and abundance (e.g., bearberry, blueberry, or alder)

Source: Beckingham and Archibald (1996).

The oil sands area is located within the Boreal Forest Natural Region and the Central Mixedwood Subregion. The following is a description of the some of the characteristics of the subregion:

· Climate is sub-arid to sub-humid, and cool continental, which is characterized by long, cold winters and short, warm summers.

· Topography is largely subdued although hill complexes and uplands are present i.e., Firebag and Thickwood Hills, and Birch Mountains.

· Dominant tree species is trembling aspen, with other common species being black spruce, white spruce, jack pine and balsam poplar.

· Succession is typically to white spruce and balsam fir. However, frequent fires often occur before this climax plant community is reached.

· Dominant soils are Organic, Gray Luvisols, Brunisols and Gleysols.

Ecosites are subunits of the Subregion, defined by their positions on an edatophic grid based on nutrient and moisture regime.

Ecosites are subdivided into Ecosite Phases based on plant canopy cover.

Ecosites phases are subdivided into Plant Community Types based on the composition and abundance of the understory vegetation. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX A Page A-3 Prepared By: The Oil Sands Vegetation Reclamation Committee

A1.2 CHARACTERISTICS OF ECOSITES, ECOSITE PHASES AND PLANT COMMUNITY TYPES IN THE OIL SANDS AREA

This section provides a brief review of the 12 ecosites described by Beckingham and Archibald (1996) for the boreal mixedwood area. Plant communities may vary based on topography and soil type. For example, dry, sandy sites may have jack pine, whereas wetlands may be treed with black spruce and/or larch.

The 12 ecosites are spatially arranged on an edatopic grid based on nutrient and moisture regime (Figure A.1). These are then subdivided into ecosite phases and plant communities as listed in Figure A.2.

· a or lichen ecosite. The nutrient regime is poor and the moisture regime is xeric to subxeric. There is one ecosite phase, namely jack pine, and three plant communities: bearberry, blueberry, or green alder. These sites occur on dry, coarse-textured, well-drained soils. Succession to black spruce is generally precluded by fire. Drought limitations are to be considered on this ecosite.

· b or blueberry ecosite. The nutrient regime is poor to medium and the moisture regime is subxeric to submesic. There are four ecosite phases associated with this ecosite. These are: pine/aspen, aspen (white birch), aspen/white spruce, and white spruce/pine. The plant communities are separated based on the cover value of blueberry, green alder and Labrador tea. Hairy wild rye can be found on these sites as well. As in the lichen ecosite, the soils are relatively dry, rapidly drained and coarse-textured. Succession may proceed to a white spruce dominated phase. Drought limitations are a factor on these sites.

· c or Labrador tea (mesic) ecosite. The nutrient regime is poor and the moisture regime is mesic. There is one ecosite phase which is jack pine/black spruce. Ground cover is dominated by Labrador tea, green alder and/or feathermoss. Successionally, this ecosite trends towards black spruce but this is rare owing to fire.

· d or low-bush cranberry ecosite. This ecosite is located centrally in the edatopic grid. It is medium in terms of nutrient regime and mesic in moisture. There are three ecosite phases: aspen, aspen/white spruce and white spruce. Several understory species determine plant community type. These include low-bush cranberry, buffaloberry, saskatoon, rose, alder, hazelnut and feathermoss. Deciduous species will succeed to white spruce. As coniferous canopy increases, understory vegetation will decline resulting in a higher moss cover. Vegetation competition is high in this ecosite. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX A Page A-4 Prepared By: The Oil Sands Vegetation Reclamation Committee

Nutrient Regime

Moisture Regime Very Poor A Poor B Med. C Rich D Very Rich E

Xeric 2

a Subxeric 3

Submesic 4 b

c Mesic 5 d

e Subhygric 6

g f Hygric 7 h

i Subhydric 8 j k

Hydric 9 l

Ecosites: a = lichen g = Labrador tea-subhygric subxeric/poor subhygric/poor b = blueberry h = Labrador tea/horsetail submesic/medium hygric/medium c = Labrador tea-mesic i = bog mesic/poor subhygric/very poor d = low-bush cranberry j = poor fen mesic/medium subhydric/medium e = dogwood k = rich fen subhygric/rich subhydric/rich f = horsetail l = marsh hygric/rich hydric/rich

Figure A.1 Edatope (moisture/nutrient grid) Showing The Location Of Ecosites For The Boreal Mixedwood Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX A Page A-5 Prepared By: The Oil Sands Vegetation Reclamation Committee Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX A Page A-6 Prepared By: The Oil Sands Vegetation Reclamation Committee

· e or dogwood ecosite. The nutrient regime is rich and the moisture regime is subhygric. There are three ecosite phases: balsam poplar/aspen, balsam poplar (aspen)/white spruce and white spruce. Indicator species of this ecosite include dogwood, river alder, and bracted honeysuckle. Another key indicator of this ecosite is the presence of ferns such as oak fern, lady fern and shield fern. These plant communities commonly occur along riverine or seepage areas which receive nutrients. They are high in species diversity and plant competition. This tends to be the most productive ecosite.

· f or horsetail ecosite. The nutrient regime is rich to very rich and the moisture regime is hygric. There are the same three ecosite phases as with the dogwood ecosite. Ground cover is dominated by horsetails. This ecosite is successional to white spruce. Sensitive to disturbance as water table may rise if trees are removed. Soil temperature and plant competition are high.

· g or Labrador tea (subhygric) ecosite. The nutrient regime is poor and the moisture regime is subhygric to hygric. There is one ecosite phase which is black spruce/jack pine. The understory vegetation is Labrador tea, feathermoss. Successionally mature stands maintain a small component of jack pine, but are black spruce dominated. Soil temperature and excess moisture may be limitations to productivity of this ecosite.

· h or Labrador tea-horsetail ecosite. The nutrient regime is medium and the moisture regime is hygric. There is one ecosite phase which is white spruce/black spruce. Main understory species are Labrador tea, horsetail, and feathermoss. Excess moisture and soil temperature limitations characterize this ecosite.

· i or bog ecosite. This ecosite is poor in nutrient regime and subhydric in moisture. Black spruce, Labrador tea, cloudberry and peatmoss characterize the ecosite. The water table maintains an edaphic climax community in the (i), (j) and (k) ecosites. Soil temperature limitations and excess moisture are considerations on these ecosites as well.

· j or poor fen ecosite. The nutrient regime is poor to medium and the moisture regime is subhydric. This ecosite is characterized by the presence of black spruce/larch, Labrador tea/dwarf birch, cloudberry/sedge and peatmoss/golden moss.

· k or rich fen ecosite. The nutrient regime is rich and the moisture regime is subhydric. This ecosite is characterized by the following species: larch, dwarf birch, sedge, golden moss and tufted moss. Other indicator species of this ecosite may include bog bean and marsh marigold. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX A Page A-7 Prepared By: The Oil Sands Vegetation Reclamation Committee

Nonforested communities also occur in the area. In addition to this, there are shrubby bogs (i) and fens (j, k) and graminoid fens (k). These simply lack the overstory canopy of trees and/or shrubs.

· l or marsh ecosite. The nutrient regime is rich and the moisture regime is hydric. These are graminoid dominated plant communities which include: bulrushes, cattails and reed grasses. Excess moisture is a limiting factor in this ecosite.

There is always considerable natural variability in the landscape with ecotones being common. Thus, not all sites can be easily distinguished and classified. This is evident, for example, in the positions of the ecosites on the grid, as they always intergrade with other, adjacent sites. In addition, species will often be found growing in areas for which they are not best adapted, or in suitable microsites.

Because of the complexity of plant communities they are very difficult to imitate or replace. However, an understanding of the natural plant communities and the many factors which influence their structure, provides a framework to assist us in re-creating these communities.

A1.3 MOISTURE REGIME CLASSES

Very Xeric Water removed extremely rapidly in relation to supply; soil is moist for a negligible time after precipitation. Xeric Water removed very rapidly in relation to supply; soil is moist for brief periods following precipitation. Subxeric Water removed rapidly in relation to supply; soil is moist for short periods following precipitation. Submesic Water removed readily in relation to supply; water available for moderately short periods following precipitation. Mesic Water removed somewhat slowly in relation to supply; soil may remain moist for significant but sometimes short periods of the year, available soil water reflects climatic input. Subhygric Water removed slowly enough to keep the soil wet for a significant part of the growing season; some temporary seepage and possible mottling below 20 cm. Hygric Water removed slowly enough to keep the soil wet for most of the growing season; permanent seepage and mottling present; possibly weak gleying. Sybhydric Water removed slowly enough to keep the water table at or near the surface for most of the year; organic and gleyed mineral soils; permanent seepage less than 30 cm below the surface. Hydric Water removed so slowly that the water table is at or above the soil surface all year; organic and gleyed mineral soils. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX A Page A-8 Prepared By: The Oil Sands Vegetation Reclamation Committee

A2.0 LITERATURE CITED

Alberta Environmental Protection. 1994. Natural Regions of Alberta. Edmonton, Alberta.

Beckingham, J.D. and J.H. Archibald. 1996. Field Guide to Ecosites of Northern Alberta. Natural Resources Canada, Canadian Forest Service, Northwest Region, North. For. Cent. Spec. Rep. 5. Edmonton, Alberta. APPENDIX B

CAPABILITIES AND DIVERSITY OF EXISTING SOILS WITHIN SOME OIL SANDS LEASES Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX B Page B-1 Prepared By: The Oil Sands Vegetation Reclamation Committee

CAPABILITIES AND DIVERSITY OF EXISTING SOILS WITHIN SOME OIL SANDS LEASES

Prepared by Leonard Leskiw, Can-Ag Enterprises Ltd.

B1.0 BASELINE SOILS AND VEGETATION

B1.1 STEEPBANK MINE

A comparison of soils and site index values for the Steepbank Mine Area was made by overlying the Alberta Vegetation Inventory polygons on the detailed soil survey. The soils ratings are based on the modal profile representing the soil series. Site indices are estimated for each polygon based on measurements from a number of sample plots located throughout the study area (Table B.1). These results are therefore approximate. In future mapping, it is recommended that soils descriptions and site index measurements be made at the same points. If suitable plots are selected, in terms of density, age and health, there is much more likelihood of observing a good correlation between soil capability and tree productivity. If random plots are located to characterize stands, as for inventory purposes, the soil capability and tree productivity relationships are weaker. It is important that the same approach be used for comparing baseline and reclaimed productivity levels.

Table B.1 Soil and Site Index Information for the Steepbank Mine Area LFH Mean Range Classificatio Parent Moisture Or Peat Soil Site In Site Soil n Material Texture Regime Thicknes Capabilit Index Index Series Surfac Sub- s y at 50 at 50 e surface (cm) Years Years Algar Gleysol M(a), F Peat SCL (b), CL hygric, 25 4 12 7-17 subhydric Firebag Brunisol F, M LS, SL SL, S xeric, 7 3 13 7-17 subxeric (Mildred) Kinosis Luvisol M, F SL, LS SCL, CL mesic, 11 2 15 13-17 subhygric McLelland Mesisol O Peat peat subhydric, >50 5 4 4 (fen) hydric McMurray Regosol F, F/M SiL, L SL, S mesic 20 2, 1 15 10-21 Gleyed Gleyed F, F/M SiL, L SL, S hygric 6 5, 4 11 5-18 McMurray Regosol Muskeg Mesisol O Peat peat hygric >50 5 10 3-17 subydric (bog) Rough Brunisol F/tar SL tar sand subxeric- 6 2, 3 13 6-19 Broken 2 sand subhygric Luvisol Rough Regosol M, M/r SL SCL subxeric- 17 3, 4 13 3-19 Broken 3 subhygric Luvisol

(a) F = fluvial; M = morainal; O = organic; r = residual (b) C = clay; L = loam; S = sand; Si = silt Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX B Page B-2 Prepared By: The Oil Sands Vegetation Reclamation Committee

B1.2 AURORA MINE

Soil capability ratings were assigned to representative profiles and these were compared with a number of site index values for selected vegetation plots for the Aurora Mine Area. The moisture regime and soil map unit were indicated for each vegetation plot, but a soil profile description was not available for each plot. Therefore, vegetation plots were selected only if the moisture regime matched that of the representative profile. Table B.2 shows a summary of the soil series, capability and site index relationship. Table B.3 shows individual plot data, thereby providing a breakdown of site index by species for each ecosite and series combination. Note that the vegetation plots were established to characterize the existing vegetation, so the site indices reflect actual growth rather than “potential” growth. In future mapping, it is necessary to have soils and site index rated at the same points, preferably at carefully selected plots (appropriate age, density, health).

Table B.2 Ecosites, Soils, Capability and Site Indices for Selected Sites at the Proposed Aurora Mine Area

Surface/ Numbe Soil Classifi- Parent Subsurfac Moisture Soil Site Index r Of Ecosite Series Cation Material(a e Regime Capabilit at 50 Years Plots ) (b) Texture y Mean Range lichen Fort Luvisol F SL/SCL subxeric 3 13.0 10-14 7 Mildred Brunisol F S/LS mesic 4 15.0 14-16 2 blueberry Dalkin Brunisol FE/F LS/S submesic 4 11.4 8-15 5 Fort Luvisol F SL/SL subxeric 3 8.8 7-12 4 low-bush Algar Gleysol L pt/L,SL,C hygric 4 11.0 10-12 2 cranberry Dover Luvisol L/M L/C mesic 3 20.8 16-26 4 Fort Luvisol F SL/SCL submesic 2 21.0 20-22 2 Mildred Brunisol F S/LS mesic 3 20.3 15-24 4 Steepbank Gleysol F/M pt/CL,SCL hygric 2 15.7 14-19 3 (aerated) horsetail Bitumount Gleysol F pt/S,LS,SL subhydric 5 9.4 8-13 4 Labrador tea Livock Luvisol F/M SiL/SiCL,CL submesic 2 11.0 9-13 2 bog Algar Gleysol L pt/L,SL,C hygric 4 11.0 11 1 Bitumount Gleysol F pt/S,LS,SL subhydric 5 6.0 5-8 3 Hartley Mesisol O/M pt/pt subhydric 5 8.2 7-11 5 McLelland Mesisol O pt/pt subhydric 5 12.5 7-23 4 Muskeg Mesisol O pt/pt hydric 5 11.0 8-14 2 Steepbank Gleysol F/M pt/CL,SCL hygric 4 8.4 5-12 5 (reduced) poor fen Bitumount Gleysol F pt/S,LS,SL subhydric 5 10.7 6-14 3 Hartley Mesisol O/M pt/pt subhydric 5 14.5 8-22 6 rich fen Hartley Mesisol O/M pt/pt subhydric 5 10.5 8-13 2 Mariana Mesisol O/M pt/pt hydric 5 14.0 12-16 2 Muskeg Mesisol O pt/pt hydric 5 11.0 10-12 2

(a) F = fluvial; M = morainal; O = organic; pt = peat (b) C = clay; L = loam; S = sand; Si = silt Note: One soil profile rating for each soil type; Number of Plots refers to the number of vegetation plots on that soil for that ecosite. Source: BOVAR Environmental (1996) updated to match Leskiw 1998 and Table B.3 Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX B Page B-3 Prepared By: The Oil Sands Vegetation Reclamation Committee

Table B.3 Selected Site Data Relating Tree Growth and Soils at the Proposed Aurora Mine Area

Ecosite Series Species % DBH Height Age Site Soil Soil Moisture cover (cm) (m) (br.hgt) Index Drainage Perviousness Regime lichen Fort Pj 25 17.5 13.7 46 14 well moderate subxeric Pj 60 17.5 11.6 65 10 rapid rapid xeric Pj (u) 5 10 10.2 32 14 rapid rapid xeric Aw (u) 5 9.5 7.8 29 13 moderate moderate subxeric Pj 70 18 14.8 55 14 moderate moderate subxeric Pj 15 15.1 12.8 41 15 rapid rapid xeric Pj 30 11.7 13.2 42 15 well rapid subxeric Mildred Pj 35 22.7 17.5 58 16 well moderate mesic Pj 40 18.7 14.2 50 14 moderate moderate submesic blueberry Dalkin Aw 7 18.5 11.9 80 8 well moderate submesic Aw (u) 5 9.5 10.5 32 15 well moderate submesic Pj 30 22 11.3 56 11 well moderate submesic Sb (u) 5 15.5 9.4 39 11 well moderate submesic Sw 20 18 11.9 48 12 well moderate submesic Fort Aw (u) 10 9.5 9.3 52 9 well moderate mesic Bw (u) 5 6 6.8 48 7 well moderate mesic Pj 10 28 11.4 46 12 well moderate mesic Sw 20 19 12.1 81 7 well moderate mesic lowbush Fort Aw 60 20.7 19.8 52 20 well moderate submesic cranberry Aw 70 15.8 23 57 22 imperfect mod-slow hygric Mildred Aw 60 18.2 21 56 20 well moderate submesic Bw 1 12.8 12.5 39 15 well moderate submesic Aw 60 13.3 18.9 32 24 moderate moderate mesic Aw 40 14.7 19.2 40 22 well moderate mesic Steepbank Aw 86 19 21.6 104 14 well moderate mesic Aw (u) 40 7.5 10.8 23 19 poor slow hygric Bw (u) 5 9 11 38 14 poor slow hygric Algar Aw 35 9 9.5 52 10 imperfect moderate hygric Sb 25 12.3 9.5 35 12 imperfect moderate hygric Dover Aw 40 22.6 20.1 70 16 moderate moderate subhygric Sw 20 31.9 26.2 72 21 moderate moderate subhygric Aw 30 12 11.6 28 17 well moderate mesic Sw 20 15.7 13.7 39 16 well moderate mesic horsetail Bitumount Bw 60 13 12.6 86 8 imperfect slow hygric Lt 5 12.5 9.8 91 13 imperfect slow hygric Sb 5 15.5 9.3 75 7 imperfect slow hygric Sw 10 11.5 11.4 54 10 imperfect slow hygric Labrador Livock Pj 5 15.2 11.9 44 13 well rapid submesic tea Sb 10 11.2 7.9 41 9 well rapid submesic bog Algar Sb 10 15 13.1 64 11 very poor slow subhydric Bitumount Sb 15 10 6.5 71 5 poor slow subhydric Pb 5 10.2 8 51 5 imperfect slow subhygric Sb 50 10 7.9 52 8 imperfect moderate subhygric Hartley Sb 5 7 6.3 42 8 poor slow subhydric Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX B Page B-4 Prepared By: The Oil Sands Vegetation Reclamation Committee

Table B.3 (cont’d)

Ecosite Series Species % DBH Height Age Site Soil Soil Moisture cover (cm) (m) (br.hgt) Index Drainage Perviousness Regime Sb 2 6.3 6.5 23 11 poor slow subhydric Sb 10 17 13 128 7 poor slow hygric Sb 8 7.8 6.6 40 8 poor slow hygric Sb 25 11.5 7.2 50 7 imperfect moderate subhygric McLelland Sb 10 8.1 18.3 35 23 poor slow subhydric Sb 25 10.8 9.5 42 11 poor slow hygric Sb 10 11 6.3 47 7 very poor slow subhydric Sb 25 13 9.6 56 9 poor slow subhygric Muskeg Sb 40 19 15.5 60 14 imperfect slow hygric Lt 2 7.5 5.1 25 8 very poor slow subhydric Steepbank Sb 25 77 9 41 11 verypoor slow subhygric Sb 20 19.4 12.3 54 12 poor slow subhydric Sb 2 8 5.4 56 5 imperfect slow hygric Sb 10 18 16.7 137 9 poor slow hygric Sb (u) 25 12.5 9.7 119 5 poor slow hygric poor fen Bitumount Lt 5 11 8.3 41 14 poor slow subhydrc Lt 5 11 7.9 74 12 poor slow subhydric Sb (u) 10 9 5.4 45 6 poor slow subhydric Hartley Lt 5 13 13.7 39 22 imperfect slow hygric Sb 5 9 7 44 8 imperfect slow hygric Lt 20 18.4 12.3 40 20 poor slow hygric Sb (u) 8 9.5 6.4 32 9 poor slow hygric Lt 10 22.5 12 62 17 imperfect slow subhydric Sb 5 18.9 11.5 51 11 imperfect slow subhydric rich fen Hartley Lt 7 12 6 179 8 very poor slow hydric Lt 5 17 7.5 42 13 poor slow hydric Mariana Lt 15 12 10.7 56 16 poor slow hydric Lt 10 17.5 8.3 82 12 very poor slow hydric Muskeg Lt 20 13.5 5.8 55 10 very poor slow hydric Sb 5 8.5 5.4 31 12 very poor slow subhydric Source: Bovar Environmental 1996. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX B Page B-5 Prepared By: The Oil Sands Vegetation Reclamation Committee

B1.3 SHELL LEASE 13

A detailed soil survey of the Shell Lease 13 project area was conducted in August, 1997. A summary of the soil and site index information is presented in Table B.4. Figure B.1 shows the relationship of soil capability and site index at 50 years based primarily on white spruce and some aspen. Trees were measured and aged at several soil inspection sites representing a cross-section of soil conditions.

B1.4 MILLENNIUM MINE, SUNCOR

A detailed soil survey of the Millennium project area was conducted in September, 1997. A summary of the soil and site index information is presented in Table B.5. Figure B.2 shows the relationship between soil capability and site index at 50 years. Species examined included mainly white spruce, black spruce and some jack pine. Trees were measured and aged at soil inspection sites.

B1.5 PERMANENT SAMPLE PLOT STUDY IN CENTRAL ALBERTA

A study of soil capability as related to site index at permanent sample plots was conducted in the agricultural fringe area in Central Alberta and the Peace River region (Leskiw et al. 1997). In this study, individual soil ratings and site index values were determined for the same points on mostly “good” plots. An adjustment was made for open jack pine stands (multiply site index by 0.60) to lower the site index values to be more representative of mean annual increment in open jack pine stands. These measurements reflect potential capability under natural growing conditions. Table B.6 provides a summary of soil and site index information, while Figure B.3 displays the resultant correlation.

B2.0 SITE INDICES

Actual site indices for ecosite phases that occur in the Syncrude Aurora Mine and the Suncor Millennium Mine are presented in Table B.7.

B3.0 LANDSCAPE AND SOIL DIVERSITY

The natural and reclaimed landscapes and soils each have a different range of conditions and diversity (Tables B.8 and B.9).

Overall, there is an opportunity to establish a similar or equivalent range of diversity in landscapes and soils, after reclamation compared to original conditions, though the specific types of landscapes and soils will likely differ. This can be accomplished through careful planning and implementation. Subsequent monitoring is required to confirm that targeted conditions are attained. Guidelines for Reclamation to Forest Vegetation October,1998 In the Alberta Oil Sands Region – APPENDIX B Page B-6 Prepared By: The Oil Sands Vegetation Reclamation Committee

Table B.4 Soils and Average Site Index by Tree Species for 60 Sites in the Shell Lease 13 Project Area

SUMMARY DESCRIPTION AVERAGE SITE INDEX BY TREE SPECIES AT 50 YEARS Capability Class LFH or Peat White Jack Black Moisture Surface/ depth (cm); Aspen Spruce Pine Spruce Overall Site Index (n) Subgroup Regime Subsurface median (range) Aw (n) Sw (n) Pj (n) Sb (n)

1 Gleyed Subhygric SL / SL 15 (4-20) - (0) 22 (3) - (0) - (0) 22 (3) Luvisol

2 Brunisol Mesic Hygric SL, LS, SiL, / L, 13 (4-35) 22 (2) 19 (10) 10 (2) 18 (1) 18 (15) Gleysol (aerated) SL, LS, residual

3 Brunisol Submesic Mesic LS, / LS, S, CL 4 (1-10) 18 (10) 18 (6) 10 (6) - (0) 16 (22) Luvisol

4 Brunisol Submesic Hygric S, LS, / S, LS, 7 (2-17) - (0) 13 (4) 10 (9) 11 (4) 11 (17) Gleysol (reduced) CL

5 Mesisol Subhydric pt / pt 72 (11-160) - (0) - (0) - (0) 11 (3) 11 (3)

1 C = clay; L = loam; S = sand; Si = silt; pt = peat Note: n = number of sites. Guidelines for Reclamation to Forest Vegetation October,1998 In the Alberta Oil Sands Region – APPENDIX B Page B-7 Prepared By: The Oil Sands Vegetation Reclamation Committee

Table B.5 Soils and Average Site Index by Tree Species for 57 Sites in the Suncor Millenium Project Area

1 Gleyed Subhygric SL / CL 8 (6-20) 19 (3) 19 (4) - (0) - (0) 19 (7) Luvisol

2 Luvisol Hygric Mesic SL, SiL / SCL, 10 (5-20) 17 (8) 17 (8) - (0) 12 (1) 17 (17) Gleysol (aerated) CL

3 Luvisol Mesic LS, / SCL, 23 (5-40) 17 (1) 17 (1) - (0) - (0) 17 (2)

4 Gleysol Hygric (reduced) pt, SL, / SCL, 20 (10-50) - (0) - (0) 8 (1) 10 (14) 10 (15) CL

5 Mesisol Subhydric pt / pt 55 (0-120) - (0) 8 (1) 10 (1) 9 (14) 9 (16)

1 C = clay; L = loam; S = sand; Si = silt; pt = peat Note: n = number of sites Guidelines for Reclamation to Forest Vegetation October,1998 In the Alberta Oil Sands Region – APPENDIX B Page B-8 Prepared By: The Oil Sands Vegetation Reclamation Committee

Table B.6 Soils and Average Site Index by Tree Species for 80 Permanent Sample Plots in Central Alberta

SUMMARY DESCRIPTION AVERAGE SITE INDEX BY TREE SPECIES AT 50 YEARS Capability Class LFH or Peat White Jack Lodgepol Black Moisture Surface/ depth (cm); Aspen Spruce Pine e Spruce Overall Site Subgroup Regime Subsurface median (range) Aw (n) Sw (n) Pj (n) Pine Sb (n) Index (n) Pl (n)

1 Gleyed Subhygric SL, L, SiL / CL, 5(5-30) 20.8 (7) 19.8 (9) 24.8 (1) 24.0 (1) 24.3 (1) 20.9 (19) Luvisol SCL (a)

2 Luvisol Mesic SiL, SL, L / CL, 8(5-20) 18.8 (53) 18.9 (25) 15.2 (2) 16.1 (30) 13.8 (7) 17.6 (117) SCL, C

3 Luvisol Submesic-mesic SiL, LS, SL / SiL, 5(0-5) 16.5 (2) 17.6 (4) - (0) 12.5 (2) - (0) 16.0 (8) Brunisol SL, CL

4 Brunisol Xeric-subxeric LS, S / S or 5(0-30) - (0) - (0) 9.5 (19) - (0) - (0) 9.5 (19) Gleysol (sandy) or SL,SCL/SL, - (0) 14.4 (4) - (0) - (0) 8.0 (2) 12.3 (6) hygric (finer) SCL

5 Mesisol Subhydric peat / peat 70(45-100) - (0) - (0) - (0) - (0) 5.7 (3) 5.7 (3)

(a) C = clay; L = loam; S = sand; Si = silt; pt = peat Note: n= number of sites. Guidelines for Reclamation to Forest Vegetation October,1998 In the Alberta Oil Sands Region – APPENDIX B Page B-9 Prepared By: The Oil Sands Vegetation Reclamation Committee

Shell

y = 0.1813x + 5.6859 2 25 R = 0.4189

15 Site Index (m) 10

0 0 5 20 4 40 3 60 2 80 1 100 Soil Class

Figure B.1 Relationship Between Site Index and Soil Index for Sites in the Shell Lease 13 Project Area Guidelines for Reclamation to Forest Vegetation October,1998 In the Alberta Oil Sands Region – APPENDIX B Page B-10 Prepared By: The Oil Sands Vegetation Reclamation Committee

Suncor Millenium

y = 0.1425x + 6.0131 25 R2 = 0.7529

Site Index (m) 10

0 0 5 20 4 40 3 60 2 80 1 100 Soil Class

Figure B.2 Relationship Between Site Index and Soil Index for Sites in the Suncor Millenium Project Area Guidelines for Reclamation to Forest Vegetation October,1998 In the Alberta Oil Sands Region – APPENDIX B Page B-11 Prepared By: The Oil Sands Vegetation Reclamation Committee

PSP Values

25.0

20.0

15.0

y = 0.188x + 4.9352 10.0 R2 = 0.7251 Site Index (m)

5.0

0.0 0 5 20 4 40 3 60 2 80 1 100 Soil Class

Figure B.3 Relationship Between Site Index and Soil Rating for 80 Permanent Sample Plots in Central Alberta Guidelines for Reclamation to Forest Vegetation October,1998 In the Alberta Oil Sands Region – APPENDIX B Page B-12 Prepared By: The Oil Sands Vegetation Reclamation Committee

Table B.7 Site Index (at 50 years) of the Ecosite Phases from Beckingham and Archibald (1996) and Those Occurring in the Aurora Mine (the latter is based on ground-truthed plots) Beckingham and Archibald (1996) Aurora Mine(a) Number Site Index at 50 Number SI Range in Std. Ecosites and Ecosite Phases (n) Years (n) (at 50 yrs) SI Dev. Variance a jack pine/lichen 62 13.4 20 14 10-18 1.96 3.84 b spen 56 15.8 6 13.2 11-16 3.06 9.37 jack pine 29 14.3 8 14.8 12-19 3.24 10.50 white spruce 28 17.5 6 14.2 7-20 4.36 18.97 b1 jack pine-aspen/blueberry aspen N/A N/A 3 14 11-16 2.65 7.00 jack pine N/A N/A 3 15 12-19 3.51 12.33 b3 aspen-white spruce/blueberry aspen N/A N/A 1 14 -- 0 0 white spruce N/A N/A 1 15 -- 0 0 b4 white spruce-jack pine/blueberry white spruce N/A N/A 5 14 7-20 4.85 23.50 jack pine N/A N/A 5 14 14-19 3.44 11.8 c jack pine-black spruce/Labrador tea jack pine 64 14.3 1 13 -- 0 0 black spruce 20 11.5 1 9 -- 0 0 d aspen 397 18.2 42 16.2 6-28 4.31 18.58 white spruce 502 16.8 24 16.3 9-26 4.14 17.17 d1 aspen/low-bush cranberry N/A N/A 26 17 6-28 5.12 26.25 d2 aspen-white spruce/low-bush cranberry aspen N/A N/A 16 16 10-19 2.55 6.50 white spruce N/A N/A 15 15 9-21 3.22 10.37 d3 white spruce/low-bush cranberry 7 19 13-26 4.75 22.57 e white spruce 175 17.8 N/A N/A N/A N/A N/A balsam poplar 38 19.7 N/A N/A N/A N/A N/A aspen 58 21.4 N/A N/A N/A N/A N/A f white spruce 175 16.4 2 13 10-16 4.24 18.00 balsam poplar 7 17.8 1 8 -- 0 0 aspen 12 19.8 1 11.0 -- 0 0 f1 balsam poplar-aspen (white birch)/horsetail white birch N/A N/A 3 8 7-9 1.15 1.33 f2 balsam poplar-white spruce/horsetail balsam poplar N/A N/A 1 8 -- 0 0 Guidelines for Reclamation to Forest Vegetation October,1998 In the Alberta Oil Sands Region – APPENDIX B Page B-13 Prepared By: The Oil Sands Vegetation Reclamation Committee

Beckingham and Archibald (1996) Aurora Mine(a) Number Site Index at 50 Number SI Range in Std. Ecosites and Ecosite Phases (n) Years (n) (at 50 yrs) SI Dev. Variance white spruce N/A N/A 2 13 10-16 4.24 18.00 g black spruce-jack pine/Labrador tea black spruce 21 9.9 3 8 7-9 1.00 1.00 jack pine 20 11.7 3 10 8-13 2.65 7.00 h white spruce-black spruce/Labrador tea white spruce 35 12.9 2 8 6-9 2.12 4.5 black spruce 15 9.5 1 7 -- 0 0 i1 treed bog (black spruce) 32 9.8 31 8 3-23 3.82 14.58 j1 treed poor fen (black spruce and larch) black spruce 9 10.4 9 8 5-11 1.94 3.78 larch 11 8.3 9 15 10-22 4.15 17.25 k1 treed rich fen (larch) 5 7.3 18 13 8-17 2.59 6.68 (a) Site Index values presented here are the average of the primary tree species occurring in each polygon. This is meant to provide an approximate level of overall ecosite phase productivity. Guidelines for Reclamation to Forest Vegetation October,1998 In the Alberta Oil Sands Region – APPENDIX B Page B-14 Prepared By: The Oil Sands Vegetation Reclamation Committee

Table B.8 Landscape and Soil Diversity in Natural Lands

LANDSCAPE SOIL

Geological Surface Expression Capability Surface Expression Subgroup Moisture Regime Capabilit Material y

Eolian Dunes, 5 to 15 m relief. 1 Dunes, 5 to 15 m relief. Dystric Brunisols Subxeric, Submesic 4 Sands Undulating plains, <15% slopes, <5 m Undulating plains, <15% slopes, <5 Gleyed Dystric Brunisols Subhygric 3 relief. m relief. Orthic and Peaty Gleysols Hygric a 4 No aspect effects. No aspect effects. Fluvial (recent) Floodplains, terraces, dissected by 1 Floodplains, terraces, dissected by Orthic Regosols Mesic 2 sandy, silty, meander meander Gleyed Regosols Subhygric 1 gravelly Scars, <5% slopes, <5 m relief. scars, <5% slopes, <5 m relief. Orthic and Peaty Gleysols Hygric a 4 No aspect effects. No aspect effects. Glacio-fluvial Meltwater channels, gently undulating, 1 Meltwater channels, gently Dystric & Eutric Brunisols Submesic, mesic 3, 4 sandy, gravelly ridged, undulating, ridged, Gleyed Brunisols Subhygric 2, 3 <9% slopes, <5 m relief. <9% slopes, <5 m relief. Orthic and Peaty Gleysols Hygric a 4 No aspect effects. No aspect effects. Glacio- Plains, <5% slopes, <2 m relief. 1 Plains, <5% slopes, <2 m relief. Orthic Gray Luvisols Mesic 2 lacustrine No aspect effects. No aspect effects. Gleyed Gray Luvisols Subhygric 1 clayey Orthic and Peaty Gleysols Hygric a 4 Morainal Undulating to rolling 5 to 30% slopes, 1 to 2 Undulating to rolling 5 to 30% Orthic Gray Luvisols Mesic 2 loamy to clayey 5 to 50 m relief. slopes, Gleyed Gray Luvisols Subhygric 1 Minimal aspect effects. 5 to 50 m relief. Orthic and Peaty Gleysols Hygric a 4 Minimal aspect effects. Rough Broken Steeply sloping river banks, 16-70% 2 to 5 Steeply sloping river banks, 16- Luvisols, Brunisols, Submesic, mesic 3, 4 Variable slopes, 70% slopes, Regosols 5 to 100 m relief. Unstable. 5 to 100 m relief. Unstable. Gleyed Soils Subhygric 2, 3 Aspect effects. Aspect effects. Gleysols Hygric a 4 Organic Bogs - level, high water table. 1 Bogs - level, high water table. Typic and Terric Mesisols Subhydric 5 fibric, mesic and Fibrisols and humic Fens - level, high water table 1 Fens - level, high water table Typic and Terric Mesisols Subhydric 5 peats and Humisols

a Hygric aerated is Class 2 or 3, reduced is Class 4. Source: Leskiw and Moskal 1997a,b. Guidelines for Reclamation to Forest Vegetation October,1998 In the Alberta Oil Sands Region – APPENDIX B Page B-15 Prepared By: The Oil Sands Vegetation Reclamation Committee

Table B.9 Landscape and Soil Diversity on Reclaimed Lands

LANDSCAPE SOILS

Reconstruction Material Surface Expression Capability Soil Type Moisture Regime Capability

Overburden with soil Side walls, 16 to 40% slopes, 2 to 3 Peat-mineral topsoil on overburden. Mesic 3 Capping 5 to 100 m relief. Subhygric 2 Aspect effects. Peat-mineral topsoil over subsoil on Mesic 2 overburden Subhygric 1 Nearly level surfaces, <2% slopes, 1 Peat-mineral topsoil on overburden. Mesic 3 <5 m relief. Subhygric 2 No aspect effects. Hygric a 4 Subhydric, Hydric 5 Peat-mineral topsoil over subsoil on Mesic 2 overburden Subhygric 1 Hygric a 4 Subhydric, Hydric 5 Tailings sand Side walls, 16 to 40% slopes, 2 to 3 Peat-mineral topsoil on tailings sand Submesic, Mesic 3 with soil capping 5 to 100 m relief. Subhygric 2 Aspect effects Peat-mineral topsoil over subsoil on Submesic, Mesic 3 tailings sand Subhygric 2 Nearly level surfaces, <2% slopes, 1 Peat-mineral topsoil on tailings sand Mesic 3 <5 m relief. Subhygric 2 No aspect effects. Hygric a 4 Subhydric, Hydric 5 Peat-mineral topsoil over subsoil on Mesic 2 tailings sand Subhygric 1 Hygric a 4 Subhydric, Hydric 5

a Hygric aerated is class 2 or 3, reduced is Class 4. Source: Leskiw and Moskal 1997c,d. Guidelines for Reclamation to Forest Vegetation May, 1998 In the Alberta Oil Sands Region – APPENDIX B Page B-16 Prepared By: The Oil Sands Vegetation Reclamation Committee

B4.0 LITERATURE CITED

BOVAR Environmental. 1996. Environmental Impact Assessment for the Syncrude Canada Limited Aurora Mine. Prepared for Syncrude Canada Limited.

Leskiw, L.A. 1998. Land Capability Classification for Forest Ecosystems in the Oil Sands Region. Revised Edition. Prepared by the Tailings Sands Reclamation Practices Working Group.

Leskiw, L.A. and T.D. Moskal. 1997a. Predisturbance Soils and Forest Soil Capability. Ratings for Syncrude Canada Ltd. Prepared for Syncrude Canada Ltd.

Leskiw, L.A. and T.D. Moskal. 1997b. Predisturbance Soils and Forest Capability Ratings for Suncor Lease 86/17. Prepared for Suncor Inc. Oil Sands Group.

Leskiw, L.A. and T.D. Moskal. 1997c. Reclaimed Soils and Forest Ecosystem Capability of Syncrude Canada Ltd. Prepared for Syncrude Canada Ltd.

Leskiw, L.A. and T.D. Moskal. 1997d. Reclaimed Soils and Forest Ecosystem Capability of Suncor Inc. Oil Sands Group. Prepared for Suncor Inc. Oil Sands Group.

Leskiw, L.A., L.K.M. Esak, L. Waterman, B. Roth and J. Pluth. 1997. Soil Capability Classification for Forestry and an Inventory of Pilot Woodlots. Prepared for Daishowa- Marubeni, Millar Western Industries Ltd. and Weyerhaeuser Canada. APPENDIX C

ACTS, REGULATIONS, POLICIES AND GUIDELINES RELEVANT TO REVEGETATION AND RECLAMATION IN THE OIL SANDS REGION, DATED JULY 1997 Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX C Page C-1 Prepared By: The Oil Sands Vegetation Reclamation Committee

ACTS, REGULATIONS, POLICIES AND GUIDELINES RELEVANT TO REVEGETATION AND RECLAMATION IN THE OIL SANDS REGION, DATED JULY 1997

Prepared by Chris Hale, Alberta Environmental Protection

Acts and Regulations · The Forest Act and associated Regulation - The Timber Management Regulations and, - Regeneration Survey Manual, with the 1997 Deciduous Amendment. · Alberta Environmental Protection and Enhancement Act, and Regulations - The Conservation and Reclamation Authority, and associated Operating Conditions for each Mine. · The Public Lands Act, and Regulations - The Mineral Surface Lease Letter of Authority, and associated Operating Conditions for each Mine. · The Canada Seed Act

Policies and Guidelines, as Authorized by Legislation · Fort McMurray-Athabasca Oil Sands Subregional Integrated Resource Plan. · Recommended Native Grasses and Legumes for Revegetating Disturbed Lands in the Green Area. · Forest Conservation Strategy. · Canadian Framework of Criteria and Indicators for Sustainable Forest Management, and Defining Sustainable Forest Management, A Canadian Approach to Criteria and Indicators, from the Canadian Council of Forest Ministers. · Forest Site Interpretation & Silvicultural Prescription Guide for Alberta. · The Alberta Timber Harvest Planning and Operating Groundrules. · C & R/97-1 Conservation and Reclamation Guidelines for Alberta. · C & R/IL/96-1 Land Capability Classification for Forest Ecosystems in the Oilsands Region. · C & R/95-1 Conservation and Reclamation Code of Practice for Alberta. · Guidelines for the Preparation of Applications and Reports for Coal and Oil Sands Operations., Alberta Land Conservation and Reclamation Council, 1991. APPENDIX D

RESEARCH ON PLANT VARIETIES Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX D Page D-1 Prepared By: The Oil Sands Vegetation Reclamation Committee

RESEARCH ON PLANT VARIETIES

Prepared By: Leanard Barnhardt and Narinder Dhir, Alberta Environmental Protection

D1.0 OVERVIEW

Research on plant varieties provide an opportunity to develop exotic or hybrid stock that may out perform native varieties in certain aspects. To conduct research on exotic or hybrid plant varieties, the proponent must submit the proposed trial, with a complete description of how it will be done and what is to be proven, to the Land and Forest Service, Forest Management Division, Genetics Section for approval. Research that meets approved design parameters will likely be approved for research.

Exotic, clonal or hybrid plant varieties used in experimental trials will not be accepted for reclamation certification until research proves the acceptability of those plants to the standards set by the Land and Forest Service. A more descriptive listing of the requirements for exotic forest tree species, clonal forestry, hybrid varieties and plant with novel traits is provided below.

D1.1 GENERAL INFORMATION REGARDING EXOTIC, CLONAL OR HYBRID PLANT VARIETIES AND PLANTS WITH NOVEL TRAITS

Exotic Forest Tree Species

· Promising Species - Siberian Larch Operational testing stage (Raivola Siberian Larch more suited for the north) - Scots Pine Research testing phase (Southwestern Siberian seed sources are more promising)

· No exotic tree species is approved for reforestation in Alberta at present. Issues of concern are: - Must not be injurious to forest health, - Insufficient test data, - Long-term insect/disease concerns, - Genetic base population, and - Ecological succession and evolutionary biology of the exotic forest. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX D Page D-2 Prepared By: The Oil Sands Vegetation Reclamation Committee

Clonal Forestry

· Definition: - A forest regeneration method in which clones or genetically identical trees are deployed and tended to produce forest crop.

· Advantage: - Highly specialized and productive varieties can be developed.

· Disadvantages:

- Risks associated with reduced genetic diversity,

- Requires continuing research and development efforts to replenish clones on periodic basis,

- Requires site specific and more intensive silviculture practice, and

- Concerns about the progeny of clonal forest.

Hybrid Varieties

· Definition:

- F1 populations derived by crossing selected clones, varieties, natural populations (or species) that are genetically dissimilar.

· No applicable example in Alberta forestry at present.

Plants With Novel Traits (PNT)

· Definition:

- Plant variety/genotype possessing characteristics that demonstrate neither familiarity nor substantial equivalence to those present in a distinct stable population of a cultivated species of seed in Canada and that have been intentionally selected, created or introduced into a population of that species through specific genetic change.

· Familiarity:

- The knowledge of the characteristics of a plant species and experience with the use of that plant species in Canada.

· Substantial Equivalence:

- The equivalence of a novel trait within a particular plant species, in terms of its specific use and safety to the environment and human health, to those in the Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX D Page D-3 Prepared By: The Oil Sands Vegetation Reclamation Committee

same species, that are in use and generally considered as safe in Canada, based on valid scientific rationale.

Test Criteria and Considerations for PNT

Familiarity: · Species · Trait introduced · Method of introduction · Cultivation practice

Substantial Equivalence: · Altered weediness potential · Gene flow to related species · Altered plant pest potential · Potential impact on non-target organisms · Potential impact on biodiversity APPENDIX E

ECOLOGICAL DIVERSITY MONITORING FRAMEWORK, DRAFT #6, PREPARED FOR THE BIODIVERSITY MONITORING WORKING GROUP (AUGUST 1997) Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX E Page E-1 Prepared By: The Oils Sands Vegetation Reclamation Committee

Ecological Diversity Monitoring Framework

Draft For Discussion (#6) August, 1997

Prepared for the Biodiversity Monitoring Working Group

Richard Schneider R. R. 2, Tofield, AB. T0B-4J0 403-662-4233 [email protected] Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX E Page E-i Prepared By: The Oils Sands Vegetation Reclamation Committee

Table of Contents

SUMMARY...... 1

E1. SCOPE ...... 2

E2. CONTEXT...... 2 E2.1 International...... 2 E2.2 National...... 3 E2.3 Provincial...... 4

E3. RATIONALE...... 5 E3.1 Monitoring As An Audit Process ...... 5 E3.2 Monitoring As Part Of An Adaptive Approach To Resource Management.... 5 E3.3 Monitoring As An Alternative To A Prescriptive Regulatory Framework ...... 5 E3.4 Monitoring To Fulfill National And International Commitments...... 5 E3.5 Monitoring As Part Of Forest Certification...... 6

E4. OBJECTIVES...... 6 E4.1 Monitoring ...... 6 E4.2 Research...... 6

E5. FRAMEWORK ...... 7 E5.1 Introduction...... 7 E5.2 Remote Sensing...... 8 E5.3 Large-Scale Sampling Network ...... 9 E5.3.1 General Sampling Design...... 9 E5.3.2 General Sampling Protocol...... 10 E5.3.3 Sampling Protocol For Aquatic Systems ...... 10 E5.3.4 Sampling Elements...... 11 E5.4 Rare/Endangered Species Program...... 13 E5.5 Research program...... 14

E6. IMPLEMENTATION ...... 14 E6.1 Administration...... 14 E6.2 Data Standards...... 15 E6.3 Funding...... 16 E6.4 Reference Areas...... 16

E7. ACKNOWLEDGMENTS...... 17

E8. LITERATURE CITED...... 18 Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX E Page E-1 Prepared By: The Oils Sands Vegetation Reclamation Committee

SUMMARY

This paper outlines a framework for monitoring ecological diversity within the forested regions of Alberta. The emphasis is on detecting changes associated with industrial use of the forest. The unit of concern is the complete forest ecosystem, including both terrestrial and aquatic habitats.

The initial section of the paper describes the international, national, and provincial context for the framework. The rationale for the framework is then discussed, and the following objectives are listed:

· To detect changes in ecological diversity that exceed the range of natural variation, across a range of spatial and temporal scales.

· To provide an “early warning” of impending irreversible changes.

· To provide reports to the public on the status of ecological diversity in Alberta in a timely and accessible manner.

· To meet Alberta’s national and international commitments for monitoring biodiversity.

· To provide data consistent with the requirements of forest certification programs.

The remainder of the paper outlines the actual framework and topics relevant to its implementation. The framework has four components:

· Remote sensing.

· Large-scale sampling network.

· Endangered/threatened species monitoring program.

· Research program.

The large-scale sampling network is discussed in detail. A systematic network of fixed sample points across the entire forested region of the province is proposed. A preliminary list of elements to be sampled is presented, along with a description of sampling protocols. The need for reference areas (i.e., controls) is discussed, and the Chinchaga watershed, Liege River watershed, and Caribou mountains are proposed as candidate reference areas.

It is proposed that a single agency run the monitoring program, with ongoing direction provided by an administrative committee comprised of representatives from the government, agencies that fund the program, and technical experts. It is suggested that funding for the program be derived from stumpage fees and similar mechanisms from all industries that use the forest. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX E Page E-2 Prepared By: The Oils Sands Vegetation Reclamation Committee

E1. SCOPE

The program for monitoring ecological diversity outlined in this paper is intended to detect changes in ecological diversity within the forested regions of Alberta. The emphasis is on detecting changes associated with industrial use of the forest (e.g., forestry, oil and gas exploration and extraction, and mining). The unit of concern is the complete forest ecosystem, including both terrestrial and aquatic habitats.

E2. CONTEXT

E2.1 International

At the international level there are two initiatives with direct relevance to Canada and Alberta. These are the United Nations Convention on Biodiversity (i.e., Rio Convention; UNEP 1992) and the Montreal Process (i.e., Santiago Declaration (Canadian Forest Service 1995). Canada is a signatory to both documents. The Rio Convention focuses on strategies for conserving biodiversity, and includes explicit requirements for the monitoring of biodiversity (Article 7). The Montreal Process was established in 1994 to develop internationally agreed upon criteria and indicators for the conservation and sustainable management of temperate and boreal forests.

The United States currently spends 650 million dollars per year on environmental monitoring, including the monitoring of forest health (CENR 1997). The nation is in the process of combining all of its various monitoring programs into a single fully integrated framework (CENR 1997). The objectives of the new framework are to “identify environmental and ecosystem trends, relate these trends to their causes and consequences, and predict the outcomes of alternative future scenarios.” (CENR 1997; p. 6). A fundamental premise of the framework is that “no single sampling design can efficiently provide all the information required to evaluate environmental conditions and to guide policy.” (CENR 1997; p. 14). Consequently, the framework has three interrelated levels:

1. Remote sensing,

2. A regional sampling network, and

3. Intensive monitoring/research sites.

Levels one and two are intended to quantify the extent, distribution, condition, and rate of change of specific environmental variables over the entire land base. Level three is intended to determine the causes of changes noted in levels one and two and to develop and test predictive models. Integration, through the use of a standard set of core variables, standards for data collection, research, and modeling, is a key feature of the framework. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX E Page E-3 Prepared By: The Oils Sands Vegetation Reclamation Committee

The United Kingdom has established an Environmental Change Network to:

1. Obtain comparable long-term data sets for major environmental variables that can be used to distinguish human-induced change from natural variation,

2. Identify and quantify environmental changes associated with human activities, and

3. Give warning of undesirable effects (National Environmental Research Council 1994).

The network currently consists of 50 sites throughout the United Kingdom, and is linked to integrated surveys that combine remote sensing and ground-based sampling. The network is operated by a consortium of 15 sponsoring organizations coordinated by the National Environmental Research Council.

Other notable examples of international environmental monitoring programs include the Global Terrestrial Observing System (UNESCO 1997) and the United Nations Economic Commission for Europe Integrated Monitoring Program (UNECE 1993). Most environmental monitoring programs were initiated in response to air and water pollution issues; however, many have since expanded in scope to incorporate the monitoring of biodiversity as well.

E2.2 National

Canada has made commitments relating to the monitoring of biodiversity in the National Forest Strategy (CCFM 1992), the Canadian Biodiversity Strategy (Environment Canada 1995), and the Rio Convention (UNEP 1992). The Canadian Biodiversity Strategy states: “Monitoring programs are required to detect and measure changes in biodiversity, to better understand functional linkages in ecosystems, and to evaluate the success or failure of biodiversity conservation and sustainable use policies and programs.” (Environment Canada 1995). In response to these commitments, the Canadian Council of Forest Ministers (CCFM) has produced a set of criteria and indicators for the sustainable management of Canada’s forests, including indicators for biodiversity and ecosystem function (Canadian Council of Forest Ministers 1995).

There are currently three main programs involved in monitoring biodiversity at the national level in Canada. The first is Environment Canada’s National Environmental Indicators Program (EC 1997). This program monitors national-level indicators of sustainable forest management, including timber harvest levels, natural disturbance trends, and regeneration after harvest. Indicators of biodiversity are planned, but have not yet been implemented (EC 1997). The second national initiative is an annual report produced by the Canadian Forest Service entitled The State of Canada’s Forests (Canadian Forest Service 1996). This report covers a range of issues related to Canadian forests and includes data on some of the indicators established by the CCFM. The final initiative, begun in 1994, is the Environmental Monitoring and Assessment Network (EMAN) (EMCO 1996). The goal of this initiative is to understand what changes are occurring in Canadian ecosystems and why. To answer these questions long-term multidisciplinary information is being gathered within a research context at a network of 78 sites across the country. The research sites are all independently funded and managed and the role of EMAN is primarily one of coordination and integration.

Monitoring is also part of the mandate of the national parks (Woodley 1997) and the Model Forest Program (CFS 1997); however, only some sites have implementing formal monitoring programs (e.g., Yoho (YNP 1997) and Kejimkujik (Drysdale and Beattie 1995)). Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX E Page E-4 Prepared By: The Oils Sands Vegetation Reclamation Committee

Two forest certification programs are being advocated in Canada, one by the Canadian Standards Association (Canadian Standards Association 1996) and the other by the Forest Stewardship Council (Elliott and Hackman 1996). They are not intended to serve as national monitoring programs; however, monitoring is an integral component of both programs. The indicators relating to biodiversity in these programs are, in general, consistent with the criteria and indicators proposed by the CCFM (Canadian Council of Forest Ministers 1995).

E2.3 Provincial

The situation with respect to the monitoring of biodiversity in Alberta is atypical relative to most other jurisdictions in that the government is in the process of transferring much of the responsibility for monitoring to industry. Several forestry companies have produced discussion papers concerning biomonitoring programs they have under consideration (Bonar 1995; Doyon and Duinker 1997; Gilmore 1997; Rabik and Larson 1997).

The government will continue to play a role in monitoring; however, the nature of that role is still being determined (H. Stelfox, pers. comm.). Currently, the Wildlife Management Division of Alberta Environmental Protection monitors the status of wildlife (primarily non-fish vertebrate species) for planning and management purposes. This activity also supports the department's policy commitment to publicly report on the status of wildlife every five years. A Biodiversity Species Observation Database is used to store and evaluate observations of species from various sources throughout Alberta. Particular attention is given to the collection and analysis of population data for species that are, or may be, at risk of permanent decline. The Biodiversity Species Observation Database and other data sources are used to help determine more formal threatened and endangered species designations for Alberta and Canada. The Alberta Natural Heritage Information Centre, managed by the Recreation and Protected Areas Division of Alberta Environmental Protection, tracks occurrences of rare features of provincial significance, primarily plant and animal species, but also geological and cultural features.

In the non-government sector, the Federation of Alberta Naturalists maintains provincial natural history databases which include: (1) the Alberta Breeding Bird Atlas, (2) Spring (May) Species Counts for plants & birds, (3) Alberta Birdlist observations and (4) Christmas Bird Count results (as reported in the Alberta Naturalist). These data are primarily from amateur naturalist sources.

The Foothills Model Forest (CFS 1997) has expressed an interest in becoming involved in the monitoring of biodiversity (e.g., through the development of this framework), as have Jasper National Park and Wood Buffalo National Park (P. Achuff, pers. comm.; N. Stolle, pers. comm.). There are three EMAN sites in the forested regions of Alberta: the Meanook Biological Research Station, the Terrestrial and Riparian Organisms, Lakes and Streams (TROLS) study area, and South Waterton Biosphere Reserve. Additional ecological research, much of it industry sponsored, is being conducted at sites that are not part of EMAN. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX E Page E-5 Prepared By: The Oils Sands Vegetation Reclamation Committee

E3. RATIONALE

E3.1 Monitoring As An Audit Process

The maintenance of ecological diversity has become a central goal of forest management in Alberta (Alberta Forest Conservation Strategy Steering Committee 1997) and across North America (Hunter 1991; Everett 1993; Grumbine 1994). Consequently, the government (on behalf of the public) and industry (as part of its internal audit process) must monitor forest ecosystems to ascertain whether or not the goal of maintaining ecological diversity is being met as resources are being extracted (Noss 1990; Christensen et al. 1996; Alberta Forest Conservation Strategy Steering Committee 1997).

E3.2 Monitoring As Part Of An Adaptive Approach To Resource Management

Because of gaps in our understanding of ecological systems, among other factors, the outcome of resource management strategies is generally uncertain (Hunter 1991; Walters and Holling 1990). Therefore, management procedures must be continually modified in response to feedback from the system in order to achieve management objectives (Kessler et al. 1992; Everett 1993). The implementation of such an adaptive approach to management requires a monitoring program to provide the necessary feedback and a research program to determine the cause of any problems that are noted and to provide options for mitigation (Ringold et al. 1996). The monitoring program should incorporate a predictive element so that irreversible deleterious changes, such as the extirpation or extinction of a species, can be prevented while it is still possible (Ligon and Stacey 1996; CENR 1997).

E3.3 Monitoring As An Alternative To A Prescriptive Regulatory Framework

Without a rigorous monitoring program in place it is unlikely that public acceptance of a flexible approach to forest management, such as ecosystem management, would be maintained (Grumbine 1994). The imposition of a rigid regulatory framework would become increasingly more likely.

E3.4 Monitoring To Fulfill National And International Commitments

Monitoring is required to fulfill Alberta’s national and international commitments, as described in Section 2. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX E Page E-6 Prepared By: The Oils Sands Vegetation Reclamation Committee

E3.5 Monitoring As Part Of Forest Certification

Monitoring is a prerequisite for forestry companies seeking to ensure market security through certification programs (Elliott and Hackman 1996). Monitoring is an integral component of both the Canadian Standards Association (Canadian Standards Association 1996) and the Forest Stewardship Council certification programs (Elliott and Hackman 1996).

E4. OBJECTIVES

E4.1 Monitoring

The objectives of the monitoring program are to:

· Detect changes in ecological diversity that exceed the range of natural variation, across a range of spatial and temporal scales.

· Provide an “early warning” of impending irreversible changes.

· Provide reports to the public on the status of ecological diversity in Alberta in a timely and accessible manner.

· Meet Alberta’s national and international commitments for monitoring biodiversity.

· Provide data consistent with the requirements of forest certification programs.

E4.2 Research

Research has an important adjunct role within the context of the monitoring framework. The objectives of the research component are to:

· Support the implementation of the monitoring program by providing insight into appropriate sampling design, sampling methodology, species selection, statistical analyses, and other related issues.

· Determine the cause of any changes in ecological diversity that are observed. This may entail short-term studies in response to specific problems, but should also include long-term research into ecological processes to provide a broader foundation for understanding changes.

· Provide management options for the mitigation of any deleterious changes that are observed.

· Develop the capability for avoiding deleterious changes by constructing predictive models based on research findings and data collected through the monitoring program. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX E Page E-7 Prepared By: The Oils Sands Vegetation Reclamation Committee

E5. FRAMEWORK

E5.1 Introduction

Ecological systems can be characterized by three key attributes: composition, structure, and function (Noss 1990). Composition refers to the constituent parts of the ecosystem, both biotic and abiotic. Species are typically used to define the unique elements of the biotic component; however, the existence of local and regional variation in genotype must also be recognized (Noss 1990). Ecosystem structure refers to the spatial organization of the constituent parts of the system, including large-scale patterns. Ecosystem function refers to how the parts interact with each other. Variety and variability within these key attributes, across a range of spatial and temporal scales, constitute ecological diversity (Naiman et al. 1993).

The key objective of the monitoring framework is to detect changes in ecological diversity. However, because the forest landscape is dynamic it is not change per se that must be detected, but changes that exceed the range of natural variability (Swanson et al. 1993; Frelich and Reich 1995; Christensen et al. 1996; Hecnar and M’Closkey 1996). This implies that the monitoring framework must not only provide data for determining the mean values of ecological attributes, but for characterizing variances as well. Furthermore, the “base case” that is used as a reference when making comparisons (see below) will have to be well characterized in terms of the expected range of natural variability.

The issue of scale is critical to the assessments that are made. The range of natural variation is scale-dependent (both temporally and spatially), so measurements must be made across multiple scales to fully quantify it (Lord 1990; Noss 1990; Knopf and Samson 1994; Steen et al. 1996). The impact of industrial activity is also scale-dependent, and assessments made at small scales cannot be directly extrapolated to larger scales. For example, the cumulative impact of many local changes may combine synergistically to cause unexpected and significant changes at the regional level, or they may simply blend into the background of natural variation. Only integrative measurements taken at large scales will be able to discern the true outcome (Bohning-Gaewse et al. 1993; Brown et al. 1995; Hecnar and M’Closkey 1996).

Two types of “base case” (i.e., control) are available as a reference for the monitoring program, each with its own set of benefits and drawbacks. The first is the industrial land base itself. Given that large-scale forestry is a relatively recent phenomenon in Alberta, excepting the Eastern Slopes, most sampling will initially occur in unharvested forest patches. Consequently, it will be possible to gain an understanding of the initial state of the system, including an estimate of the range of natural variation, based on the initial years of sampling and reference to historical inventory data. Such a control is desirable because of its direct spatial correspondence to the treatment area. The drawback of this control is that it only provides a “snap-shot” of the system at a specific period of time. Eventually, most of the sample points in the industrial land base will be harvested or otherwise impacted. The indirect effects of regional forest harvesting may affect sample points in an even shorter period of time.

The second type of “base case” is a contemporary control based on a system of fixed reference (benchmark) areas and parks. Contemporary controls will increase in importance, relative to the static controls, as climate change and the establishment of roads begin to exert a significant influence on the landscape and as different types of data are required because of changing societal priorities (Kurz et al. 1995). The drawbacks of this type of control are that they are spatially separated from the treatment areas and that they all have been impacted to some extent by roads, fire suppression, and other human influences. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX E Page E-8 Prepared By: The Oils Sands Vegetation Reclamation Committee

Multiple strategies for collecting data will be necessary to address all of the objectives of the monitoring framework. The following are the key components:

· Remote sensing.

· Large-scale sampling network.

· Endangered/threatened species monitoring program.

· Research program.

E5.2 Remote Sensing

Conservation ecologists contend that biota are often adapted to structural attributes of forest landscapes such as the size, age, juxtaposition, and interspersion of forest patches (Hunter 1993; Bunnell 1995). If a relationship between the viability of biotic populations and patch metrics does exist, then changes to landscape patterns through industrial use of the forest may lead to altered biodiversity (Hunter 1993; Bunnell 1995). For this reason the long-term monitoring of landscape metrics should be a key component of the monitoring framework.

Data collected through remote sensing technologies (both satellite and aircraft) are well suited to measuring spatial metrics dealing with patch size, shape, and general composition. They provide an opportunity for monitoring at the largest spatial scales without the need for subsampling, though sampling at these spatial scales generally precludes fine-scale measures (e.g., the composition or abundance of animal species). Once these photogrametric data are processed and housed within a Geographical Information System (GIS), it becomes possible to characterize the forest landscape using a host of metrics such as provided by FragStats (McGarigal and Marks 1994).

Another key role of remote sensing is to track changes in the supply of vegetative communities, particularly those that are rare or are likely to be lost through industrial use of the forest (e.g., older age classes). For most forested regions, fine-scale classification of vegetative communities is currently based on Alberta Vegetation Inventory (AVI) polygons (DEP 1994a) but eventually it may be possible to use an ecological classification system (e.g., ecosite phase; DEP 1994b). The mean area of each community type, and the temporal variability of this mean, across a range of spatial scales, are the key summary measures to be made. The spatial distribution of the community types should also be described. A good example of the utility of AVI-type inventories is for the detection of the unmixing of hardwood/softwood species that may be caused by certain harvesting and silvicultural practices.

Many of the aforementioned analyses are already done by forestry companies, but there is a need for the analyses and reporting to be done in a uniform and consistent manner so that meaningful comparisons across space and time can be made. The data should be updated on a regular (5-10 year) basis so that an analysis of trends can be done. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX E Page E-9 Prepared By: The Oils Sands Vegetation Reclamation Committee

E5.3 Large-Scale Sampling Network

A large-scale sampling network will be required for ground-truthing the measurements made through the remote sensing program and for collecting various types of data that cannot be collected remotely. Because this component of the monitoring framework has received little attention to date, and has a great need for an integrated design, it will be discussed in detail below.

E5.3.1 General Sampling Design

The level of detail of the sampling network will necessarily be fairly coarse given the immense spatial scale involved and limitations on the financial resources available. There will be a tradeoff between the number of sample points that can be established and the amount of data that can be collected at each point.

A sampling design based on the systematic placement of fixed sample points is arguably the best approach to take, primarily because of its flexibility. With this approach, stratification takes place at the analysis stage, instead of prior to sampling. Therefore, regardless of the landscape unit that may be of interest (e.g., Natural Subregion, forest management area, etc.), or the type of industrial impact being investigated (forest harvesting, oil and gas exploration, etc.), the distribution of sample points will always be appropriate. Given the long time frame of the monitoring program, maintaining options is a major consideration. Experience has shown that our delineation of landscape units changes continually over time, as new data are gathered, new questions arise, and new management designations are applied (Department of Environmental Protection 1994a,b). From a logistical perspective, a systematic design is also efficient to implement, as it is not necessary to spend time finding sample points that meet specific criteria. Finally, data arising from a systematic design lends themselves to wide applicability for future scientific research, the nature of which we may not even anticipate at the present time.

The major drawback of a systematic sampling design is that rare vegetative communities will not be well represented. It should be possible to achieve a sample size that will adequately represent the major successional trajectories of forests in the province, but rarer stages will undoubtedly be missed. These communities will therefore need to be emphasized in the remote sensing component of the monitoring framework (Section 5.2). High profile communities could also be monitored through specialized programs, along the lines of endangered species monitoring (Section 5.4).

An alternative approach to the systematic sampling design would be to initially stratify the landscape based on an ecological classification scheme such as the Natural Regions of Alberta (Department of Environmental Protection 1994a,b). In theory this would provide a more balanced sampling of the different types of vegetative communities and may increase the statistical efficiency of the monitoring program (Bourdeau 1953; Scott et al. 1981). However, there are several problems with this approach as well:

· Ecological classifications to the level of ecosite phase have only been completed for a small fraction of the province, and there are no plans to complete the classifications in the near future.

· Many rare vegetative communities, such as those associated with older age classes of forest, are transient phenomena that move around the landscape over time (Frelich and Reich 1995). Furthermore, different vegetative communities can occur on the Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX E Page E-10 Prepared By: The Oils Sands Vegetation Reclamation Committee

same physical site, depending on the disturbance history and other factors (Frelich and Reich 1995). Consequently, as disturbances alter the landscape, the initial scheme of stratification will be lost (given a fixed array of sample points).

· An increase in statistical efficiency may not apply for all analyses. For example, given a limited number sample points overall, a distribution of points based on ecological criteria may decrease the statistical power for contrasting landscapes based on non- ecological criteria such as Forest Management Agreement (FMA) boundaries.

Other sampling designs, such as randomly selected sample points, also exist. The decision on which design to implement should be based on a detailed statistical power analysis, using the best available estimates for the means and variances of the various data measurements (Murtaugh 1996; Thomas 1996). A power analysis will also be required to determine the overall number of sample points in the system. As a ballpark estimate, it would seem that at least one sample point per township will be required to provide adequate coverage. This translates into approximately 3,800 points for the entire Green Zone of the province.

E5.3.2 General Sampling Protocol

Each sample point in the systematic grid of points would be permanently identified and located by a set of Geographic Positioning System (GPS) coordinates. It can be anticipated that access to most points will have to be by helicopter. Sampling would be conducted by a team of two or four personnel using a fixed protocol (Scott et al. 1981; Oliver and Beattie 1993; Oliver and Beattie 1996). The local variance in the attributes that are measured would be estimated by sampling along linear transects that extend out from the central point and in quadrats of fixed size and at a fixed distances from the central point. An areal photograph could be taken of the site to record vegetative structure and local physiography.

Given adequate sampling intensity it should be possible to estimate the mean and variance of sampled attributes for the local region surrounding each sample point. By pooling data from many points it should also be possible to estimate the mean and variance of the sampled attributes at very large scales. What is lacking is link between the two scales. Specifically, we must understand what happens between the 100-200 meters that surround a sample point and the ten kilometers that separate adjacent sample points (given one point per township). It would be possible to address this issue by extending the sampling zone to the intermediate scale on a subset of the sampling points. For example, at every tenth sample point additional measurements could be taken at fixed intervals out to perhaps a few kilometers from the central point. Similarly, for the temporal scale, a subset of points could be resampled every year to provide an estimate of the annual variance, and provide the context for interpreting fluctuations that are observed in the sites that are sampled less frequently.

E5.3.3 Sampling Protocol For Aquatic Systems

Monitoring of aquatic systems is somewhat problematic using a systematic sampling design because most sample points will fall in terrestrial habitats. One solution would be to sample the water source closest to the systematic sample point. It may also be desirable to take additional measurements from selected larger streams. Together with data from the systematic grid, the data from larger streams could be used provide integrated assessments of selected watersheds. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX E Page E-11 Prepared By: The Oils Sands Vegetation Reclamation Committee

E5.3.4 Sampling Elements

The selection of the elements to be monitored is central to the design of the monitoring program. For large-scale monitoring it is desirable to standardize the types of data that are collected (CENR 1997). This will increase the power of the program to detect changes, increase its cost effectiveness, and facilitate the reporting of the results in a consistent and meaningful manner. Standardized monitoring will also help ensure that changes that are widespread are not erroneously attributed to localized land use practices.

The following is a preliminary list of the core elements that should be measured at each sample point. Abiotic elements are included for later incorporation into multivariate statistical analyses. The biotic elements are intended to provide full coverage of ecological diversity, including composition, structure, and function. Note that many additional measurements pertaining to ecological structure, particularly at larger scales, are made through the remote sensing component of the framework (Section 5.2).

Abiotic:

· Elevation

· Slope

· Aspect

· Soil (nutrient levels and selected physical, chemical, and biological properties)

· Local physiography

· Distance to nearest linear disturbance (e.g., road, seismic line)

· Presence of industrial facilities (e.g., wellsite) within a given radius

Terrestrial flora:

· Composition of vegetative overstorey and understorey using Ecosite Phase and maturity class classification

· Density of snags

· Amount and distribution of coarse woody debris

· Percentage of forest cover within a defined radius

· Percent green tree retention (in recently disturbed sites)

· Estimated abundance of indicator species and composition of indicator assemblages

Terrestrial fauna:

· Estimated abundance of indicator species and composition of indicator assemblages Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX E Page E-12 Prepared By: The Oils Sands Vegetation Reclamation Committee

· Estimated abundance of selected exotic species

Aquatic systems:

· Abiotic measures, including: dissolved oxygen content, amount of large woody debris and stage of rot class, silt loading (from road construction), and temperature and light penetration (via automated long-term sensors)

· Composition of emergent vegetation, larger aquatic invertebrates, small fish species, and benthic fauna

Ecological function (our understanding of ecological function is limited and further research will be required to better define what should be monitored):

· Growth rate of trees

· Nutrient cycling

· Successional trajectory

· Natural disturbance events:

- Large-scale measures including rate of disturbance by type (e.g., forest harvesting, seismic exploration, fire, insect outbreak, disease) and percent of all disturbances due to industrial use

- Small-scale measures including tree fall rates and gap initiation

· Hydrology

The preceding list of elements is fairly general. A considerable amount of effort will be required by the agencies involved in implementing the program to define the specific elements to be monitored. These decisions should be based on scientifically defensible principles to ensure the greatest probability of achieving the goals of the program and to help secure public acceptance of the program. Economic factors will also need to be considered at some point, providing a “reality-check” on what can realistically be implemented. A workshop setting may be best suited for the simultaneous consideration of all perspectives. Participants should include researchers with a broad range of technical expertise and representatives from the various agencies that will ultimately be responsible for implementing the program.

Given the vast number of species existing in forested ecosystems, and the difficulties inherent in monitoring them, the use of surrogate measures for ecological composition will be unavoidable. Indicator species are typically used for this purpose (Pearson 1994; CCFM 1995), though there is much concern associated with their use (Van Horne 1983; Block et al. 1987; Landres et al. 1988). The indicator species that are chosen should represent a broad range of taxonomic groups so that the assessment of composition is as comprehensive as possible. Too narrow a focus during the selection of the indicators will weaken the program and foreclose future options (in the same way that a pre-stratified sampling design is limiting; Section 5.3.2). The following is a list of key criteria that are desirable in species to be used as indicators for monitoring biodiversity (Noss 1990; Pearson 1994; Bonar 1995): Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX E Page E-13 Prepared By: The Oils Sands Vegetation Reclamation Committee

· Distributed across all of the regions to be monitored

· Potential to act as a surrogate for other similar species

· Easy and relatively inexpensive to monitor

· Preexisting data available

· Biology of the species is reasonable well understood

· Sensitive to industrial impacts (able to provide an early warning of problems)

· Sufficient sample size for monitoring

· Stable population dynamics

To circumvent the deficiencies of indicator species, the monitoring of species assemblages should be attempted wherever possible (Kremen 1992). For example, instead of monitoring the abundance of a single species of bird, the presence of all birds along a linear transect (based on vocalizations) could be recorded. Statistical techniques for analyzing changes in such species assemblages have been developed for the analysis of Breeding Bird Survey data and other similar datasets (Bohning-Gaese et al. 1993; James et al. 1996; Flather and Sauer 1996). Such analyses provide a much broader assessment of species diversity than the weak extrapolations provided by individual indicator species.

The monitoring of genetic diversity is even more difficult than the monitoring of species diversity. It is unlikely that direct assessments can be made for more than a few species, such as tree species that are genetically manipulated as part of silvicultural programs. Assessments for other species will have to be indirect, based on the status of population sizes and distributions. Because techniques for the analysis of genetic diversity are advancing rapidly, it would be prudent to archive selected samples for potential analyses in the future (Hedrick and Miller 1992).

E5.4 Rare/Endangered Species Program

Rare and endangered species have a high public profile and must be incorporated into the monitoring framework. However, because of their rarity they are extremely difficult to monitor and specially designed programs are required for the task (Green and Young 1993; Kunin and Gaston 1993). It would be most efficient to integrate the monitoring of these species into the individual recovery plans that are being developed for endangered species in Alberta by the government and various stakeholders. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX E Page E-14 Prepared By: The Oils Sands Vegetation Reclamation Committee

E5.5 Research program

There are many university, industry, and government research initiatives in Alberta dedicated to sustainable forest management (SFMNCE 1997). It would be most efficient to address the research needs of the monitoring program (Section 4.2) through these existing programs, rather than establishing a new program. However, it will be necessary to develop an efficient management structure to coordinate the research efforts and to ensure that all objectives are achieved in a timely manner. The use of the Sustainable Forest Management Network of Centers for Excellence (SFMNCE 1997) should be explored for this purpose. If this organization proves too unwieldy it may be necessary for the agencies that implement the monitoring program to develop their own management entity to oversee and coordinate monitoring-related research.

In addition to its supportive role, research can also be used to directly monitor certain elements of ecological diversity. Studies that compare the effects of various land-use practices to the effects of natural disturbances (Niemela et al. 1993; Siitonen and Martikainen 1994) can be construed as a form of monitoring, albeit, at small spatial and temporal scales. However, the data from these studies is most useful for advancing our understanding of ecological processes, rather than its direct use for monitoring. While every bit of information helps, small-scale research studies are no substitute for dedicated long-term monitoring of the cumulative effects of industrial practices over the entire land base.

One of the key deficiencies of the monitoring program is its lack of predictive power. Because of lags in many ecological processes, and various threshold phenomena, changes may not be detected until a critical state has been reached (With and Crist 1995; Brawn and Robinson 1996; Ligon and Stacey 1996). Therefore, as part of the research program, models based on research findings and data from the monitoring program should be developed to improve our ability to predict and mitigate negative consequences of industrial impacts before they become critical (CENR 1997).

E6. IMPLEMENTATION

E6.1 Administration

While multiple agencies will be involved in developing and funding the monitoring framework, it would be most efficient to establish a single entity to run the program (Scott et al. 1981). This would ensure that measurements and analysis are completely standardized, and it should be more cost effective than having each agency develop and run a small-scale program on its own. Furthermore, running the program at arm’s length from the industry partners that fund it may be critical in obtaining public acceptance of the program and its results.

The organization that runs the monitoring program would be responsible for the following tasks:

· Collecting data from the large-scale sampling network, based on a defined protocol.

· Acquiring forest inventory updates in GIS format from industry and government sources as they become available, along with updated satellite images as required. Proprietary inventory data would be kept confidential.

· Summarizing the data and conducting specified analyses. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX E Page E-15 Prepared By: The Oils Sands Vegetation Reclamation Committee

· Producing an annual report of the results.

· Storing the data and maintaining an Internet web site to provide access to it.

· Archiving samples.

· Running a quality control program to ensure that sampling and other operations continually meet required standards.

Ongoing direction, with respect to sampling protocols, data standards, and issues pertaining to funding, would be provided to the monitoring organization by an administrative committee. This committee would be comprised of representatives from the government, agencies that fund the program, and technical experts. The administrative committee would also provide direction for the research program associated with the monitoring program (Section 5.5).

The interpretation of the annual results of the monitoring program will require careful consideration. There may be profound changes in a single element of ecological diversity, or subtle changes in many. The range of natural variation may clearly be exceeded, or there may only be a progressive trend away from the mean. Changes may widespread or localized. Clearly, it will not be a simple matter to determine when sufficient change has occurred such that action is required, and such a determination is probably beyond the scope of the organization that does the monitoring. The best course may be to submit the results of the monitoring program to a scientific panel on a regular basis for detailed review. To maintain public acceptance of the monitoring program the recommendations of the panel should be made public. Implementation of the recommendations could be done through the administrative committee and by companies that are affected directly.

E6.2 Data Standards

Standardization of sampling protocols and methods of analysis will be critical for making meaningful comparisons across wide areas (Ringold et al. 1996; CENR 1997). The establishment of a technical working group for the development of these standards would be useful. To permit national and international comparisons, consideration should be given to the standards that are currently being developed for the monitoring of biodiversity by the EMAN initiative in Canada (EMCO 1996), and the National Environmental Monitoring Initiative of the United States (CENR 1997). Research may be required to determine the optimal methodologies for sampling and analysis where well-defined procedures have not yet been established.

Standards and defined protocols need to be established for the following elements of the program:

· Design of the large-scale sampling network and total number of sample points required (based on statistical power analyses; Section 5.3.1)

· Frequency of resampling individual sample points

· Spatial pattern of annual sampling. For example, a different Natural Region could be completely sampled each year, or a fraction of the points could be sampled across the entire land base each year Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX E Page E-16 Prepared By: The Oils Sands Vegetation Reclamation Committee

· Data formats, particularly for GIS data

· Elements to be sampled (Section 5.3.4)

· Sampling methodology for each element (including length of transects, number and size of quadrats, measurements taken, etc.)

· Data summarization and analysis

It is also important that the monitoring framework be compatible with the criteria and indicators developed by the CCFM and used by forest certification programs (Appendix A). The monitoring framework as outlined, together with provincial-level data already collected by the government, should be sufficient to satisfy all of the CCFM indicators relating to ecological diversity and ecosystem function. Issues pertaining to data formats and reporting may still need to be addressed. National-level indicators (e.g., global carbon cycles) and socio-economic indicators are beyond the scope of the monitoring framework.

E6.3 Funding

Given the vast area that must be monitored, and the fact that access to most sample points will have to be by helicopter, it can be expected that the monitoring program will be an expensive undertaking. If industry is asked by the government to fund the program, then the most equitable approach would be to apportion the required fees based on resource usage per company. The stumpage fee system currently used by the forestry sector is an example of such an approach. In fact, it may be reasonable to direct some of the money currently collected from stumpage fees to the monitoring program, instead of to the productivity-related activities it is now used for. The cumulative impact of the oil and gas industry is on par with that of the forest industry and therefore its contribution to the monitoring program should also be on the same scale. Fees could be assigned based on the number of square meters of forest impacted by seismic exploration, drilling, and road building.

E6.4 Reference Areas

As discussed in Section 5.1, two forms of control, or “base case”, are available for the monitoring program: a “snapshot” of the land base prior to industrial use, based on the initial years of sampling, and a contemporary control based on a system of ecological reference areas and parks. As both types of controls have serious deficiencies (see Section 5.1), it is critical that both be used together, if valid assessments are to be made.

The need for reference areas is documented in the Alberta Forest Conservation Strategy (Alberta Forest Conservation Strategy Steering Committee 1997) and it is recommended that the Special Places 2000 program be used as the mechanism for their establishment. However, the mandate of Special Places 2000 has been narrowly interpreted and the establishment of reference areas is not currently under consideration.

The lack of a provincial program notwithstanding, the selection of suitable reference areas may be fairly straight forward, because few options remain with regard to their placement. In northern Alberta, the Chinchaga watershed, the Liege River watershed, and the Caribou Mountains have been proposed as potential reference areas. Though not entirely pristine, these regions contain large core areas that are not traversed by roads and have had minimal exposure to industrial Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX E Page E-17 Prepared By: The Oils Sands Vegetation Reclamation Committee activity to date. The forestry companies with cutting rights in these areas support their establishment as reference areas, though the oil and gas sector has not yet given its support. These three areas, together with Wood Buffalo National Park, would provide representation of seven Natural Subregions: Central Mixedwood, Wetland Mixedwood, Peace River Lowlands, Subarctic, Boreal Highlands, and Upper and Lower Foothills (Department of Environmental Protection 1994a,b). Three additional Natural Subregions would remain in need of representation: Dry Mixedwood, and the Upper and Lower Foothills in the southern half of the province. These latter three Subregions have already been heavily impacted by industrial use and, because of a high level of industrial commitment, there are few options available for the establishment of large reference areas. Creative solutions will be required, possibly involving the Little Smoky river and an expansion of Notikewan Provincial Park.

In addition to the issue of representation, candidate reference areas must also be capable of maintaining natural ecological processes over the long term (i.e., over 100 years). Because of this requirement the reference areas must be spatially fixed and industrial use must be prohibited within them. It is also necessary that the areas be large, so that they are not significantly influenced by industrial use of the adjacent forest, and so that ecological processes that operate at large spatial scales are maintained.

Two key processes that operate at large spatial scales are the movements of wide-ranging species and natural disturbance events such as fire. Even with fire suppression, large fire events still occur, as evidenced by the Mariana Lakes fire in 1995 (1,300 km2). These infrequent catastrophic events may be qualitatively different than the many smaller disturbances that occur in intervening years and they may be one of the key processes that maintain ecological diversity in the boreal forest (Romme and Despain 1989; Greenberg et al. 1994; Bessie and Johnson 1995; Romme et al. 1995). The literature is vague with regards to specific size requirements for the maintenance of ecological processes, but most reports suggest that the areas must be greater than the largest disturbance event (Pickett and Thompson 1978; Baker 1992). For northern Alberta, this implies an area of several thousand square kilometers.

The final key requirement of candidate reference areas is that they contain a sufficient number of sample points to act as a statistically meaningful controls. If one sample point is established per township, then there would be 11 points per thousand square kilometers. This again implies that the reference areas will have to be several thousand kilometers in size, though statistical power analyses will be required to provide a more robust estimate.

Areas of several thousand square kilometers are possible for the Chinchaga, Liege, and Caribou Mountains, but may be difficult to establish elsewhere.

E7. ACKNOWLEDGMENTS

I thank the review committee, comprised of Stan Boutin, Dan Farr, Brett Purdy, Chris Spytz, and Brad Stelfox, for the help, advice, and insight they provided throughout the development of the framework. I also thank the following people for providing feedback and sharing their perspectives on monitoring with me: Peter Achuff, Harry Archibald, Rick Bonar, Steve Brechtel, Gordon Court, Ian Corns, Dan Gilmore, Susan Hannon, David Langor, Phil Lee, Brent Rabik, Jim Schieck, Fiona Schmiegelow, David Shindler, John Spence, Harry Stelfox, Norman Stolle, Mike Sullivan, Dale Vitt, Jan Volney, and Shawn Wasel. I also thank Kerry Grisley, with the Alberta Ecological InfoService, for her assistance with the literature review. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX E Page E-18 Prepared By: The Oils Sands Vegetation Reclamation Committee

E8. LITERATURE CITED

AFCSSC (Alberta Forest Conservation Strategy Steering Committee). 1997. Alberta Forest Conservation Strategy. Alberta Department of Environmental Protection, Edmonton, AB.

Baker, W. 1989. Landscape ecology and nature reserve design in the Boundary Waters Canoe Area, Minnesota. Ecology 70:23-35.

Baker, W. 1992. The landscape ecology of large disturbances in the design and management of nature reserves. Landscape Ecol. 7:181-194.

Bessie, W.C. and Johnson, E.A. 1995. The relative importance of fuels and weather on fire behavior in subalpine forests. Ecology 76:747-762.

Block, W.M., Brennan, L.A., and Gutierrez, R.J. 1987. Evaluation of guild-indicator species for use in resource management. Environ. Manag. 11:265-269.

Bohning-Gaese, K., Taper, M.L., and Brown, J.H. 1993. Are declines in North American insectivorous songbirds due to causes on the breeding range. Cons. Biol. 7:76-83.

Bonar, R. 1995. Biodiversity Indicators for Forest Managers - An Example. Unpublished discussion paper.

Bourdeau, P.F. 1953. A test of random versus systematic ecological sampling. Ecology 34:499- 511.

Brawn, J.D. and Robinson, S.K. 1996. Source-sink population dynamics may complicate the interpretation of long-term census data. Ecology 77:3-12.

Brown, J., Mehlman, D.W., and Stevens, G. 1995. Spatial variation in abundance. Ecology 76:2028-2043.

Bunnell, F. 1995. Forest-dwelling vertebrate faunas and natural fire regimes in British Columbia: patterns and implications for conservation. Cons. Biol. 9:636-644.

CCFM (Canadian Council of Forest Ministers). 1992. National Forest Strategy. Sustainable Forests: A Canadian Commitment. Canadian Forest Service. Ottawa, ON.

CCFM (Canadian Council of Forest Ministers). 1995. Defining Sustainable Forest Management: A Canadian Approach to Criteria and Indicators. Canadian Forest Service, Ottawa, ON.

CENR (Committee on Environment and Natural Resources). 1997. Integrating the Nation's Environmental Monitoring and Research Networks and Programs: A Proposed Framework. National Science and Technology Council, Washington, D.C.

CFS (Canadian Forest Service). 1995. Criteria and Indicators for the Conservation and Sustainable Management of Temperate and Boreal Forests. Canadian Forest Service, Hull, PQ.

CFS (Canadian Forest Service). 1996. The State of Canada's Forests 1995-1996. Canadian Forest Service, Ottawa, ON. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX E Page E-19 Prepared By: The Oils Sands Vegetation Reclamation Committee

CFS (Canadian Forest Service). 1997. The Model Forest Network Home Page. Canadian Forest Service. URL: http://mf.ncr.forestry.ca.

Christensen, N.L. 1996. The report of the Ecological Society of America Committee on the Scientific Basis for Ecosystem Management. Ecol. Applic. 6:665-691.

CSA (Canadian Standards Association). 1996. A Sustainable Forest Management System: Guidance Document. Canadian Standards Association, Toronto, ON.

DEP (Department of Environmental Protection). 1994a. Alberta Vegetation Inventory Standards Manual. Department of Environmental Protection, Edmonton, AB.

DEP (Department of Environmental Protection). 1994b. Natural Regions of Alberta. Department of Environmental Protection, Edmonton, AB.

Doyon, F. and P. Duinker. 1997. Biodiversity Assessments at the Species Level: Species Selection for Millar Western Industries Ltd. and Alberta Newsprint Company FMA Territories in Central Alberta (BAP Report #2). Unpublished discussion paper.

Drysdale, C. and B. Beattie. 1995. Integrated Ecological Monitoring and Atmospheric Issues in Kejimkujik National Park. Environmental Monitoring and Assessment Network. URL: http://www.cciw.ca/eman-temp/reports/publications/intro.html.

EC (Environment Canada). 1995. Canadian Biodiversity Strategy. Environment Canada, Hull, PQ.

EC (Environment Canada). 1997. The National Environmental Indicator Series. Environment Canada. URL: http://www1.ec.gc.ca/~ind/

Elliott, C. and Hackman, A. 1996. Current Issues in Forest Certification in Canada. World Wildlife Fund, Toronto, ON.

EMCO (Environmental Monitoring Coordinating Office). 1996. EMAN’s Goals, Objectives, and Deliverables: 1996 Declarations. EMAN Occasional Paper Series # 3. URL: http://www.cciw.ca/eman-temp/reports/publications/intro.html.

Everett, R. 1993. Eastside Forest Ecosystem Health Assessment: Executive Summary. U.S. Department of Agriculture, Washington, D.C.

Flather, C.H. and Sauer, J.R. 1996. Using landscape ecology to test hypotheses about large- scale abundance patterns in migratory birds. Ecology 77:28-35.

Frelich, L.E. and Reich, P.B. 1995. Spatial patterns and succession in a Minnesota southern- boreal forest. Ecol. Monog. 65:325-346.

Gilmore, D. 1997. Canfor Draft Measures and Indicators Document. Unpublished discussion paper.

Green, R.H. and Young, R.C. 1993. Sampling to detect rare species. Ecol. Applic. 3:351-356.

Greenberg, C.H., Neary, D.G., and Harris, L.D. 1994. Effect of high-intensity wildfire and Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX E Page E-20 Prepared By: The Oils Sands Vegetation Reclamation Committee

silvicultural treatments on reptile communities in sand-pine scrub. Cons. Biol. 8:1047- 1057.

Grumbine, E. 1994. What is ecosystem management? Cons. Biol. 8:27-38.

Hecnar, S.J. and M'Closkey, T.M. 1996. Regional dynamics and the status of amphibians. Ecology 77:2091-2097.

Hedrick, P.W. and Miller, P.S. 1992. Conservation genetics: techniques and fundamentals. Ecol. Applic. 2:30-45.

Hunter, M. L. 1991. Coping with ignorance: the coarse-filter strategy for maintaining biodiversity. Pages pp. 266-281 in Kohn, K., ed. Balancing on the Brink of Extinction. Island Press, Washington, D.C.

Hunter, M.L. 1993. Natural fire regimes as spatial models for managing boreal forests. Biol. Conser. 72:115-120.

James, F.C., McCulloch, C.E., and Wiedenfeld, D.A. 1996. New approaches to the analysis of population trends in land birds. Ecology 77:13-27.

Kessler, W., Salwasser, H., Cartwright, C., and Caplan, J. 1992. New perspectives for sustainable natural resources management. Ecol. Applic. 2:221-225.

Knopf, F. and Samson, F. 1994. Scale perspectives on avian diversity in western riparian ecosystems. Cons. Biol. 8:669-676.

Kremen, C. 1992. Assessing the indicator properties of species assemblages for natural areas monitoring. Ecol. Applic. 2:203-217.

Kunin, W.E. and Gaston, K.J. 1993. The biology of rarity: patterns, causes and consequences. Trends Ecol. Evol. 8:298-301.

Kurz, W. A., Apps, M. J., Stocks, B., and Volney, J. 1995. Global climate change: disturbance regimes and biospheric feedbacks of temperate and boreal forests. Pages 119-133 in Woodwell, G. M. and MacKenzie, F. T., eds. Biotic Feedbacks in the Global Climate System. Oxford University Press, New York.

Landres, P.B., Verner, J., and Thomas, J.W. 1988. Ecological uses of vertebrate indicator species: a critique. Cons. Biol. 2:316-328.

Ligon, J.D. and Stacey, P.B. 1996. Land use, lag times and the detection of demographic change: the case of the acorn woodpecker. Cons. Biol. 10:840-846.

Lord, J.M. 1990. Scale and the spatial concept of fragmentation. Cons. Biol. 4:197-201.

McGarigal, K. and B.J. Marks. 1994. Fragstats. Spatial pattern analysis program for quantifying landscape structure. Version 2.0. Forest Science Department, Oregon State University, Corvallis, OR.

Murtaugh, P.A. 1996. The statistical evaluation of ecological indicators. Ecol. Applic. 6:132-139. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX E Page E-21 Prepared By: The Oils Sands Vegetation Reclamation Committee

Naiman, R., Decamps, H., and Pollock, M. 1993. The role of riparian corridors in maintaining regional biodiversity. Ecol. Applic. 3:209-212.

NERC (National Environmental Research Council). 1994. The United Kingdom Environmental Change Network. Institute of Terrestrial Ecology, Cumbria, UK.

Niemela, J., Langor, D., and Spence, J.R. 1993. Effects of clear-cut harvesting on boreal ground- beetle assemblages (Coleoptra: Carabidae) in western Canada. Cons. Biol. 7:551-560.

Noss, R.F. 1990. Indicators for monitoring biodiversity: a hierarchical approach. Cons. Biol. 4:355-364.

Oliver, I. and Beattie, A. 1993. A possible method for the rapid assessment of biodiversity. Cons. Biol. 7:562-571.

Oliver, I. and Beattie, A.J. 1996. Designing a cost-effective invertebrate survey: a test of methods for rapid assessment of biodiversity. Ecol. Applic. 6:594-607.

Pearson, D.L. 1994. Selecting indicator taxa for the quantitative assessment of biodiversity. Phil. Trans. R. Soc. Lond. B 345:75-79.

Pickett, S. and Thompson, J.N. 1978. Patch dynamics and the design of nature reserves. Biol. Conser. 13:27-37.

Rabik, B. and Larson, K. 1997. Development of Alberta-Pacific’s Biomonitoring Program. Unpublished discussion paper.

Ringold, P.L., Alegria, J., and et. al. 1996. Adaptive monitoring design for ecosystem management. Ecol. Applic. 6:745-747.

Romme, W.H. and Despain, D.G. 1989. The Yellowstone fires. Sci. Am. 261:37-46.

Romme, W.H., Turner, M., Wallace, L., and Walker, J. 1995. Aspen, elk, and fire in northern Yellowstone National Park. Ecology 76:2097-2106.

Scott, J.M., Jacobi, J.D., and Ramsey, F.L. 1981. Avian surveys of large geographical areas: a systematic approach. Wildl. Soc. Bull. 9:190-200.

SFMNCE (Sustainable Forest Management Network of Centres for Excellence). 1997. The Sustainable Forest Management Network of Centres for Excellence Home Page. URL: http://www.biology.ualberta.ca/sfm/

Siitonen, J. and Martikainen, P. 1994. Occurrence of rare and threatened insects living on decaying Populus tremula: a comparison between Finnish and Russian Karelia. Scand. J. of Forest Res. 9:185-191.

Steen, H., Ims, R.A., and Sonerud, G. 1996. Spatial and temporal patterns of small-rodent population dynamics at a regional scale. Ecology 77:2365-2372.

Swanson, F.J., Jones, J., Wallin, D., and Cissel, J. 1993. Eastside Forest Ecosystem Health Assessment: Natural Variability - Implications for Ecosystem Management. U.S. Dept. of Agriculture, Washington, D.C. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX E Page E-22 Prepared By: The Oils Sands Vegetation Reclamation Committee

Thomas, L. 1996. Monitoring long-term population change: why are there so many analysis methods? Ecology 77:49-58.

UNECE (United Nations Economic Commission for Europe). 1993. Manual for Integrated Monitoring, Program Phase 1993-1996. Environment Data Center, National Board of Waters and Environment, Helsinki, Finland.

UNEP (United Nations Environment Programme). 1992. Convention on Biological Diversity. UNEP Publication Na. 92-7807. Geneva, Switzerland.

UNESCO (United Nations Educational Scientific and Cultural Organization). 1997. The Global Terrestrial Observing System. UNESCO Man and the Biosphere Program. URL: http://www.unesco.org/mab/home/mabis.htm.

Van Horne, B. 1983. Density as a misleading indicator of habitat quality. J. Wildl. Manag. 47:893- 901.

Walters, C. and Holling, C. 1990. Large-scale management experiments and learning by doing. Ecology 71:2060-2068.

With, K. and Crist, T.O. 1995. Critical thresholds in species' responses to landscape structure. Ecology 76:2446-2459.

Woodley, S. 1997. Indicators of Ecological Integrity. Environment Canada. Compendium of Sustainable Development Indicator Initiatives. URL: http://www1.ec.gc.ca/

YNP (Yoho National Park). 1997. Ecosystem monitoring and Yoho National Park. Environmental Monitoring and Assessment Network. URL: http://www.cciw.ca/eman-temp/reports/publications/intro.html. APPENDIX F

A HISTORY OF TERRESTRIAL RECLAMATION IN THE OIL SANDS REGION Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX F Page F-1 Prepared By: The Oil Sands Vegetation Reclamation Committee

TABLE OF CONTENTS

PAGE

F1. INTRODUCTION 2 F1.1 HISTORY OF LAND RECLAMATION TECHNIQUES AT SUNCOR...... 2 F1.1.1 Soil Reconstruction...... 4 F1.1.2 Revegetation...... 5 F1.2 RECLAMATION RESEARCH AND MONITORING...... 7 F1.2.1 Vegetation and Soil Monitoring of Reclaimed Land ...... 8 F1.3 HISTORY OF LAND RECLAMATION TECHNIQUES AT SYNCRUDE...... 10 F1.3.1 Surface Preparation...... 10 F1.3.2 Application of Fertilizer...... 11 F1.3.3 Revegetation...... 12 F1.3.4 Monitoring ...... 12

F2. REVEGETATION SUCCESS ...... 12 F2.1 RECLAIMED TAILINGS SAND ...... 12 F2.2 RECLAIMED OVERBURDEN...... 14 F2.3 SUMMARY OF VEGETATION ESTABLISHMENT IN RECLAIMED AREAS...... 14

F3. LITERATURE CITED...... 23 Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX F Page F-2 Prepared By: The Oil Sands Vegetation Reclamation Committee

A HISTORY OF TERRESTRIAL RECLAMATION IN THE OIL SANDS REGION

Prepared by: Earl Anderson – Sycrude Canada Limited Tom Coolern - Syncrude Canada Limited Steve Tuttle – Suncor Energy Inc. Carl Warner – AGRA Earth & Environmental Limited

F1. INTRODUCTION

Reclamation of terrestrial vegetation in the oil sands region has been ongoing since 1971 at Suncor and since 1976 at Syncrude. The majority of reclamation has occurred on relatively steep slopes (e.g., 27% to 40% slope angles) of tailings sand dykes and overburden dumps. The information presented in this section provides a summary of the methods used to reclaim these areas and the success of these methods.

F1.1 HISTORY OF LAND RECLAMATION TECHNIQUES AT SUNCOR

The current reclamation method at Suncor has been developed as a result of many years of scientific study, demonstration plots and serendipity. The original intent for implementing reclamation at Suncor in 1971 was to develop an erosion controlling cover of grasses. This goal was achieved through the placement of approximately 10 cm of peaty soil onto the reclamation site, tilling this soil into the overburden or tailings sand, then seeding to grass through use of a hydroseeder or seed drill. This method was successful in developing a self-sustaining cover of grass and legumes on the reclaimed areas. Although a sprinkler system was used at one time to assist the establishment of the grasses, this proved to be unnecessary, as the grass cover successfully developed under typical levels of precipitation received in the region.

However, the reclamation focus at Suncor has changed from one of erosion control only, to one where the development of a self-sustaining ecosystem in tune with the region has become the priority. One of the primary reasons for changing this goal was due to a shift in the type of soil being used as an amendment. Instead of utilizing peat alone, the peat layer along with a portion of the underlying soil was taken as a reclamation amendment. This provided for the addition of more fine components to the amendment, which allowed for a better quality soil to be placed on the reclamation site. In an attempt to eliminate areas where insufficient soil amendment was placed, the depth of application was increased to 20 cm.

In addition to the change in the depth of the amendment placed on a reclamation area, the amendment material source has also been altered. Muskeg soil had been excavated and placed in stockpiles for future use. Reclamation material needs were then drawn directly from these stockpiles. During the reclamation programs of 1983 and 1984, the source of the muskeg soil changed to deposits located in unmined areas where disturbance was minimal. The undisturbed material was excavated and placed directly on the reclamation site. The result of using partially frozen in situ pockets of topsoil was noteworthy because native seeds and root fragments transferred with the soil became established and grew rapidly on the reclamation sites. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX F Page F-3 Prepared By: The Oil Sands Vegetation Reclamation Committee

Seeding methods and mixtures have also changed. Hydroseeding of grasses and legumes on the reclamation areas is no longer included as part of the seeding program. Barley is now used as a nurse crop on all reclaimed sites because grasses proved to be over-competitive to trees, shrubs, and forbes developing on the reclamation sites. Aerial seeding is undertaken on new reclamation sites after which the areas are raked lightly with harrows. Raking of a shallow cover of soil on the barley seeds promotes early seed germination thus hastens erosion control for the newly reclaimed site. The barley cover does not hinder the development of the emerging vegetation but still functions in reducing erosion. Additionally, it leaves a stubble capable of trapping snow which is needed to protect the young tree seedlings during the winter.

A further refinement to the process of applying topsoil has been to excavate and haul soil building materials during the winter months when the insitu seed and roots are dormant. Spreading of the muskeg soil on the reclamation site is then completed in early spring. The usual result is the emergence of a variety of native woody stemmed plants, forbes, wildflowers and grasses. This prolific vegetative growth provides a erosion controlling cover which is diverse, in tune with the ecosystems found in the region, and easily capable of meeting the goals of the program.

Suncor has adopted, as part of its reclamation strategy, a planting program designed to provide a diverse mixture of woody stemmed species on the reclaimed areas. The species mixture includes wild rose, raspberry, gooseberry, saskatoon and chokecherry, all of which are used by birds and mammals. Tree species are also planted to provide ecosystem diversity and health. The planting program, in conjunction with the profusion of native plants developing from the soil amendment, provides for a diverse vegetative community on the reclamation sites. Sufficient numbers of trees are planted to meet the Alberta Environmental Protection - Lands and Forest Service (AEP-LFS) establishment criteria for a commercial stand (Alberta Environmental Protection 1994). The numbers of trees on reclamation areas should also be able to provide for a variety of wildlife and human uses.

Suncor’s vision for reclamation includes the construction of stable landforms and re-establishment of productive, self-sustaining ecosystems which will provide land use capabilities equivalent to those of the pre-mining environment. The following general operational and reclamation criteria form the basis for reclamation program design:

· Structures will be geotechnically stable.

· Discharge of earth materials through surface erosion processes will be controlled to rates which are acceptable to the environment.

· Discharge of surface and seepage waters will be managed to ensure an acceptable level of impact on the Athabasca River.

· The ecosystems re-established on disturbed lands will be fully self-sustaining and will mature naturally without presenting significant risk to resident or migratory wildlife, or plant species.

· Fully reclaimed lands will be maintenance-free, thereby justifying reclamation certification.

Development of methodologies to achieve the Suncor reclamation objectives requires an Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX F Page F-4 Prepared By: The Oil Sands Vegetation Reclamation Committee understanding of the principal processes that influence ecosystem development. The types of vegetation and soil that will develop on the Suncor lease are dependent on climate, topography, parent material, drainage and time. However, of these environmental factors only parent materials and topography can be modified to any extent by Suncor. The other factors are essentially fixed by virtue of the oil sands mining and extraction methodologies or by natural conditions over which Suncor has no control.

The Suncor reclamation approach is to reconstruct a “soil” from tailings or overburden parent material and muskeg soil (peat and mineral fines). This creates a mixture that is capable of sustaining an initial erosion controlling plant cover. The reconstructed soil is also designed to be capable of supporting the growth of tree species which were found in the pre-mined forest communities and exist in areas adjacent to the lease.

F1.1.1 Soil Reconstruction

The restoration of soil capabilities to a state equal to or greater than predisturbed conditions required the definitions of reconstructed soil conditions. The design specifications ensure that the reconstructed soil provides: · Adequate moisture supply, · Adequate nutrient supply, and · Acceptable erosion control.

Three natural soil types have been selected as references to represent comparable pre-mining capabilities for the three main post-mining landforms (i.e., tailings sand dykes, tailings sand plateaus, and overburden dykes/dumps). Physical and chemical parameters of these reference soils are used to assess the effectiveness of reconstructed soils in meeting the above specifications and in meeting the Suncor long term reclamation goals.

The mining operation creates a variety of land forms that must be reclaimed, including: · Tailings sand dykes, · Tailings sand plateaus, · Overburden dumps and dykes, · End pit (wall and floor), · Tailings ponds, · Oversize dump, and · Ancillary areas such as coke and sulphur storage pads, ponds and roads.

Suncor utilizes a capping methodology for seedbed preparation. The capping methodology involves distribution of salvaged muskeg soil to the various reclamation areas followed by spreading to an average depth of 20 cm. Spreading is followed by application of fertilizer and a barley nurse crop which is mixed into the seedbed by harrows. This results in minimal mixing of the amendment with the spoil thereby favouring establishment of native species from seed and root fragments in the amendment material. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX F Page F-5 Prepared By: The Oil Sands Vegetation Reclamation Committee

F1.1.2 Revegetation

The primary objectives of the Suncor revegetation program are to:

· Provide an erosion-controlling plant cover on tailings dyke slopes and overburden dump slopes, and

· Establish a permanent, self-sustaining cover of forest species.

Secondary objectives are to provide habitats which are suitable for wildlife use and which have possibilities for recreation.

Suncor has developed a revegetation program based on field trials and operational experience to aid in the achievement of these objectives. The revegetation program involves: seeding of reclamation areas with ground covers designed to control erosion; area fertilization; and establishment of appropriate woody plant species. Measures are also taken to encourage the native plant invasion onto reclaimed sites.

The distribution of the vegetation types present on the Suncor lease prior to development was related to the type of surficial materials, soils and drainage regime. The Suncor revegetation program is aimed at establishing four main vegetation types on the three main reclamation landforms (tailings sand plateaus, tailings sand slopes, and overburden dumps. The reclamation starter vegetation types are:

· Pine Forest - This vegetation type will be established on the edges of tailings sand plateaus and tailing sand slopes.

· Poplar-White Spruce/Shrub - This vegetation type will be established on the moister areas of the tailings sand plateaus.

· White Spruce-Poplar/Shrub Community - This vegetation type will be established on the overburden dumps and more mesic sites on tailings dyke slopes (lower portions of the slopes and/or areas with northerly aspects).

· Wetland Complex - This vegetation type will be established on poorly drained areas of the tailings sand plateaus. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX F Page F-6 Prepared By: The Oil Sands Vegetation Reclamation Committee

Revegetation methodology includes fertilizer application, herbaceous ground cover establishment, woody plant establishment and revegetation area maintenance. Each of these are reviewed below.

F1.1.2.1 Fertilization

During the 1970s, fertilizer was applied and incorporated into the surface amendment, followed by annual maintenance applications for all reclaimed areas at Suncor. Composition and application rates were based on agricultural production recommendations for forage crops and annual soil monitoring results of plant available nutrients. In the mid 1980's, annual maintenance applications were reduced to minimize the potential for competition between the developing herbaceous vegetation and the planted woody seedlings. At this time, fertilizer is applied during the initial years of an area's reclamation as an aid for the rapid development of an erosion-controlling vegetative cover. Annual fertilization is then discontinued so that the developing herbaceous cover will not compete vigorously with planted woody seedlings. The starter fertilizer formulation incorporated into the seedbed is essentially the same for tailings sand and overburden. Rates and composition are determined from initial field trials and annual monitoring. Maintenance rates are determined from criteria such as soil tests and cover performance. Typical maintenance periods, which depend on the rate of ground cover establishment, are limited to 2 to 3 years after reclamation for reclaimed overburden and 3 to 4 years for tailings sand.

F1.1.2.2 Herbaceous Ground Cover Establishment

The strategy for ground cover establishment has been to develop a vegetative cover that does not become overly competitive with outplanted woody stock, yet is still able to control erosion. Early programs prescribed agronomic species at high seeding rates which established easily and provided nearly immediate erosion control. However, these agronomic species also became restrictive to the establishment of trees and shrubs. The current approach is to seed barley either by helicopter (primary method) or hydroseeding following seedbed preparation. Barley, an annual cereal species, provides nearly immediate erosion control in the first growing season. Additionally, it produces a litter and root biomass that further controls erosion in succeeding growing seasons. Native plants may easily invade the areas or regenerate from muskeg soil applied during seedbed preparation, while outplanted woody stock performance is also greatly enhanced. Results from a study conducted by Hardy BBT Limited (1990) for the Alberta Reclamation Research Technical Advisory Committee (RRTAC), as well as results from the annual Suncor reclamation monitoring program continues to verify the success of this approach.

F1.1.2.3 Woody Plant Establishment

Activities to establish woody plants on Suncor reclamation areas began in 1972. Since then, over 1 750 000 trees and shrubs of various species have been planted. Suncor research projects on woody plant establishment have included evaluation of: plant container types, planting time, effect of ground cover density on woody plant survival, fertilizer amendments, species selection, direct seeding, and planting of hardwood cuttings. Assessment plots are established in reclamation operational areas for inclusion in the annual monitoring program. The results of these research efforts have been used to refine the operational afforestation program. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX F Page F-7 Prepared By: The Oil Sands Vegetation Reclamation Committee

The selection of species and proportions of each species in the planting mix is based on existing field conditions and the vegetative type expected to develop on the site. The species selected are representative of different stages of vegetative succession in the region. This means that as the woody cover develops and the micro-environment modifies to provide favourable conditions for later successional and climax species, these species will be present and able to take advantage of the change in conditions. In this manner, the process of natural succession is accelerated towards the conceptual vegetation covers. Tree and shrub seedlings are planted at an average total density of 2500 stems/ha. This planting density was selected so that sufficient numbers are planted to ensure adequate stocking of each species after initial mortality and to permit the establishment of volunteer plants. The species composition numbers are subject to change depending on the availability of planting stock and specific reclamation site conditions. The planting program is designed to provide a diverse mixture of woody stemmed species on the reclaimed areas. The species mixture includes wild rose, raspberry, gooseberry, saskatoon, and chokecherry, all of which are used by birds and mammals. Tree species are also planted to provide ecosystem diversity and health. The planting program in conjunction with the profusion of native plants developing from the soil amendment provides for a diverse vegetative community on the reclamation sites.

Seedlings are propagated from seed and cuttings collected from the Fort McMurray area. Outplanting periods are early spring and late summer depending on logistics and availability of reclaimed areas. Planting is undertaken as soon as possible after soil reconstruction has been completed.

F1.2 RECLAMATION RESEARCH AND MONITORING

Suncor conducts annual monitoring programs in reclaimed areas specifically to assess herbaceous vegetation growth as well as soil physical and chemical properties. Results of these monitoring programs have been documented and are reported to Alberta Environmental Protection in the annual Development and Reclamation report.

Annual assessments of tree and shrub survival and growth have been conducted in areas where a known number of seedlings have been outplanted. In addition, several other studies related to land reclamation, as well as groundwater monitoring and fine tailings handling, have been undertaken. Suncor will continue research in these areas to ensure that reclamation goals are achieved. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX F Page F-8 Prepared By: The Oil Sands Vegetation Reclamation Committee

F1.2.1 Vegetation and Soil Monitoring of Reclaimed Land

Vegetation and soil characteristics in reclaimed areas on the Suncor Lease are routinely monitored for Suncor by both in-house staff and external personnel. The monitoring program consists of regular annual vegetation cover assessment and soil sampling from areas reclaimed within the past three to four years. Additionally, a detailed assessment and sampling of older reclaimed areas is completed every fifth year. This information provides input into the Graphical Information System (GIS) (Arc Info & Arc View) which is used by Suncor as a planning tool, to relate the monitoring data to the soils and plant development on the reclamation areas.

F1.2.1.1 Weather Data

The development of a vegetative cover can be affected by meteorological conditions. Therefore, the amount of precipitation, temperature and the seasonal variations of these factors is very important when assessing area development. The annual reclamation monitoring program has shown that normal precipitation and temperature together with a relatively even distribution of precipitation during the growing season results in superior vegetation growth in all reclaimed areas.

F1.2.1.2 Soil Sampling

Soil samples are collected as part of the reclamation program, in conjunction with the vegetation assessment. On reclaimed tailings sand, three layers were sampled per transect at most sites including the amended layer (approximately 0 to 15 cm), the layer immediately below the amended layer (15 to 30 cm) and the 45 to 60 cm layer. Samples were taken from three locations (i.e., approximately the end points and the middle of each transect). These were bulked to form one composite sample per layer per transect. In most cases, portions of several samples were bulked again to form one composite sample representing a specific reclamation area which in turn are submitted for laboratory analysis. Since soil properties of the lower layers (analyzed the year the site was reclaimed) tend to change much more slowly than at the surface, it was considered unwarranted to repeat the analysis of the lower layers within such a short time period.

F1.2.1.3 Vegetation Assessment

Herbaceous Species Vegetation cover and height data are collected from assessment plots located on transect line(s) running through the reclamation area. These transects are determined through review of aerial photographs of the areas. At each location, a permanent 30 m transect is positioned at midslope, parallel to the contour. Ten 0.10 m2 quadrats are systematically placed along the transect and the average vegetation height (of herbaceous species only), percent living plant cover (by individual species) and dead plant cover are estimated within each quadrat. Tree (Woody Stemmed Vegetation) Assessments

Suncor uses two methods for assessing woody stemmed vegetation establishment, growth and performance. The Primary Assessment method is used to assess reclaimed areas three to four Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX F Page F-9 Prepared By: The Oil Sands Vegetation Reclamation Committee years after tree planting has been completed on a reclamation site. The Regeneration Survey method is conducted on all areas that have been revegetated for a period of 8 years or greater.

Primary Assessment: The Primary Assessment procedure is employed to monitor the success of establishing woody stemmed vegetation on a reclamation site over the short term. Collected data is used to plan maintenance treatments for individual sites (e.g., fill-in planting and fertilization). The results from the survey method, while variable, provide an indication about the need for additional work on a reclamation area. Areas expected to develop into a commercial forest stand are to have 1200 stems/ha at the end of the assessment period. A survey may indicate that insufficient woody stemmed plants have become established to meet the Alberta Environmental Protection - Land and Forest Services (AEP-LFS) Regeneration Standard. Should this occur, fill-in planting is implemented or other mitigative work is completed. Tailings sand slopes that were reclaimed primarily for erosion control are also assessed. If these areas do not meet the forest stand criteria, they may be replanted or left as open wildlife habitat.

The main detrements to tree establishment on reclamation sites are a southerly aspect and the presence of a grass mat. Excessive grass growth, in conjunction with the southern exposure aspect, result in severe drying conditions which stress trees or cause tree mortality. This may result in the requirement to undertake additional tree planting work.

Regeneration Survey: The second method of assessing the vegetation involves utilization of the Regeneration Survey procedure developed by the AEP-LFS. The assessment survey must be conducted by a certified person using the AEP-LFS Regeneration Survey Manual (Alberta Environmental Protection 1994).

The AEP-LFS mixedwood establishment survey is utilized on the Suncor site to help evaluate tree survival and native invasion on reclamation areas. According to the survey criteria, acceptable trees, eight years following planting, include 50+ cm spruce, or 100+ cm jack pine, larch, aspen, balsam poplar, or birch. These species, once they have reached a commercial size, all have potential for use in lumber or pulp industries. The AEP-LFS establishment survey was modified slightly to include the presence of other woody stemmed plants because Suncor's reclamation focus is to develop a diverse ecosystem on the reclamation sites. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX F Page F-10 Prepared By: The Oil Sands Vegetation Reclamation Committee

F1.3 HISTORY OF LAND RECLAMATION TECHNIQUES AT SYNCRUDE

Reclamation of lands disturbed by Syncrude has been taking place since 1976; initially on construction disturbances and then subsequently on mine/plant wastes, both in-pit and out-of-pit. Syncrude’s reclamation program has been continually modified in response to regulatory changes and to information derived from monitoring. Some of the strategies have met with substantial success, in terms of forest ecosystem initiation, while others have not. Reclamation techniques that have been employed by Syncrude can be summarized as follows:

Reclamation Surface Amendment for Surface Amendment for Revegetation Species Technique Overburden Tailings Sand A None 15 cm peat Agronomic grass/legume seed mix B 10 to 20 cm peat 15 cm peat over 10 cm clay Agronomic grass/legume; wide variety of trees & shrubs C 1 m Holocene/Pleistocene 15 cm peat over 10 cm clay Low rates of material on (assumed) grass/legume; wide saline/sodic overburden variety of trees & shrubs D 1 m suitable surface material on 50 cm ± 20 cm suitable Emphasis on trees - (assumed) saline/sodic surface material few species, no overburden seeding except on steep slopes. E 50 cm suitable surface material 70 cm±10 cm suitable Oats/barley nurse crop; on non-saline / non-sodic surface material emphasis on trees - overburden (construction specs) few species (monoculture plots) F 50 cm suitable surface material 35 to 50 cm suitable surface Barley nurse crop; on non-saline / non-sodic material to create class 2/3 emphasis on trees - 1 overburden (construction specs) soil capability for forestry conifer + 1 deciduous (poplar or dogwood)

F1.3.1 Surface Preparation

Syncrude’s objective in soil reconstruction is to produce a self-sustaining soil cap over the overburden and tailings sand materials that meets or exceeds the predisturbance soil capability levels. The soils used include both the surface organic soils and the underlying mineral horizons that meet the “Suitable Soil” criteria. The suitable soils are delineated by soils surveys and auger drilling programs. The mineral soils often have some chemical and physical limitations such as high pH (7-8) and high clay contents. The underlying bedrock units of the Clearwater and McMurray Formations have high salinity and sodicity and are not used as reclamation materials. However, they make up the largest volumes in the overburden dumps and in-pit structures and require a thicker soil cap to prevent salt exsolution. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX F Page F-11 Prepared By: The Oil Sands Vegetation Reclamation Committee

The creation of more upland sites in the post mining area has enabled Syncrude to create a more productive site for forest development – one of the main vegetation objectives in the final landscape. The reclamation soils are used to cap the following upland mining structure:

· Tailing Dykes · External Sand Storage Dumps · External Overburden Dumps · In-pit Containment Structures

During the initial years (1978 to 1983) of reclamation at Syncrude, tailings sand slopes were reclaimed by capping with a nominal 10 cm layer of clayey overburden followed by a 15 cm layer of peat and incorporating these amendments into the sand using a cultivator and discs or a rotovator.

The second reclamation approach evaluated at Syncrude began in 1986 when the amendment material was applied as a single 50 cm cap. Amendment material used for reclamation was rated as having either Fair or Good reclamation suitability based on criteria set out by the Soil Quality Criteria Working Group (1987). After placement, the amendment was prepared using a rome disc bedder which left prominent ridges (30 to 50 cm high parallel to the contour on slopes).

Subsequent reclamation of tailings sand increased the depth of the amendment material to 70 cm and efforts were made to increase the surface organic matter content. For example, if the reclamation material placed on the site had less than 20% organic (peat) by volume, a nominal 15 cm top dressing of higher organic peat (50 to 100% by volume) material was placed at the surface. The thickness of the lower material was reduced to 55 cm such that the total amendment thickness was maintained at 70 cm. The current reclamation approach is based on the Land Capability Classification for Forest Ecosystems in the Oil Sands Region (Leskiw 1998). In general, according to this classification, a 35 to 50 cm cap of suitable surface soil can effectively provide a class 2 or 3 capability for forestry in the landscape.

In the initial years (1978-79) of overburden reclamation at Syncrude, no surface amendment was applied. However, due to poor revegetation success, these areas have subsequently (in 1990) been capped with a 100 cm cap of material rated as Fair or Good reclamation suitability. Areas of overburden reclaimed in 1982 and 1983 were capped with a 10 to 20 cm layer of peat. In 1984, overburden reclamation methodology was altered to include a 100 cm cap of Fair or Good reclamation suitability material placed over overburden materials that were assumed to be saline or sodic, followed by discing or plowing to incorporate fertilizer and break up lumps. In more recent years, where the overburben has been demonstrated to be non-saline and non-sodic, the thickness of the cap material has been reduced to 50 cm.

F1.3.2 Application of Fertilizer

In the early years of reclamation, fertilizer was applied after initial placement of reclamation material and followed by annual maintenance applications up until 1982. Since then, fertilizer Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX F Page F-12 Prepared By: The Oil Sands Vegetation Reclamation Committee has only been applied the first year except in rare situations where deficiency symptoms are evident and then only if there is not a competing grass sward. It has always been the practice to till the starter fertilizer into the surface amendment material by discing in order to incorporate the phosphate and organics and to smooth the surface.

F1.3.3 Revegetation

The main objectives of vegetation establishment at Syncrude are:

· To create a natural self-sustaining ecosystem, · To stabilize slopes and prevent erosion, and · To meet end land use objectives as agreed upon by the stakeholders.

Revegetation techniques at Syncrude closely parallel those described above for Suncor. In the early years, agronomic grasses and legumes were applied at high rates, a practice that was discontinued for all areas in 1985 in an effort to promote revegetation of woody species. From 1986 to 1989 reclaimed areas were not seeded. This resulted in a significant amount of erosion in some sloped areas. Consequently, beginning in 1990, reclaimed slopes of tailings sand or overburden have been seeded with an annual barley or oats crop whereas relatively level areas have not been seeded. Exceptions to this approach are the large areas of relatively level overburden and tailings sand that have been revegetated with species that provide forage for bison.

Revegetation techniques for woody species is also very similar to the approach described above for Suncor, although Syncrude places more emphasis on trees than shrubs in their planting program. Dominant species are jack pine, white spruce, aspen and red-osier dogwood.

F1.3.4 Monitoring

Syncrude’s reclamation monitoring program has been continually modified in response to regulatory changes and to information derived from monitoring. The program will continue to change as more information becomes available and corporate regulatory philosophies mature. A 20 m x 20 m tree assessment plot is established in each reclaimed area designated for forestry to monitor tree growth and survival. Tree measurements are taken in the first year and then at five year intervals. The assessment plots are also used to monitor native species establishment and biodiversity. A soil monitoring program has also been established to measure soil thickness and chemical and physical parameters of the reclaimed soils. This also provides the opportunity to classify the soil based on the Land Capability Classification of Forest Ecosytems in The Oil Sands Region (Leskiew 1998).

F2. REVEGETATION SUCCESS

F2.1 RECLAIMED TAILINGS SAND

A detailed study of vegetation on reclaimed tailings sand structures was undertaken in 1995 at both Suncor and Syncrude. For comparison, study plots (measuring 10 x 10 m) were also sampled in adjacent natural stands of the four dominant upland vegetation community types in the vicinity. A summary of the plots sampled is shown in Table F.1. Table F.2 shows the Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX F Page F-13 Prepared By: The Oil Sands Vegetation Reclamation Committee average number of species and percentage cover of four main vegetation groups, while the percentage cover of all species encountered is provided in Tables F.3 and F.4.

By far the greatest number of sites assessed in reclaimed areas (47 at Suncor, 23 at Syncrude) were relatively old (14 to 24 years) where the peat amendment was incorporated into the sand and an agronomic grass/legume mix was applied Table F.1. Only 12 sites were assessed where the current reclamation approach was implemented whereby the amendment is applied as a cap and only an annual barley is seeded the first year. These sites range from 4 to 10 years old and are referred to as "No Seed" in Table F.1. Only 2 sites occurred on level terrain (Dyke 2 Plateau). Thickness of the surface amended soil layer ranged from an average 16 to 33 cm except for the non-seeded sites on the Mildred Lake Settling Basin (MLSB) at 53 cm. Thickness of the surface litter layer (undecomposed plant material) averaged 1 cm or less at all sites. Rooting depth (of the herbaceous vegetation) averaged 32 to 53 cm, with no obvious relationship with site age or treatment. It should be noted that there is a considerable amount of subjectiveness in this measurement.

Characteristics of the natural forest stands differed from the reclaimed sites in several ways. Most importantly, all of the natural sites were on level terrain. Thickness of the organic soil layer averaged 14 to 22 cm for the mixedwood sites, comparable to reclaimed areas, but only 2 to 4 cm at the jack pine and aspen forest communities. Litter layer thickness ranging from 2 to 8 cm was considerably greater than reclaimed sites, but rooting depth of the herbaceous species was comparable averaging 42 to 50 cm.

The relatively high numbers of grasses and legumes on reclaimed sites was the most obvious difference compared to the natural stands. Not surprisingly, this difference was most apparent when comparing the reclaimed sites that were seeded, where grasses and legumes comprised the greatest number of species. In contrast, other herbaceous species were the most abundant group among the natural communities. The average percentage cover per plot (Table F.2) clearly illustrates the dominance of grasses and legumes in seeded reclaimed areas whereas trees provided most of the cover in natural stands. Grass and legume cover ranged from 50 to 100 percent on seeded sites but was virtually absent in natural forest stands. Trees provided less than 10 percent cover at seeded sites and between 20 and 30 percent cover at older non- seeded sites (Dyke 2 Plateau and MLSB), compared with 55 to 90 percent cover in the natural forest communities.

The percentage cover of other herbaceous species was comparable between the non-seeded and natural stands, but the actual species providing the cover were different. Fireweed, sow thistle and sedge provided almost all of this cover on the reclaimed sites but were virtually absent in the natural stands (Table F.4).

The similarity between the reclaimed sites and each of the natural forest communities was assessed by calculating the "coefficient of community" for each reclaimed site and natural forest site. This coefficient provides a numerical comparison of the number of species common to both sites. The value of the coefficient for each of the reclaimed site groups compared with each of the natural forest communities is given in Table F.5. In general, the results indicate there was little similarity in terms of species composition between any of the reclaimed areas with the natural stands. The oldest reclaimed sites on Suncor’s Tar Island Dyke (TID) and Syncrude’s MLSB seeded to grasses and legumes typically had values of 0.10 or less. Sites seeded to native grasses (Dyke 2) and sites not seeded (TID and MLSB) had slightly higher values, Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX F Page F-14 Prepared By: The Oil Sands Vegetation Reclamation Committee typically between 0.1 and 0.2 or more. In most cases, the species that were common between the sites were the trees and shrubs planted as part of the reclamation program.

F2.2 RECLAIMED OVERBURDEN

A study using consistent methodology has not been conducted to compare vegetation establishment on reclaimed overburden at the two minesites and in nearby natural forest stands. However, both Suncor and Syncrude have conducted long-term monitoring of vegetation in these areas. Table F.6 presents data on the average number of species and percentage cover from a variety of sites monitored on reclaimed overburden. The data has been separated into the same four vegetation groups as presented above for the reclaimed tailings sand sites.

All of the sites monitored and presented in Table F.6 occur on overburden slopes with an average 35% slope angle. The sites monitored at Suncor were typically reclaimed by placing a 15 to 20 cm cap while those at Syncrude were capped with up to 100 cm of amendment material.

The data show a similar trend to reclaimed tailings sand with respect to the influence of seeding on long-term vegetation cover. At both Suncor and Syncrude, sites that were seeded were dominated by grasses and legumes in terms of percentage cover. This is most evident for the Suncor sites where trees, shrubs and other herbaceous species were almost non-existent at the seeded sites even after 15 to 17 years. In contrast, all four groups were well represented at the sites not seeded. The difference at the Syncrude sites was less apparent. However, it is important to note that the methodology for vegetation cover monitoring is different at Syncrude and therefore the data should not be compared directly with the Suncor data. Trees are not included in the Syncrude survey and much more area is assessed per site compared to Suncor.

F2.3 SUMMARY OF VEGETATION ESTABLISHMENT IN RECLAIMED AREAS

The grass and legume mixtures that were seeded in the early years of reclamation at both Suncor and Syncrude were highly effective in establishing an erosion resistant vegetation cover on reclaimed overburden and tailings sand. These vegetation communities have persisted for over 20 years and have resisted the establishment of native species either through natural invasion or planting programs. The dominant species at these reclaimed sites are typically the seeded agronomics: awnless bromegrass, creeping red fescue, and alfalfa. Natural invasion is occurring very slowly. It is important to note however, that there are areas up to 0.5 ha in size where dense stands of trees and shrubs have been successfully established at both minesites in these formerly seeded areas.

Sites that were not seeded or only seeded to an annual barley have typically become dominated by a variety of herbaceous species that provide close to 100% total areal cover within a few years after reclamation. Dominant species include perennial sow thistle, fireweed, sweet clover and hawksbeard. The seeded barley provides effective erosion control during the first year and in most instances, the invading species maintain this control in the following years. Trembling aspen, balsam poplar and a variety of willows and other native shrubs frequently invade these areas within a few years of reclamation. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX F Page F-15 Prepared By: The Oil Sands Vegetation Reclamation Committee

Table F.1 Vegetation Plots Sampled on Reclaimed Tailings Sand in 1995 at Suncor, Syncrude, and Nearby Natural Forest Stands.

Suncor Syncrude Natural Forest Stands TID TID Dyke 2 MLSB Spruce Dyke 2 TID MLSB Mixed- Seeded Seeded Plateau 2.3.1.2.1.1 Dominated Jack Pine Aspen Seeded No Seed Seeded wood (old) (base) No Seed o Seed Mixedwood Number of Plots 47 6 8 2 3 23 7 2 2 2 2 Age* (years) 20-24 13-15 7-12 10 4 14-15 8-9 144 198 51 98 Topographical Position 40% 40% 40% slope Level 40% 27% 27% level level level level slope slope slope slope slope Organic Soil Thickness 19 16 23 26 33 31 53 14 22 2 4 (cm) Litter Thickness (cm) 1 <1 1 1 <1 1 1 8 4 2 4 Rooting Depth (cm) 53 49 35 32 34 45 40 46 43 50 42

* Age of Syncrude and Suncor plots is based on the year the site was reclaimed. Age of Natural Forest Stands is based on the oldest tree cored in the stand.

Note: TID, Dyke 2, Dyke 2 Plateau and MLSB are tailings sand structures at Suncor and Syncrude. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX F Page F-16 Prepared By: The Oil Sands Vegetation Reclamation Committee

Table F.2 Average Number of Species and Percentage Cover in Plots Sampled on Reclaimed Tailings Sand in 1995 at Suncor, Syncrude, and Nearby Natural Forest Stands

Suncor Syncrude Natural Forest Stands TID TID Dyke 2 Spruce Dyke 2 TID MLSB MLSB Mixed- Seeded Seeded Plateau Dominated Jack Pine Aspen Seeded No Seed Seeded No Seed wood (old) (base) No Seed Mixedwood Average No. Species: Trees 1.9 0.7 2.4 2.3 2.0 0.8 2.4 3.5 4.0 2.5 2.5 Shrubs 0.8 1.2 2.1 1.3 2.5 1.5 1.9 3.5 5.5 1.0 4.5 Grasses/Legumes 4.2 4.5 4.9 2.7 2.0 4.0 3.4 0.0 0.0 1.0 0.5 Other Herbaceous 3.3 2.0 1.6 6.3 4.5 2.7 3.1 5.5 9.5 2.5 7.0

Average Cover (%): Trees 11.9 2.2 4.6 5.2 23.5 10.3 26.6 63.0 90.0 63.0 55.3 Shrubs 6.7 1.6 3.1 2.2 49.5 15.6 10.0 33.8 38.5 15.0 34.0 Grasses/Legumes 52.6 99.6 85.4 2.7 8.5 85.4 39.1 0.0 0.0 3.0 1.0 Other Herbaceous 10.6 8.9 1.7 107.8 43.8 14.0 13.7 43.5 45.3 33.5 42.3 Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX F Page F-17 Prepared By: The Oil Sands Vegetation Reclamation Committee

Table F.3 Average Percentage Cover of Tree and Shrub Species in Plots Sampled on Reclaimed Tailings Sand in 1995 at Suncor, Syncrude, and Nearby Natural Forest Stands

Suncor Syncrude Natural Forest Stands TID TID Dyke 2 Dyke 2 TID MLSB MLSB Mixed- Spruce Seeded Seeded Seeded Plateau No Seed Seeded No Seed wood Dominated Jack Pine Aspen (old) (base) No Seed Mixedwood Trees Alder sp. 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 2.5 25.0 7.5 Aspen 4.0 0.0 0.0 0.0 1.3 0.6 15.7 12.5 12.5 0.5 40.0 Balsam Fir 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.5 5.0 0.0 0.0 Balsam Poplar 1.1 0.2 1.8 22.5 0.2 0.1 0.9 7.5 1.5 0.0 0.3 Green Ash 0.5 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Jack Pine 1.3 0.0 1.9 0.0 0.0 7.5 4.3 0.0 0.0 37.5 0.0 Lodgepole Pine 3.0 0.2 0.3 0.5 3.7 0.0 0.0 0.0 0.0 0.0 0.0 Manitoba Maple 0.4 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Siberian Elm 0.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Tamarack 0.0 0.0 0.0 0.0 0.0 0.5 0.0 0.0 0.0 0.0 0.0 White Birch 0.1 1.7 0.0 0.0 0.0 0.0 0.0 12.5 1.0 0.0 0.0 White Spruce 1.2 0.2 0.6 0.5 0.0 1.6 5.7 30.0 67.5 0.0 7.5 Shrubs Blueberry sp. 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 2.5 15.0 7.5 Canada Buffaloberry 0.0 0.0 0.0 0.0 0.0 0.2 0.0 0.0 1.5 0.0 1.5 Chokecherry 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Common Caragana 3.6 0.0 0.0 0.0 0.0 11.2 0.0 0.0 0.0 0.0 0.0 Gooseberry 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.5 0.0 0.0 0.0 Labrador Tea 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 20.0 Low Bush Cranberry 0.0 0.0 0.0 0.0 0.0 0.0 0.0 14.5 5.0 0.0 0.0 Raspberry sp. 0.0 0.0 0.0 0.0 1.0 1.1 0.0 0.0 0.3 0.0 0.0 Red-osier Dogwood 0.3 1.1 0.1 1.0 0.0 2.6 6.6 20.0 17.5 0.0 0.0 Rose sp. 0.0 0.0 1.8 0.5 0.2 0.0 0.1 10.3 4.0 0.0 2.5 Saskatoon 0.0 0.3 0.8 0.0 0.0 0.4 0.0 0.0 0.3 0.0 1.5 Shrubby Cinquefoil 0.0 0.3 0.0 0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.0 Silverberry 0.0 0.0 0.0 0.5 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Snowberry 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1.5 5.0 0.0 0.0 Willow sp. 2.6 0.0 0.4 47.5 1.0 0.1 3.3 0.0 2.5 0.0 1.0 Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX F Page F-18 Prepared By: The Oil Sands Vegetation Reclamation Committee

Table F.4 Average Percentage Cover of Herbaceous and Nonvascular Species in Plots Sampled on Reclaimed Tailings Sand in 1995 at Suncor, Syncrude, and Nearby Natural Forest Stands

Suncor Syncrude Natural Forest Stands TID TID Dyke 2 TID MLSB Dyke 2 MLSB Mixed- Spruce Dominated Seeded Seeded Plateau No No Jack Pine Aspen Seeded Seeded Wood Mixedwood (Old) (Base) No Seed Seed Seed Grasses/Legumes Alfalfa 12.7 6.7 9.3 0.0 0.0 16.6 13.6 0.0 0.0 0.0 0.0 Alsike Clover 0.2 0.1 3.9 0.0 0.0 1.2 1.4 0.0 0.0 0.0 0.0 Awnless Brome 23.4 40.2 1.9 0.0 0.2 22.5 3.1 0.0 0.0 0.0 0.0 Bent Grass sp. 0.0 0.0 0.0 0.0 0.0 2.2 0.0 0.0 0.0 0.0 0.0 Birdsfoot Trefoil 0.0 0.7 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Cicer Milk Vetch 0.0 33.3 0.0 0.3 0.0 1.7 0.1 0.0 0.0 0.0 0.0 Crested Wheatgrass 1.3 1.7 0.0 0.0 0.0 7.2 0.0 0.0 0.0 0.0 0.0 Foxtail Barley 0.0 0.0 0.1 0.0 0.3 0.0 0.1 0.0 0.0 0.0 0.0 Hair Grass 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Kentucky Bluegrass 2.2 0.0 1.5 7.5 0.3 0.2 2.9 0.0 0.0 3.0 0.5 Red Clover 0.0 0.0 0.0 0.0 0.0 00 0.0 0.0 0.0 0.0 0.0 Red Fescue 11.9 16.5 65.3 0.5 0.0 33.5 5.7 0.0 0.0 0.0 0.0 Sweet Clover sp. 0.1 0.0 0.1 0.0 1.0 1.2 11.4 0.0 0.0 0.0 0.0 Timothy 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Wheatgrass sp. 1.2 0.5 3.3 0.0 0.8 1.4 0.7 0.0 0.0 0.0 0.0 White Clover 0.0 0.0 0.0 0.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX F Page F-19 Prepared By: The Oil Sands Vegetation Reclamation Committee

Table F.4 (cont’d)

Suncor Syncrude Natural Forest Stands TID TID Dyke 2 TID MLSB Dyke 2 MLSB Mixed- Spruce Dominated Seeded Seeded Plateau No No Jack Pine Aspen Seeded Seeded Wood Mixedwood (Old) (Base) No Seed Seed Seed Other Herbaceous Aster sp. 0.1 0.0 0.0 0.0 0.2 0.1 0.1 1.0 1.0 0.0 0.3 Bearberry 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 25.0 15.0 Bishop's Cap 0.0 0.0 0.0 0.0 0.0 0.0 0.0 5.0 4.0 0.0 0.0 Bluebell 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Bog Cranberry 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 8.0 5.0 Bunchberry 0.0 0.0 0.0 0.0 0.0 0.0 0.0 7.5 10.0 0.0 5.0 Canada Thistle 0.0 0.0 0.0 0.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Chickweed sp. 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Club Moss sp. 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 7.5 10.0 Coltsfoot 0.0 0.0 0.0 0.0 0.0 0.0 0.0 2.5 0.0 0.0 2.5 Columbine sp. 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1.0 0.0 0.0 Common Dandelion 5.2 0.0 0.0 0.0 1.3 0.3 2.1 0.0 0.0 0.0 0.0 Fireweed 2.8 2.5 0.3 2.0 40.0 1.0 2.6 0.0 0.0 0.0 0.3 Geranium sp. 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Goldenrod sp. 0.0 0.2 0.0 0.0 0.2 0.0 0.0 0.0 0.0 0.0 0.0 Grape Fern 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Hawkweed sp. 0.2 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Hawk's Beard sp. 0.2 0.1 0.0 0.0 4.7 0.0 0.9 0.0 0.0 0.0 0.0 Horsetail sp. 0.0 0.0 0.0 0.0 0.7 0.0 0.0 12.5 14.0 0.0 0.0 Lamb's Quarter 0.0 0.0 0.3 0.0 1.7 0.0 0.1 0.0 0.0 0.0 0.0 Mustard sp. 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Northern Bedstraw 0.0 0.0 0.0 0.0 0.0 0.0 0.3 0.5 4.5 0.0 0.0 Northern Twinflower 0.0 0.0 0.0 0.0 0.0 0.0 0.0 10.0 3.5 0.0 10.0 Pea Vine 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.5 Pepper Grass 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX F Page F-20 Prepared By: The Oil Sands Vegetation Reclamation Committee

Table F.4 (cont’d)

Suncor Syncrude Natural Forest Stands TID TID Dyke 2 TID MLSB Dyke 2 MLSB Mixed- Spruce Dominated Seeded Seeded Plateau No No Jack Pine Aspen Seeded Seeded Wood Mixedwood (Old) (Base) No Seed Seed Seed Other Herbaceous Perennial Sow Thistle 1.6 5.3 0.9 0.3 58.3 8.8 6.6 0.0 0.0 0.0 0.0 Cinquefoil sp. 0.0 0.0 0.0 0.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Rush sp. 0.0 0.2 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Sedge sp. 0.0 0.5 0.0 40.5 0.8 0.0 0.0 0.0 0.0 0.0 0.0 Stinging Nettle 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Strawberry sp. 0.0 0.0 0.0 0.3 0.0 0.1 0.0 0.5 2.5 0.0 0.0 Tall Lungwort 0.0 0.0 0.0 0.0 0.0 0.0 0.0 3.5 2.8 0.0 0.0 Tartary Buckwheat 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Unidentified sp. 0.1 0.0 0.1 0.0 0.0 0.1 0.0 0.0 0.0 0.0 1.3 Violet sp. 0.0 0.0 0.0 0.0 0.0 0.1 0.0 0.5 0.0 0.0 0.0 Wintergreen sp. 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 2.0 0.5 2.5 Yarrow sp. 0.2 0.2 0.1 0.3 0.0 1.3 1.0 0.0 0.0 0.0 0.0 Moss and Lichen Moss sp. 2.69 10.3 14.6 2.8 10.7 4.1 16.7 17.5 25.0 10.0 2.5 Lichen sp. 2.6 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Bare Ground 2.4 4.1 0.9 0.0 15.0 4.3 33.9 0.0 0.0 0.5 0.0 Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX F Page F-21 Prepared By: The Oil Sands Vegetation Reclamation Committee

Table F.5 Average Coefficient of Community* for Reclaimed Tailings Sand Sites Compared with Natural Forest Stands

Reclaimed Sites Natural Suncor Syncrude Forest Site Type TID Seeded TID Seeded Dyke 2 TID Dyke 2 Plateau MLSB MLSB (old) (base) Seeded No Seed No Seed Seeded No Seed Mixedwood 2 0.10 0.10 0.24 0.08 0.29 0.05 0.21 X 0.06 0.03 0.11 0.12 0.08 0.04 0.14 Spruce Dominated 13 0.10 0.08 0.19 0.16 0.19 0.08 0.21 Mixedwood 16 0.10 0.05 0.16 0.18 0.22 0.07 0.26 Jack Pine 18 0.11 0.00 0.13 0.17 0.06 0.07 0.22 19 0.10 0.00 0.13 0.07 0.06 0.05 0.11 Aspen 14 0.12 0.04 0.21 0.21 0.22 0.07 0.22 22 0.11 0.07 0.19 0.19 0.15 0.07 0.17 Number of Sites Averaged 47 6 8 3 2 23 7

* Coefficient of Community = 2C÷(S1 + S2) where C = Number of common species; S1 = Number of species in the Natural Forest Site; and S2 = Number of species in the Reclaimed Site. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX F Page F-22 Prepared By: The Oil Sands Vegetation Reclamation Committee

Table F.6 Average Number of Species and Percentage Cover in Plots Sampled On Reclaimed Overburden at Suncor and Syncrude

Suncor Syncrude Waste Area 5, Waste Area 8, Waste Area 16 Waste Area Waste Area Waste Area 16 Waste Area 19 No Seed S-4 Seeded S-4 No Seed Seeded No Seed Number of Sites 5 8 16 7 2 Age (Years) 15 - 17 10 - 15 6 - 9 10 10, 11

Average No. Species: Trees 0.0 1.0 2.8 N/A* N/A Shrubs 0.0 0.9 7.6 1.1 0.0 Grasses/Legumes 3.6 2.1 2.1 3.3 3.0 Other Herbaceous 0.6 2.5 3.6 5.6 9.5

Average Cover (%): Trees 0.0 23.7 27.1 N/A N/A Shrubs 0.0 6.9 14.5 7.6 0.0 Grasses/Legumes 90.9 32.8 30.4 43.3 32.5 Other Herbaceous 1.9 13.4 34.0 32.0 45.0

*N/A = Not Applicable (Trees are not included in the vegetation cover monitoring assessment conducted by Syncrude) Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX F Page F-23 Prepared By: The Oil Sands Vegetation Reclamation Committee

F3. LITERATURE CITED

Alberta Environmental Protection. 1994. The Alberta Regeneration Survey Manual (Addendum April 1997). Publication No.: Ref. 70.

Hardy BBT Limited. 1990. Natural Plant Invasion into Reclaimed Oil Sands Mine Sites. Alberta Land Conservation and Reclamation Council Report No. RRTAC 90-3. 65 pp.

Leskiw, L.A. 1998. Land Capability Classification for Forest Ecosystems in the Oil Sands Region. Revised Edition. Prepared by Tailings Sands Reclamation Practices Working Group.

Soil Quality Criteria Working Group. 1987. Soil Quality Criteria Relative to Disturbance and Reclamation. Revised. Edmonton, AB. 56 pp. APPENDIX G

SEED ZONES AND SOURCES Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX G Page G-1 Prepared By: The Oil Sands Vegetation Reclamation Committee

TABLE OF CONTENTS

PAGE

G1. SEED ZONES AND SOURCES...... 2

G2. DRAFT SEED ZONE DESCRIPTIONS...... 2 G2.1 DRAFT SEED ZONE 8A...... 2 G2.1.1 Geography...... 2 G2.1.2 Topography and Soils ...... 5 G2.1.3 Climate ...... 5 G2.1.4 Ecology...... 5 G2.1.5 Summary...... 5 G2.2 DRAFT SEED ZONE 9B...... 6 G2.2.1 Geography...... 6 G2.2.2 Topography and Soils ...... 6 G2.2.3 Climate ...... 6 G2.2.4 Ecology...... 6 G2.2.5 Summary...... 7 G2.3 DRAFT SEED ZONE 10B...... 7 G2.3.1 Geography...... 7 G2.3.2 Topography and Soils ...... 8 G2.3.3 Climate ...... 8 G2.3.4 Ecology...... 8 G2.3.5 Summary...... 8 G2.4 DRAFT SEED ZONE 11A...... 9 G2.4.1 Geography...... 9 G2.4.2 Topography and Soils ...... 9 G2.4.3 Climate ...... 10 G2.4.4 Ecology...... 10 G2.4.5 Summary...... 10

G3. LITERATURE CITED ...... 12 Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX G Page G-2 Prepared By: The Oil Sands Vegetation Reclamation Committee

SEED ZONES AND SOURCES

Prepared by: Bert Ciesielski, Alberta Environmental Protection

G1. SEED ZONES AND SOURCES

Current provincial seed provenance rules require seed or propagule material used on Crown lands to be native to the Natural Subregion and be obtained from within 50 miles (80 km) and within 500 feet (150 m) elevation from the planting site. All tree seed collected for use on Crown land must be registered with the Land and Forest Service, Tree Improvement Center. All trees planted on Crown land must come from registered seed sources.

The Forest Management Division is currently reviewing the provincial seed provenance rules to more definitively show the locations of zones where suitable seed can be obtained. Publication is expected in the fall of 1998. The issue of movement across zone boundaries is still to be discussed and will vary depending on the type of boundary. Where genetic and ecological gradients are steep or abrupt, or zones are bounded by natural subregion boundaries, movement across zone boundaries will differ from those within relatively homogenous Natural Subregions. The initial discussion has been to use allowable movement within a township (6 miles, 9.6 km) across seed zone boundaries within Boreal Natural Subregions. Natural Regions and Subregions of Alberta are summarized in Alberta Environmental Protection (1994).

The current draft seed zone map for the Athabasca Oil Sands Region is shown in Figure G.1. The new seed zones were developed for trees but are also intended to cover shrubs and provide a framework for herbaceous vegetation as well. Descriptions of the new seed zones in the Athabasca oil sands region follows.

G2. DRAFT SEED ZONE DESCRIPTIONS

G2.1 DRAFT SEED ZONE 8A

G2.1.1 Geography

This seed zone is situated in northeastern Alberta and consists of an area east of Fort McMurray. The zone extends from 58 05' N to 55 49' N latitude, and from 110 00' W to 111 49' W longitude. Elevations range from 300 to 660 metres and the total area is approximately 17,040 km2. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX G Page G-3 Prepared By: The Oil Sands Vegetation Reclamation Committee

Figure G.1 Draft Provincial Seed Zone Map for the Athabasca Oil Sands Region Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX G Page G-4 Prepared By: The Oil Sands Vegetation Reclamation Committee

Table G.1 Summary of Climatic Data for Draft Seed Zone 8A Stations

Climate Mean Mean Mean Mean Frost Growing Day Station June- Summer Annual Annual Free Degree Length And August Precip Temp Precip Period Days (Longest) Elevation Temp J-J-A >10°C (mm) (°C) (mm) (Days) (Hours) (°C) (Days)

Anzac 14.6 224.5 0.0 503.5 93 527.6 18 (495 m)

Bitumont 14.9 217.7 - - - - 18 LO (349 m)

Cowpar 14.6 250.2 - - 105 - 18 LO (563 m)

Fort 15.1 216.1 -0.2 471.9 84 575.8 18 McMurray (369 m)

Gordon 14.6 233.6 - - 96 - 18 Lakes LO (488 m)

Johnson 14.1 227.5 - - 93 - 18 Lakes LO (549 m)

Mildred - - - - 88 - 18 Lake (310 m)

Muskeg 14.0 234.0 - - 101 - 18 Mountain LO (652 m)

Tar Island 15.9 178.4 -0.1 408.5 109 686.5 18 (288 m)

Mean 14.8 221.1 -0.1 461.3 95.4 596.6 18 Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX G Page G-5 Prepared By: The Oil Sands Vegetation Reclamation Committee

G2.1.2 Topography and Soils

The topography for most of the zone is generally level with areas of gently rolling to rolling terrain. Parent materials are commonly a mixture of sandy alluvial glacial till and gravelly outwash. Eutric Brunisols and Podzols are common on these coarse-textured upland sites and occupy approximately 40% of the zones land area. Gray Luvisolic soils occur occasionally on finer textured upland till sites. Organic and Gleysolic soils represent approximately 60% of the land area and are located in low lying areas. Peat depths of up several feet have been measured, and permafrost has been encountered.

G2.1.3 Climate

There are several climatic stations for this area (Table G.1) with three stations recording annual data. The remaining seasonal stations have a mean June-August temperature of 14.8°C and average total precipitation for the same period of 221.1 mm. The mean annual temperature is - 0.1°C with an annual total precipitation of 461.3 mm. Growing season temperatures appear typical for core areas of the Central Mixedwood Subregion but with growing season amounts appear to be higher.

G2.1.4 Ecology

Pure jack pine forest types on coarse textured mineral soils make up a large portion of the vegetation in this zone. Understories are predominantly composed of lichens with lesser amounts of bearberry, low bilberry, bog cranberry and prickly rose. Mixed jack pine and black spruce types in upland to bog transition areas are also common. Mixed black spruce and tamarack forest types which occupy extensive areas of bog have a ground cover of Labrador tea and peat mosses or when richer fens are present, dwarf birch, sedges and brown mosses. Areas north of the Clearwater River and east of the Athabasca River are dominated by these types.

Early successional trembling aspen and balsam poplar forest types which occur on finer textured upland sites in the southern portion of the seed zone and along the Athabasca River typically have understory species including low bush cranberry, beaked hazelnut, prickly rose, red osier dogwood, march reed grass, sarsaparilla, dewberry, cream-coloured peavine, pink wintergreen and twin flower. Later successional white spruce forest types which are generally restricted to areas along the Clearwater and Athabasca River valleys, have a greater proportion of feather mosses in the understory. Small amounts of paper birch occur throughout the zone and occasionally forms pure stands.

G2.1.5 Summary

Pure jack pine forest types rather than mixedwood are most common in this seed zone. The terrain is variable with organic depressions and areas of dunes occurring within a matrix of shallow sandy soils. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX G Page G-6 Prepared By: The Oil Sands Vegetation Reclamation Committee

G2.2 DRAFT SEED ZONE 9B

G2.2.1 Geography

This seed zone is situated in northeastern Alberta and consists of the area west of Fort McMurray including the Thickwood Hills. The zone extends from 58 05' N to 56 20' N latitude, and from 111 05' W to 114 55' W longitude. The elevation range is from 250 to 600 metres and the total area is approximately 13,970 km2. Its boundaries fall within the Central Mixedwood Subregion of the Boreal Forest Natural Region.

G2.2.2 Topography and Soils

Most of this seed zone has level to depressional topography with areas of gently rolling ridges. Parent materials are primarily fine textured undulating lacustrine plain and sandy and gravelly tills. Poorly to imperfectly drained organic and Gleysolic soils occupy approximately 60% of the land area and peat depths of several feet have been recorded. Occasional permafrost occurs in these organic soils at 40 and 50 centimetres below the peat surface.

Mineral soils on better drained sites have Gray Luvisols on fine textured materials and Eutric Brunisols or Podzols on coarse textured materials. These mineral soils are generally in the form of islands surrounded by organic soils. Although it does not represent a large area, the Athabasca River valley is an important topographical feature in this zone and is associated with fluvial benches, level to depressional areas, gently rolling terrain and sand dunes.

G2.2.3 Climate

Five meteorological stations represent this area (Table G.2). Of these, only Fort McMurray records annual climate data. The mean June-August temperature of 14.4°C appears fairly typical for the Central Mixedwood Subregion but the average total June-August precipitation of 219.5 mm for the growing season and annual precipitation of 471.9 mm indicates greater precipitation. The mean annual temperature for Fort McMurray of -0.2°C is slightly lower than the 0.5°C reported as typical for the Central Mixedwood Subregion. The frost free period of 86.3 days and growing degree days above 10°C of 575.8 is also fairly typical for core boreal mixedwood areas.

G2.2.4 Ecology

Trembling aspen, balsam poplar and white spruce mixedwood forest types make up a major portion of the vegetation in this zone, particularly in well drained areas along the Wabasca and Athabasca rivers. These forest types are usually dominated initially by the poplars which are increasingly replaced by white spruce and balsam fir over time. Understories in these mixedwood types commonly include lowbush cranberry, beaked hazelnut, prickly rose, red-osier dogwood, marsh reed grass, sarsaparilla, dewberry, cream colored peavine, pink wintergreen and twin flower. Feather mosses are also present and their cover increases with increasing stand age and conifer composition.

Pure stands of balsam poplar occur along watercourses and in wet depressions. Paper birch is commonly present as a minor component in mixedwood stands and occasionally forms pure stands.

Sparse black spruce forest types are found on nutrient poor acidic bogs which are common in the central part of this zone between the Wabasca and Athabasca rivers. This treed muskeg Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX G Page G-7 Prepared By: The Oil Sands Vegetation Reclamation Committee commonly has a ground cover of Labrador tea and sphagnum mosses. On richer fens, tamarack, dwarf birches, sedges and brown mosses are more common.

On dry sandy upland sites, jack pine forest types occupy the dunes and alluvial deposits and commonly have understories of lichen bearberry, low bilberry, bog cranberry and prickly rose.

G2.2.5 Summary

This seed zone is fairly typical of the core area of the Central Mixedwood Subregion. Forest types along the Wabasca and Athabasca rivers are the most productive due to better drainage while a large area in the central part of the zone is dominated by unproductive treed muskeg.

Table G.2 Summary of Climatic Data for Draft Seed Zone 9B Station

Climate Mean Mean Mean Mean Frost Growin Day Station June- Summer Annual Annual Free g Length And August Precip Temp Precip Period Degree (Longest) Elevation Temp J-J-A Days (°C) (mm) (Days) >10°C (Hours) (°C) (mm) (Days)

Chipewyan 14.1 188.9 - - 103 - 18 Lakes LO (369 m)

Ells River LO 14.2 212.9 - - 78 - 18 (610 m)

Fort 15.1 216.1 -0.2 471.9 84 575.8 18 McMurray (369 m)

Thickwood 14.2 259.9 - - 87 - 18 Hills LO (604 m)

Mildred - - - - 88 - 18 Lakes (310 m)

Mean 14.4 219.5 -0.2 471.9 86.3 575.8 18

G2.3 DRAFT SEED ZONE 10B

G2.3.1 Geography

This seed zone is situated in northeastern Alberta and consists of the low lying areas north of the Pelican Mountains, south of the Peerless Lake Highlands and east toward Fort McMurray. The zone extends from 55 10' N to 56 45' N latitude, and from 110 45' W to 115 15' W longitude. Elevations range from 360 to 700 metres with a total area of approximately 17,740 km2. Its boundaries correspond to these of the Central Mixedwood Subregion of the Boreal Forest Natural Region. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX G Page G-8 Prepared By: The Oil Sands Vegetation Reclamation Committee

G2.3.2 Topography and Soils

Most of this seed zone has level to depressional topography with some gently rolling ridges. Parent materials are primarily sandy alluvium and gravelly outwash materials associated with undulating morainal plain. Gray Luvisolic soils are most prevalent on fine textured parent materials, while Eutric Brunisols soils are common on coarse textured upland sites. The majority of mineral soils occur as islands surrounded by poorly drained organic and Gleysolic soils which occupy approximately 75% of the total area. Peat depth of up to three metres has been measured and permafrost has occasionally been encountered.

The zone is dissected by the Athabasca River which is associated with level to depressional bog areas, gently rolling terrain, dunes and fluvial benches. Soils covering the zone range from poorly drained organics to rapidly drained Brunisols and Podzols soils, which are associated with these landforms.

G2.3.3 Climate

There is climatic information from both seasonal and yearly meteorological stations for this zone (Table G.3). These stations have a mean June-August temperature of 14.6°C and an average total precipitation of 227.8 mm for the same period. The frost free period is 98 days long and the number of growing degree days above 10°C is 586.8 days. The mean annual temperature recorded at Fort McMurray and Wabasca is 0.2°C with mean annual precipitation of 466.1 mm.

G2.3.4 Ecology

The eastern two thirds of this zone are dominated by black spruce forest types on poorly drained acidic bogs with typical ground cover of Labrador tea and sphagnum mosses. Where nutrient conditions are richer, bog species are complemented or replaced by tamarack, dwarf birches, sedges and brown mosses.

Mixedwood forest types composed of varying amounts of trembling aspen, balsam poplar and white spruce are common in the western third of this zone and along the Athabasca River valley. Understory vegetation in the better drained types is commonly composed of species such as low bush cranberry, beaked hazelnut, prickly rose, red osier dogwood, marsh reed grass, sarsaparilla, dewberry, cream-coloured peavine, pink wintergreen and twin flower. As the deciduous component in these stands declines over time and as canopy closure occurs with the increase of later successional white spruce and balsam fir types, the proportion of feather mosses in the understory increase. These later successional types occur but are not widespread in the area due to frequent fires.

Pure stands of balsam poplar are common along the Athabasca River valley and along streams. Paper birch is a minor component in mixedwood stands throughout the zone and occasionally forms pure stands. Jack pine forest types are frequent on coarse textured upland sites and commonly have understories of reindeer lichen, bearberry, low bilberry, bog cranberry and prickly rose.

G2.3.5 Summary

This zone has a climate which is similar to Central Mixedwood areas to the north. It is warmer and drier during the growing season than core boreal mixedwood areas to the west and has coarser textured soils and poorer drainage. Areas in the west of this zone and along the Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX G Page G-9 Prepared By: The Oil Sands Vegetation Reclamation Committee

Athabasca River are productive forest lands but much of the zone is too poorly drained to support commercial stands.

Table G.3 Summary of Climatic Data for Draft Seed Zone 10B Stations

Climate Mean Mean Mean Mean Frost Growth Station June- Summer Annual Annual Free Degree Day And August Precip Temp Precip Period Days Length Elevation Temp J-J-A >10°C (Longest) (°C) (mm) (Days) (°C) (mm) (Days) (Hours)

Fort 15.1 216.1 -0.2 471.9 84 575.8 18 McMurray Airport (369 m)

Grande 14.1 238.8 - - 80 - 18 LO (533 m)

Livock 14.1 243.5 - - - - 18 LO (579 m)

Muskwa 14.3 223.4 - - 113 - 18 LO (625 m)

Wabasca 15.2 217.1 0.6 460.2 115 597.7 18 RS (545 m)

Mean 14.6 227.8 0.2 466.1 98 586.8 18.0

G2.4 DRAFT SEED ZONE 11A

G2.4.1 Geography

This seed zone is situated in northeastern Alberta and consists of the front slopes of the Birch Mountains which extends from 57 00' N to 58 12' N latitude, and from 111 44' W to 114 36' W longitude. It represents lower elevations of the Birch Mountains lying between 450 and 925 metres and has a total area of approximately 7,470 km2. Its boundaries correspond to the Boreal Highlands Subregion of the Boreal Forest Natural Region.

G2.4.2 Topography and Soils

The topography of the zone varies from gently rolling to hilly and parent materials are primarily clay loam to sandy loam, non-calcareous, glacial tills. Gray Luvisolic soils have developed on well-drained upland sites while Brunisols and Podzols occur on coarser textured alluvial materials. Organic and Gleysolic soils are found in wet depressional areas and cover about 30% of the land area. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX G Page G-10 Prepared By: The Oil Sands Vegetation Reclamation Committee

G2.4.3 Climate

Climatic information for this zone is limited; four meteorological stations represent the area (Table G.4). These stations have an average June-August temperature of 12.8°C and average total precipitation for the same period of 267.5 mm. The frost free period is less than 85 days and May frosts are frequent.

G2.4.4 Ecology

Black spruce and jack pine forest types, with occasional jack-lodgepole pine hybrids at higher elevations make up a major portion of the vegetation in this zone. It is common for black spruce in this zone to occupy upland as well as wet sites. Peatland complexes composed of nutrient- poor black spruce bogs and nutrient-rich fens are common. Understory species in these forest types are commonly composed of Labrador tea and sphagnum mosses. For fens, the most common species are tamarack, dwarf birch, sedges and brown mosses. Areas of permafrost are reported for the zone.

Pure jack pine forest types on dry, sandy sites contain typical understory species such as bearberry, low bilberry, bog cranberry, prickly rose and reindeer lichen. Trembling aspen, balsam poplar and white spruce mixedwood forest types are not common in this zone and tend to be more frequent at lower elevations. Understories in these stands are fairly typical for mixedwood types but have fewer species due to the cooler climate. At higher elevations and in moister areas, these mixedwood types may be composed entirely of balsam poplar and white spruce. Paper birch is a minor component in many of these mixedwood types and occasionally forms pure stands.

G2.4.5 Summary

Seed zone 11a is part of a noncontiguous seed zone 11. It is cooler and moister during the growing season than adjacent areas at lower elevations and is dominated by coniferous black spruce and jack pine forest types. Mixedwood forest types which are more common at lower elevations and in areas of better drainage are more productive for timber than other types. The most similar seed zones are zone 11b, 11c, 11d and 11e. Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX G Page G-11 Prepared By: The Oil Sands Vegetation Reclamation Committee

Table G.4 Summary of Climatic Data for Draft Seed Zone 11A Stations

Climate Mean Mean Mean Mean Frost Growth Days Station June-August Summer Annual Annual Free Degree Length And Temp Precip Temp Precip Period Days (Longes Elevation J-J-A >10°C t) (°C) (°C) (mm) (Days) (mm) (Days) (Hours)

Birch 12.9 268.2 - - 88 - 18 Mountain LO (853 m)

Buckton 12.3 267.7 - - 69 - 18 LO (792 m)

Jean Lake 13.2 287.1 - - 83 - 18 LO (762 m)

Legend 12.7 247.1 - - 78 - 18 LO (911 m)

Mean 12.8 267.5 - - 80.0 - 18 Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX G Page G-12 Prepared By: The Oil Sands Vegetation Reclamation Committee

G3. LITERATURE CITED

Alberta Environmental Protection. 1994. Natural Regions and Subregions of Alberta: Summary. Publication No. 1/531. Edmonton, Alberta. 18 pp. APPENDIX H

NATIVE PLANTS SUITABLE FOR RECLAMATION IN THE CENTRAL MIXEDWOOD SUBREGION OF THE BOREAL FOREST NATURAL REGION Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX H Page H-1 Prepared By: The Oil Sands Vegetation Reclamation Committee

Table H.1 Native Plants Suitable for Reclamation in the Central Mixedwood Subregion of the Boreal Forest Natural Regiona

Habitat Preference Scientific Name Common Name Soil Texture Soil Moisture Soil Tolerance Grasses and Grass-like Plants Agrostis scabra tickle grass fine to coarse mesic to dry drought, acidic Beckmannia syzigachne slough grass Medium Wet alkaline, flood Bromus ciliatus fringed brome -- Mesic -- Calamagrostis canadensis bluejoint fine to coarse wet to mesic acidic, flood, drought, saline Calamagrostis inexpansa northern reed grass -- wet to mesic flood Calamagrtostis montanensis plains reed grass -- mesic to dry -- Calamagrostis stricta narrow-leaved reed -- wet to mesic -- grass Carex aquatilis water sedge Fine Wet alkaline, flood Carex atherodes awned sedge fine to medium Wet drought, alkaline Carex aurea golden sedge -- wet to mesic -- Carex capillaris hair-like sedge -- wet -- Carex obtusata blunt sedge -- mesic to dry -- Carex raymondii -- -- wet to mesic -- Carex rostrata beaked sedge -- wet to mesic -- Carex siccata hay sedge coarse dry -- Deschampsia cespitosa tufted hair grass fine to medium wet to mesic saline, alkaline, acidic Elymus canadensis Canadian wild rye medium to coarse mesic to dry alkaline Elymus innovatus hairy wild rye fine to coarse mesic to dry -- Festuca saximontana rocky mountain -- dry -- fescue Glyceria borealis northern manna fine wet flood grass Glyceria grandis tall manna grass fine wet flood Glyceria striata fowl manna grass fine wet flood Koeleria macrantha/cristata June grass medium to coarse mesic to dry drought, alkaline Oryzopsis asperifolia white-grained ------mountain rice grass Oryzopsis pungens northern rice grass fine to coarse dry -- Phragmites common reed grass fine to coarse wet to mesic alkaline australis/communis Poa arida plains bluegrass -- mesic to dry saline Poa glauca bluegrass coarse -- -- Poa palustris fowl bluegrass -- wet to mesic -- Schizachne purpurascens purple oat grass ------Trisetum spicatum spike trisetum coarse mesic to dry drought, acidic, alkaline Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX H Page H-2 Prepared By: The Oil Sands Vegetation Reclamation Committee

Table H.1 (cont’d)

Habitat Preference Scientific Name Common Name Soil Texture Soil Moisture Soil Tolerance Forbs Achillea millefolium common yarrow medium to coarse wet to dry drought, acidic Anemone canadensis Canada anemone -- wet -- Anemone multifida cut-leaved anemone coarse mesic -- Anemone parviflora small wood -- mesic -- anemone Arabis lyrata lyre-leaved rock fine dry -- cress Aralia nudicaulis wild sarsaparilla -- wet to mesic -- Artemisia campestris northern wormwood -- dry -- Aster borealis rush aster medium wet -- Aster ciliolatus Lindley’s aster ------Aster conspicuus showy aster ------Astragalus alpinus alpine milk vetch -- mesic -- Astragalus dasyglottis purple milk vetch -- mesic -- Astragalus striatus ascending purple -- dry -- milk vetch Comandra umbellata bastard toad-flax -- mesic to dry -- Cornus canadensis bunchberry -- mesic acidic Epilobium angustifolium fireweed fine to coarse mesic alkaline, acidic Epilobium ciliatum northern willow herb fine to coarse wet to mesic alkaline, acidic Erigeron canadensis horseweed -- dry -- Erigeron glabellus smooth fleabane -- mesic -- Fragaria vesca woodland strawberry coarse mesic -- Fragaria virginiana/glauca wild strawberry -- wet to mesic -- Galium boreale northern bedstraw -- wet to mesic -- Hedysarum boreale northern sweet vetch fine to coarse mesic to dry drought, saline, alkaline Heracleum lanatum cow parsnip -- mesic -- Heuchera richardsonii alum-root -- wet to mesic -- Hieracium umbellatum narrow-leaved -- dry -- hawkweed Lathyrus ochroleucus cream-colored pea medium wet to mesic -- vine Linnaea borealis twin-flower -- mesic -- Maianthemum canadense lily-of-the-valley -- mesic -- Melampyrum lineare cow wheat fine dry -- Mentha arvensis wild mint -- wet flood Mertensia paniculata tall lungwort -- mesic -- Mitella nuda Bishop’s-cap -- wet to mesic -- Oxytropis splendens showy locoweed ------Parnassia palustris northern grass of -- wet -- parnassus Petasites sagittatus arrow-leaved -- wet flood coltsfoot Potentilla pensylvanica prairie cinquefoil -- mesic -- Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX H Page H-3 Prepared By: The Oil Sands Vegetation Reclamation Committee

Table H.1 (cont’d)

Habitat Preference Scientific Name Common Name Soil Texture Soil Moisture Soil Tolerance Forbs - Continued Potentilla tridentata three-toothed fine dry -- cinquefoil Rubus arcticus dwarf raspberry -- wet to mesic acidic Rubus chamaemorus cloudberry -- wet to mesic acidic Rumex occidentalis western dock -- wet to mesic -- Scutellaria galericulata marsh skullcap -- wet -- Senecio pauperculus balsam groundsel -- wet to mesic saline Senecio triangularis brook ragwort -- wet to mesic -- Smilacina stellata star-flowered -- wet to mesic -- Solomon’s-seal Solidago canadensis Canada goldenrod fine to coarse mesic to dry drought Solidago spathulata mountain goldenrod -- mesic -- Stachys palustris hedge nettle -- wet -- Thalictrum venulosum veiny meadow rue -- wet to mesic -- Vicia sparsifolia/americana American vetch medium to coarse mesic -- Shrubs Alnus crispa green alder medium to coarse wet to mesic acidic, flood, alkaline Amelanchier alnifolia saskatoon medium to coarse mesic to dry drought, acidic, alkaline Arctostaphylos uva-ursi bearberry medium to coarse mesic to dry drought, acidic Cornus stolonifera red osier dogwood fine to medium wet to mesic alkaline, flood, acidic Ledum palustre/groenlandicum Labrador tea -- wet acidic Lonicera dioica twining honeysuckle ------Lonicera involucrata bracted honeysuckle medium to coarse wet to mesic -- Prunus pensylvanica pin cherry -- mesic -- Ribes hudsonianum Hudson Bay currant -- wet to mesic -- Ribes lacustre bristly black current medium wet to mesic -- Rosa acicularis prickly rose fine to coarse mesic acidic, drought, flood Rubus idaeus raspberry fine to coarse mesic drought, saline, acidic Salix candida hoary willow -- -- saline Salix glauca smooth willow fine to coarse wet to mesic acidic Salix maccalliana velvet-fruited willow -- wet to mesic -- Shepherdia canadensis Canadian buffalo- medium to coarse mesic to dry drought, saline, berry acidic, alkaline Vaccinium myrtillus low bilberry -- mesic to dry -- Vaccinium vitis-idaea cow-berry -- mesic to dry acidic Viburnum edule low-bush cranberry medium to coarse wet to mesic -- Guidelines for Reclamation to Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX H Page H-4 Prepared By: The Oil Sands Vegetation Reclamation Committee

Table H.1 (cont’d)

Habitat Preference Scientific Name Common Name Soil Texture Soil Moisture Soil Tolerance Trees Larix laricina tamarack medium wet to mesic acidic Picea glauca white spruce fine to coarse mesic acidic, drought, flood Picea mariana black spruce Medium wet to mesic acidic Pinus banksiana jack pine Coarse mesic to dry acidic, drought, alkaline Populus balsamifera balsam poplar fine to coarse wet to mesic flood, saline, alkaline Populus tremuloides trembling aspen fine to coarse mesic flood, drought, alkaline a Adapted from the original table in: Gerling, H.S., M.G Willoghby, A. Schoepf, K.E. Tannas and C.A. Tannas. 1996. A Guide to Using Native Plants on Disturbed Lands. Alberta Agriculture, Food and Rural Development and Alberta Environmental Protection. 247 pp. APPENDIX I

STANDARDS AND GUIDELINES FOR OPERATING BESIDE WATERCOURSES FROM: TIMBER HARVEST PLANNING AND OPERATING GROUND RULES DRAFT: Guidelines for Reclamation of Terrestrial Vegetation October 1998 In the Alberta Oil Sands Region – APPENDIX I Page 1 Prepared By: The Oil Sands Vegetation Reclamation Committee

Standards and Guidelines for Operating Beside Watercourses Prepared by Garry Ehrentraut, Northern Forest Products Ltd. Watercourse Physical Description Portion of Year Channel Fish and Wildlife Land Use Impact Roads, Landings and Watercourse Classification Water Flows Development Concerns Bared Areas Protective Buffers

Large * Major streams or rivers * All year * Unvegetated channel * Resident fish populations. * Water quality often reflects * Not permitted with 60m of the * No disturbance or Permanent * Well-defined flood plains width greater than * Important over-wintering all upstream land use impacts high-water mark or from water removal of merchantable * Valley usually exceeds 5m. habitat. and natural processes. Source areas within that buffer timber within 60m of the 400m in width * Important feeding & rearing * Primarily sedimentation of * May be permitted within 60 to high water mark except habitat. stream channels . 100m of the high water mark where specifically with written approval of a approved in the Annual Forest Officer. Operating Plan. Small * Permanent streams * All year, may * Banks and channel * Significant insect * Primarily sedimentation of * Not permitted within 30m of the * No disturbance or Permanent * Often small valleys freeze in the well-defined. population stream channels. high-water mark or from water removal of merchantable * Bench (floodplain) winter * Channel width 0.5m * Important spawning and * Fish populations sensitive to Source areas within that buffer. timber within 30m of the development to 5m. rearing habitat. situation. * May be permitted within 30 to high water mark except * Resident fish populations. * Loss of streambank fish 100 m of the high water mark where specifically * Overwintering for non- habitat. with written approval of a approved in the Annual migratory species. Forest Officer. Operating Plan.

Intermittent * Small stream channels * During wet * Distinct channel * Food production areas. * Sedimentation from bank and * Not permitted within 30m * Buffer of brush and lesser * Small springs are main season or development. * Potential spawning for streambed damage will of the high-water mark or from vegetation to be left source outside periods storms. * Usually channel is spring spawning species. damage fish habitat down water source areas within undisturbed along the of spring runoff and * Dries up unvegetated. * Drift invertebrate stream. that buffer. channel. heavy rainfalls during drought. * Channel width to populations in pools and * Width of buffer will vary 0.5m riffles. according to soils, * Some bank topography water source development. areas and fisheries values. * Treed buffer is not required unless specifically requested by a Forest Officer. Ephemeral * Often a vegetated draw * Flows during * Little or no channel * Situation may impact fish * Sedimentation downstream * Construction not permitted * Buffer of lesser vegetation or immediately development. habitat. due to ground disturbance. within a watercourse or a in wet gullies to be left after rain or * Channel is usually water source area. undisturbed. snowfall vegetated.

\\EDMOXP3A\ESESDdata\Website Development Project\Topic Areas\Land Reclamation\standard\Append-i.doc DRAFT: Guidelines for Reclamation of Terrestrial Vegetation October 1998 In the Alberta Oil Sands Region – APPENDIX I Page 2 Prepared By: The Oil Sands Vegetation Reclamation Committee

Standards and Guidelines for Operating Beside Watercourses - Concluded

Watercourse Physical Description Portion of Year Channel Fish and Wildlife Land Use Impact Roads, Landings and Watercourse Classification Water Flows Development Concerns Bared Areas Protective Buffers

Water source * Areas with saturated * All year * N/A * Potential high value to fall * Disturbance may cause * Construction not permitted * Treed buffer of at least 20m Areas (except soils or surface flow * May or may spawners. stream sedimentation. unless approved in the Annual on all streams. muskegs) not freeze * Potential high use areas * Interruption of winter flow Operating Plan. * No harvest of merchantable in winter. for terrestrial wildlife. may disrupt fish egg * No log decks permitted trees or disturbance of incubation. * The number of stream crossings lesser must be minimized. vegetation unless approved * No disturbance of organic duff by the Annual Operating layers or removal of lesser Plan. vegetation. * Buffer width may be altered according to its potential to produce surface water, provided it is approved in the Annual Operating Plan.

Lakes (little or * Large water collection * Normally * N/A * Important fish bearing * Aesthetic values may be * Not permitted within 100m * On lakes exceeding 16 ha in no recreation, areas permanently filled frozen in the habitat. disrupted. of the high-water mark without area, there will be no waterfowl, or with water winter. * Interruption of winter flow written approval of a Forest disturbance of timber within sport fishing may disrupt fish egg Officer. 100m of the high potential. incubation. water mark except where specifically approved in the Annual Operating Plan.

Lakes (with * Large water collection * Normally * N/A * Important fish bearing * Aesthetic values may be * For shorelines not located * On lakes exceeding 4 ha in recreational, areas permanently filled frozen in the habitat. disrupted. within reserved areas, no area, there will be no waterfowl or with water winter. * Interruption of winter flow disturbance will be permitted disturbance or removal of sport fishing may disrupt fish egg within 200m of the high water timber within 100m potential) incubation. mark without the written of the high water mark approval of the Forest except where specifically Superintendent. approved in the Annual Operating Plan.

\\EDMOXP3A\ESESDdata\Website Development Project\Topic Areas\Land Reclamation\standard\Append-i.doc APPENDIX J

WILDLIFE POPULATIONS AND HABITAT CAPABILITY IN THE OIL SANDS REGION Guidelines for Reclamation of Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX J Page J-i Prepared By: The Oils Sands Vegetation Reclamation Committee

TABLE OF CONTENTS

PAGE

J1. SUMMARY OF WILDLIFE POPULATIONS IN THE OIL SANDS REGION ...... 1 J1.1 UNGULATES ...... 1 J1.2 LARGE CARNIVORES...... 1 J1.3 SMALL TERRESTRIAL FURBEARERS ...... 2 J1.4 AQUATIC FURBEARERS ...... 2 J1.5 OTHER MAMMALS...... 3 J1.6 WATERBIRDS ...... 3 J1.7 RAPTORS ...... 4 J1.8 UPLAND GAME BIRDS ...... 4 J1.9 OTHER BIRDS...... 5 J1.10 REPTILES AND AMPHIBIANS...... 5 J1.11 SPECIES AT RISK ...... 5 J2. SELECTION OF TAR GET WILDLIFE SPECIES...... 7

J3. GENERAL HABITAT REQUIREMENTS FOR TARGET WILDLIFE SPECIES...... 10 J3.1 MOOSE ...... 10 J3.1.1 Food Requirements...... 10 J3.1.2 Cover Requirements ...... 10 J3.1.3 Special Habitat Requirements...... 10 J3.1.4 Landscape Components ...... 11 J3.2 BLACK BEAR ...... 11 J3.2.1 Food Requirements...... 11 J3.2.2 Cover Requirements ...... 12 J3.2.3 Special Habitat Requirements...... 12 J3.2.4 Landscape Components ...... 12 J3.3 SNOWSHOE HARE...... 12 J3.3.1 Food Requirements...... 12 J3.3.2 Cover Requirements ...... 13 J3.3.3 Landscape Components ...... 13 J3.4 RED SQUIRREL ...... 13 J3.4.1 Food Requirements...... 13 J3.4.2 Cover Requirements ...... 14 J3.4.3 Special Habitat Requirements...... 14 Guidelines for Reclamation of Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX J Page J-ii Prepared By: The Oils Sands Vegetation Reclamation Committee

TABLE OF CONTENTS (Continued)

PAGE

J3.5 RUFFED GROUSE ...... 14 J3.5.1 Food Requirements...... 14 J3.5.2 Cover Requirements ...... 14 J3.5.3 Special Habitat Requirements...... 15 J3.6 FISHER...... 15 J3.6.1 Food Requirements...... 15 J3.6.2 Cover Requirements ...... 15 J3.6.3 Special Habitat Requirements...... 15 J3.6.4 Landscape Components ...... 16 J3.7 GREAT GREY OWL...... 17 J3.7.1 Food Requirements...... 17 J3.7.2 Cover Requirements ...... 17 J3.7.3 Special Habitat Requirements...... 17 J3.7.4 Landscape Components ...... 17 J3.8 MICROTINES: RED-BACKED VOLE ...... 18 J3.8.1 Food Requirements...... 18 J3.8.2 Cover Requirements ...... 18 J3.8.3 Special Habitat Requirements...... 18 J3.8.4 Landscape Components ...... 18 J3.9 MICROTINES: DEER MOUSE...... 19 J3.9.1 Food Requirements...... 19 J3.9.2 Cover Requirements ...... 19 J3.9.3 Special Habitat Requirements...... 19 J3.10 PASSERINES...... 20 J3.10.1 Food Requirements...... 20 J3.10.2 Cover Requirements ...... 20 J3.10.3 Landscape Component...... 20 J4. ECOSITES AND LAN DSCAPE PATTERNS THAT WILL PROVIDE HABITAT FOR TARGET WILDLIFE SPECIES...... 21 J4.1 MOOSE ...... 21 J4.1.1 Food and Cover Requirements ...... 21 J4.1.2 Landscape Component...... 22 J4.2 BLACK BEAR ...... 30 J4.2.1 Food and Cover Requirements ...... 30 J4.2.2 Landscape Component...... 30 Guidelines for Reclamation of Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX J Page J-iii Prepared By: The Oils Sands Vegetation Reclamation Committee

TABLE OF CONTENTS (Continued)

PAGE

J4.3 SNOWSHOE HARE...... 31 J4.3.1 Food and Cover Requirements ...... 31 J4.3.2 Landscape Component...... 31 J4.4 RED SQUIRREL ...... 31 J4.4.1 Food and Cover Requirements ...... 31 J4.4.2 Landscape Requirements ...... 32 J4.5 RUFFED GROUSE ...... 32 J4.5.1 Food and Cover Requirements ...... 32 J4.5.2 Landscape and Special Habitat Requirements...... 32 J4.6 FISHER...... 33 J4.6.1 Food and Cover Requirements ...... 33 J4.7 GREAT GREY OWL...... 33 J4.7.1 Food and Cover Requirements ...... 33 J4.7.2 Landscape Component...... 34 J4.8 CAPE MAY WARBLER ...... 34 J4.8.1 Food and Cover Requirements ...... 34 J4.8.2 Landscape Component...... 34 J4.9 OVENBIRD ...... 35 J4.9.1 Food and Cover Requirements ...... 35 J4.9.2 Landscape Component...... 35 J4.10 WARBLING VIREO...... 35 J4.10.1 Food and Cover Requirement...... 35 J4.10.2 Landscape Component...... 35 J4.11 RED-BACKED VOLE...... 36 J4.11.1 Food and Cover Requirements ...... 36 J4.12 DEER MOUSE ...... 36 J4.12.1 Food and Cover Requirements ...... 36 J5. LITERATURE CITED ...... 37 Guidelines for Reclamation of Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX J Page J-1 Prepared By: The Oils Sands Vegetation Reclamation Committee

J1. SUMMARY OF WILDLIFE POPULATIONS IN THE OIL SANDS REGION

Prepared by Judy Smith and Greg Wagner, BOVAR Environmental

The following is an account of wildlife species known to occur in the oil sands region. Discussion is focused on wildlife species of economic, recreational and political importance including ungulates, carnivores, furbearers, waterbirds, raptors, upland game birds and species ‘at risk’.

J1.1 UNGULATES

Moose are the most common ungulate in the oil sands region. Population densities vary within the region based on habitat availability, hunting pressures and predation. In general, moose densities in northeastern Alberta are low compared to densities reported from central and northwestern Alberta. In at least part of the region, moose are known to undertake seasonal movements from upland areas to wintering areas in the Fort Hills or along the Athabasca and other major river valleys.

The oil sands region is near the northern limits of the range of both white-tailed and mule deer. Mule deer have historically been sporadically distributed across the boreal forest in northern Alberta. White-tailed deer, however, are more recent arrivals. They were first noted in the Fort McMurray area during the 1950s following the clearing of land for agriculture. Both species occur in low densities in the region. Populations are probably limited by severe winter weather conditions, including the presence of deep, crusted snow.

Elk were formerly more widely distributed in the oil sands area, but now are largely confined to areas along the Athabasca River south of Fort McMurray. Small numbers of woodland caribou are occasionally sighted in the region. Major populations, however, are largely restricted to the Birch Hills and Algar Lake area. Historically, herds of barren ground caribou made seasonal migrations as far south as Fort McMurray along the eastern boundary of the oil sands region. However, herds have not been observed in the area in over thirty years. Small herds of wood bison are occasionally observed and hunted in the oil sands region immediately south of Wood Buffalo National Park. But, major herds are mainly restricted to the park and areas north of the park.

J1.2 LARGE CARNIVORES

Coyotes are the most abundant large carnivore in the region. Timber wolves also occur. Radio-telemetry studies conducted in the late 1970s established densities at 1 animal/151 km2, which is within the range of densities reported for wolves from other localities in North America. Activities were concentrated in open and disturbed habitats, although elsewhere wolves are more commonly associated with forested habitats. Densities of black bear in the region have been estimated at 2 to 5.6 km2/bear, which are relatively high estimates compared to those reported from other areas of the boreal forest. Deciduous and mixedwood forests provide the highest quality habitats for black bear in the region. The historic range of the grizzly bear includes the oil sands region. Populations are greatly diminished from the past, but there have Guidelines for Reclamation of Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX J Page J-2 Prepared By: The Oils Sands Vegetation Reclamation Committee been some recent signs that the species still occurs in the region. Trapping records indicate that historically lynx were much more abundant in the region than they are currently. However, lynx populations follow the approximately ten year population cycle of the snowshoe hare and numbers are expected to increase as hare populations rebound.

J1.3 SMALL TERRESTRIAL FURBEARERS

The snowshoe hare is the most common terrestrial furbearer in the region, but undergoes dramatic population cycles so numbers can vary from year to year. The snowshoe hare is a staple prey species of the coyote, lynx and fisher, and affects population cycles in at least the first two species. The hare is also the most commonly trapped and hunted species in the region. Red squirrels are also abundant, occurring in forested habitats with a major pine or spruce component. Ermine are considered to be common to abundant and least weasels uncommon in the oil sands region. Both species are important to the trapping industry. These species show a preference for open tamarack-bog birch habitats, followed by black spruce- tamarack and cleared peatland habitats. Recently, marten populations have increased in northeastern Alberta. The species occurs mainly in continuous tracts of closed canopy, mature coniferous or mixedwood forests and, to a lesser extent, in open black spruce bogs and hardwood stands. The fisher is also considered to be an uncommon species that is trapped with some regularity in the region. Red fox also occur in the region, but in relatively low densities. The low numbers are attributable to competition with wolves and coyotes. Wolverine, which have large home ranges (200 to 500 km2), also occur in low densities.

J1.4 AQUATIC FURBEARERS

Beaver comprise a significant proportion of the total number of furbearers trapped in the region. They are widely distributed, occupying virtually all of the low gradient streams and standing bodies of water in the area. Muskrats occur in relatively low densities and are largely restricted to a few lakes and wetlands within the region. Mink densities are relatively high and fur harvest data indicates that mink are consistently trapped in low numbers within the Fort McMurray area. The species shows a preference for riparian shrub habitats. River otter occur in low numbers throughout the region, with the exception of the higher densities observed around the Calumet leases. Guidelines for Reclamation of Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX J Page J-3 Prepared By: The Oils Sands Vegetation Reclamation Committee

J1.5 OTHER MAMMALS

Regionally, seven species of mice and vole have been recorded. The white-footed deer mouse, red-backed vole and meadow vole are the most abundant species, while meadow jumping mouse, northern bog lemming, yellow-cheeked vole and heather vole occur in much lower numbers and are frequently restricted to specialized habitat types. Five species of shrew - masked, pygmy, dusky, water and arctic - are also expected to occur in the region. Of these, the masked and pygmy shrew are the most common. Several other species of mammals are known or expected to occur in the oil sands area. These include five species of bat (little brown bat, big brown bat, hoary bat, Keen’s bat, and silver-haired bat), badger, raccoon, woodchuck, porcupine, northern flying squirrel and chipmunk. Although some are relatively abundant in the region, none are of particular economic significance.

J1.6 WATERBIRDS

The Peace-Athabasca Delta, a major staging, breeding and moulting area for waterbirds, lies to the north of the oil sands region. Substantial numbers of migrating waterbirds therefore pass through the region in the spring and fall on their way to and from the delta. There is, however, relatively small amount of wetland habitat in the region and waterbird use is restricted to a few key wetlands. Three lakes have been identified as regionally important: McClelland and Ronald lakes for staging waterfowl, and Namur Lake as a breeding location for California gulls. Seven areas or lakes have been identified as locally important: Kearl, Audit and Algar lakes for staging ducks; Algar Lake for moulting ducks; the Birch Mountains uplands for migrating waterbirds and breeding trumpeter swans; Namur Lake for breeding white pelicans and herring gulls; and Gardiner and Eaglenest lakes for non-breeding concentrations of pelicans.

Most of the region is not heavily utilized by waterfowl during spring migration, except for a few lakes and some areas on the Athabasca River. Areas supporting the highest spring densities of waterbirds include McClelland and Little McClelland lakes. The region is used to a greater extent by fall migrants. Kearl, McClelland, Little McClelland and Gordon lakes are used by large numbers of diving and dabbling ducks, American coots and tundra swans. Numbers of staging fall migrant, however, vary substantially between both years and areas. The main migrant waterbird groups in the region are diving ducks, dabbling ducks, American coot and gulls. Little information is available on the migration or staging of other waterbird species in the region. Low numbers of migrant shorebirds, grebes, loons, swans and geese have been recorded over the last twenty years.

Included among the species migrating through the area is the endangered Whooping Crane, which nests in and near Wood Buffalo National Park and winters along the coast of Texas. During the mid-1970s, whooping cranes (lone birds or small flocks) were observed flying over Lease 17 and in 1974 two birds were flushed from Ruth Lake.

There is limited waterfowl production in the oil sands region, with the exception of three lakes, McClelland, Saline and Horseshoe, where mean densities of breeding pairs and broods are comparable to the densities associated with the prairie potholes. Most other wetlands in the Guidelines for Reclamation of Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX J Page J-4 Prepared By: The Oils Sands Vegetation Reclamation Committee region are characterized by low fertility and sparse vegetation, which create poor availability of food, nesting cover and brood cover.

Ten shorebird species are known to breed in the region, with greater and lesser yellowlegs, killdeer and spotted sandpiper being the most common.

J1.7 RAPTORS

Twenty-four species of raptors are known to occur as migrants, seasonal residents or permanent residents in the Fort McMurray region. Specific surveys have been conducted in the region to locate large and rare raptors (i.e., bald and golden eagle, osprey, peregrine falcon). Bald eagle nests have been found around several lakes in the region, with the greatest densities of nesting eagles occurring in the Gardiner Lake area. Ospreys are considered rare in the region, although a few nest sites are located around Namur Lake. Peregrine falcons and golden eagles only occur as migrants.

Common breeding raptors include the red-tailed hawk, broad-winged hawk, northern harrier, American kestrel, great horned owl, northern hawk owl and great grey owl. The red-tailed hawk prefers an interspersion of mature forest and grassland; it uses the muskeg and marshes for hunting, and the forest edge for resting. The broad-winged hawk is rarely seen in the open and prefers deciduous habitat or more open mixedwood areas with numerous low perches; it avoids disturbed areas and human inhabitation. Northern harriers hunt and nest along marshy areas and cutlines. American kestrels nest in tree cavities and prefer open areas with scattered perches. Great horned owls and northern hawk owls utilize forests and forest edge. Great grey owls nest in forested areas adjacent to bogs and fens, which are inhabited by its favourite prey item - the meadow vole. Other species that are known or thought to breed in the region include: sharp-shinned hawk, goshawk, merlin, short-eared owl, long-eared owl and boreal owl.

In addition to golden eagle and peregrine, migrants through the region include: Cooper’s hawk, Swainson’s hawk, rough-legged hawk, and gyrfalcon. Snowy owls are winter residents in the region.

J1.8 UPLAND GAME BIRDS

Four species of upland game birds occur in the region. Three of these species are year-round residents. The spruce grouse is the most abundant species, with coniferous dominated forests being the preferred habitat for much of the year. Deciduous or mixedwood forest with dense shrub understories are occupied by ruffed grouse, which are the second most abundant species in the region. Smaller numbers of sharp-tailed grouse occupy grassy and/or shrubby areas and are often found in recently cleared habitats.

The willow ptarmigan is a winter resident and reaches the southern limits of its wintering range in the tar sands region. They are most often found in habitats with a willow component or in trembling aspen and riparian communities. Guidelines for Reclamation of Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX J Page J-5 Prepared By: The Oils Sands Vegetation Reclamation Committee

J1.9 OTHER BIRDS

Other bird species occurring in the region are representative of four taxonomic groups (goatsuckers, kingfishers, woodpeckers and passerines). More than 80 species belonging to these groups have been observed in the oil sands region as migrant, breeding or overwintering species. Goatsuckers and kingfishers are represented by one species each, the common nighthawk and belted kingfisher, respectively. Seven species of woodpecker have also been encountered. Passerines are represented by more than 70 species. Over half of these are neotropical migrants.

Breeding bird surveys have been conducted in a variety of vegetation communities in the region. The numerically dominant species recorded during these surveys include: Tennessee warbler, least flycatcher, American redstart, ovenbird, Swainson’s thrush, chipping sparrow and palm warbler. The highest densities of breeding birds have been recorded in mature, mixedwood forest communities. In contrast, jack pine stands contained the least number of breeding birds.

J1.10 REPTILES AND AMPHIBIANS

Three species of amphibians, wood frog, striped chorus frog and the Canadian Toad, are known to occur in the oil sands region. A fourth species, western toad, likely occurs along the Athabasca River. The red-sided garter snake is the only reptile species known to occur in the region. It has been recorded at Kearl Lake and in the Birch Mountains. The absence of this species in other parts of the region may reflect a lack of landforms suitable for hibernacula.

J1.11 SPECIES AT RISK

Several species occurring in the region are identified as being ‘at risk’ either nationally or provincially. The Committee on the Status of Endangered Wildlife in Canada (COSEWIC) is responsible for identifying species or populations that are ‘at risk’ in Canada. Under the COSEWIC system ‘at risk’ species are designated under the following categories: endangered, threatened, and vulnerable. Alberta Environmental Protection (AEP) has also assigns status rankings to all wildlife species occurring in the province. Under the Alberta system ‘at risk’ species have been assigned to red and blue list categories. Red list species are considered at risk of extirpation within the province. The blue list is comprised of species that may be at risk of extirpation in the province. Species on the Yellow B list are considered to be naturally rare but not in decline; naturally rare and have clumped breeding distributions or associated with habitats that are or may be deteriorating. The following table presents national and provincial status designations for ‘at risk’ species occurring in the oil sands region. Guidelines for Reclamation of Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX J Page J-6 Prepared By: The Oils Sands Vegetation Reclamation Committee

Alberta Environmental Species COSEWIC Ranking Protection Ranking (Wildlife Management Division 1996) Wood Bison Threatened Red Grizzly Bear Vulnerable Blue Woodland Caribou Vulnerable Blue Wolverine Vulnerable Blue Peregrine Falcon Endangered Red Whooping Crane Endangered Red Caspian Tern Vulnerable Yellow B Bay-breasted Warbler Not Designated Blue Black-throated Green Warbler Not Designated Blue Cape May Warbler Not Designated Blue Short-eared Owl Vulnerable Blue Trumpeter Swan Not Designated Blue Canadian Toad Not Designated Red Guidelines for Reclamation of Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX J Page J-7 Prepared By: The Oils Sands Vegetation Reclamation Committee

J2. SELECTION OF TARGET WILDLIFE SPECIES

Prepared by Judy Smith and Greg Wagner, BOVAR Environmental

The justification for selecting the following species of wildlife as target species in order to define design criteria for reclaimed landscapes is presented below.

Moose. The moose is the most abundant ungulate species in the oil sands region. It is an economically important game animal for both subsistence and recreational hunting, and is a preferred species for wildlife viewing and photography. Moose are important in the food chain, providing food for predators, particularly the wolf, and scavengers in the region. They are largely browse-dependent, occupying a variety of forest types that provide edges or disturbed areas of early successional vegetation. The shrub and ground strata of deciduous, mixed and coniferous forests and shrublands are used for both food and cover. Moose are on the Green List in Alberta (Wildlife Management Division 1996).

Black Bear. The black bear is an abundant species in the region and is important for both recreational and subsistence hunting. Black bears are large, solitary omnivores that occupy large home ranges through the forested regions of North America. Habitat use by black bears is influenced primarily by the seasonal availability of food and by proximity to cover. It is dependent on a mosaic of habitat types, particularly mid-successional to mature deciduous and mixedwood forests with productive herbaceous and shrub understories. The black bear is on the Green List in Alberta (Wildlife Management Division 1996).

Snowshoe Hare. The snowshoe hare is common and widely distributed resident of the region, and provides an important food source for subsistence trappers and recreational hunters. This herbivore is an important prey species for carnivores in the area, including lynx, coyote, marten and fisher. Young forests with well-developed understories that provide abundant food and cover are the preferred habitat. It is on the Green List in Alberta (Wildlife Management Division 1996).

Red Squirrel. The red squirrel is widely distributed across the boreal forest of North America. In the northern part of its range, food availability is the most important factor determining habitat selection. Although red squirrels consume a variety of different items, spruce generally constitute the largest part of the diet, particularly in winter. In the oil sands area, the red squirrel is commonly found in mature forest communities dominated by coniferous species, particularly white spruce, and, to a lesser degree, black spruce. It is extensively trapped in the region and is an important prey species for a number of carnivores and raptors. The red squirrel is on the Green List in Alberta (Wildlife Management Division 1996).

Ruffed Grouse. The ruffed grouse is an important game bird for both recreational and subsistence hunters. Grouse is an important prey species in the area. The ruffed grouse is a non-migratory, widely distributed species, occurring in a broad band of deciduous and mixed deciduous-coniferous forest habitats across North America. Habitat use is influenced by the proximity and quality of cover and by the seasonal availability of food. In Alberta, Ruffed Grouse primarily occupy aspen forest utilizing the ground, shrub and tree canopy strata for obtaining Guidelines for Reclamation of Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX J Page J-8 Prepared By: The Oils Sands Vegetation Reclamation Committee food and cover. The ruffed grouse is a Green Listed species in Alberta (Wildlife Management Division 1996).

Fisher. The fisher is a large, arboreal member of the weasel family that was formerly widely distributed across forested areas in Canada and western and northeastern parts of the U.S. Population in the southern parts of its range have been greatly reduced because of overharvesting and habitat loss. In the oil sands area it typically occurs in mature coniferous dominated forests with a closed canopy. It is an opportunistic feeder and its diet includes small mammals, birds and carrion. When available, snowshoe hares make up a large part of the diet. It is also a specialized predator of the porcupine. Fisher are an economically important species to the trapping industry and are frequently harvested in the oil sands area. In Alberta, the fisher is designated as Yellow B List species (Wildlife Management Division 1996)

Great Gray Owl. Great Gray Owls are distributed across the boreal forests of North America and Eurasia where they occur in coniferous, deciduous, and mixed forests, typically along forest margins in muskegs, marshes, and wet meadows. The Great Gray Owl as previously considered vulnerable across its Canadian range, but was removed from the COSEWIC list in 1996. The species is on the Yellow List in Alberta (Wildlife Management Division 1996).

Microtines: Red-backed Vole. The distribution of the red-backed vole closely coincides with the boreal forests throughout North America. Although this species occupies a wide range of plant communities, it is most common in mature forests. In northern Alberta, they occupy a variety of habitats, utilizing both the ground and shrub strata for obtaining food and cover. They avoid fields, clearings, and other unforested habitats unless an abundance of protective ground litter and shrub cover is present. Red-backed voles are an important prey species for various carnivores and raptors. They are on the Green List in Alberta (Wildlife Management Division 1996).

Microtines: White-footed Deer Mouse. The white-footed deer mouse is one of the most widely distributed small rodent species in North America. It is a ubiquitous species that occupies a variety of habitat types ranging from grasslands to forested communities. In the oil sands region it most frequently occurs in balsam poplar and mixed wood forests and early successional habitats. The presence of dense shrub and ground cover appears to be a prerequisite for its occurrence in these areas. The deer mouse is an important prey species of hawks, owls, weasels and foxes as well as an important predator of insects detrimental to regenerating and mature forests. This species is on the Green List in Alberta (Wildlife Management Division 1996).

Passerines: Cape May Warbler. The Cape May Warbler is a tree-nesting wood warbler of the boreal forest. Cape May Warblers breed in mature white spruce stands within coniferous and mixedwood forests, preferring open stands and stand edges. This species is on the Blue List in Alberta (Wildlife Management Division 1996).

Passerines: Ovenbird. The ovenbird is a neotropical migrant that breeds in forested areas of Canada and the northeastern U.S. It is a fairly common summer resident in the oil sands region inhabiting mature deciduous or mixed wood forests with limited understory development. They Guidelines for Reclamation of Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX J Page J-9 Prepared By: The Oils Sands Vegetation Reclamation Committee feed and nest on the ground. Ovenbirds winter in tropical regions from the Gulf of Mexico to northern Columbia and Venezuela. The ovenbird is on the provincial Green List (Wildlife Management Division 1996).

Passerines: Warbling Vireo. The warbling vireo is a neotropical migrant that breeds in boreal, subalpine and deciduous forest communities throughout Canada and the U.S. It inhabits mature deciduous forests with open canopies and well developed understories. Nests are generally built in the upper canopy of large deciduous trees. The species is insectiverous and generally forages in the shrub stratum of forests or edge areas adjacent to forests. They often feed on hairy caterpillars, which are generally avoided by other species. It winters in central America from Mexico to Guatemala. The warbling vireo in on the Green List in Alberta (Wildlife Management Division 1996). Guidelines for Reclamation of Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX J Page J-10 Prepared By: The Oils Sands Vegetation Reclamation Committee

J3. GENERAL HABITAT REQUIREMENTS FOR TARGET WILDLIFE SPECIES

Prepared by Greg Wagner and Wayne Condon, BOVAR Environmental

In the majority of cases, habitat suitability is governed by broad biological and physical characteristics such as shrub or tree canopy cover, slope and topographic characteristics. The following section outlines the habitat requirements for the selected terrestrial target wildlife species. The information has been divided into: food requirements, cover requirements, special habitat requirements, and landscape components.

J3.1 MOOSE

J3.1.1 Food Requirements

· Food and cover are provided by the shrub and ground strata of deciduous, mixed and coniferous forests, as well as shrubland habitat types. · During the fall/winter season, deciduous browse is used almost exclusively, although actual plant species composition of the diet may vary between areas. · In the spring/summer season, herbaceous (i.e., forbs, graminoids) and aquatic vegetation, in addition to deciduous browse species, are frequently important dietary items. However, the degree to which these items are utilized varies widely from area to area. · Of all the browse species known to be utilized by moose, various willow species are consistently reported to be one of the most utilized shrubs throughout the North American range. Within the oil sands area, willow, saskatoon, red-oiser dogwood, low- bush cranberry, birch, alder, chokecherry, pincherry and, to a lesser degree, aspen and balsam poplar are important browse items.

J3.1.2 Cover Requirements

· For much of the year, habitats selected for their food producing abilities also provide moose with adequate cover, particularly against predators. · In winter, moose demonstrate preferences for areas supporting both high browse yield habitats and mature coniferous forests offering thermal cover and reduced snow depths. · In summer, lakes and ponds which may be used as aquatic foraging areas also offer open-water relief from insect harassment and high temperatures.

J3.1.3 Special Habitat Requirements

· Calving areas are predominately located in isolated muskegs, riparian areas, or marshy sites interspersed with islands and isolated forest patches. Often such areas are associated with lakes, streams or marshes.

· Home range size varies based upon age, sex and seasonal use. Seasonal home ranges seldom exceed 5-10 km2. Guidelines for Reclamation of Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX J Page J-11 Prepared By: The Oils Sands Vegetation Reclamation Committee

· Snow depths > 90 cm can severely restrict movement. However, within the oil sands area, mean snow depths rarely reach these levels and probably plays a minor role in habitat selection.

J3.1.4 Landscape Components

· In general, an interspersion of habitat types, rather than a homogenous vegetation community, is considered prime moose habitat, since such an interspersion ensures that adequate cover and forage are located nearby.

· Typically, moose prefer edge habitats spending most of their foraging time within 100 m of cover and are rarely seen more than 200 m from foraging areas.

· Optimal interspersion of foraging and cover areas is a ratio of 65:35.

· Moose may undergo seasonal migrations, although migratory tendencies may vary widely within and between geographical areas. In at least part of the oil sands region, moose make seasonal movements between summering areas in uplands and wintering areas along major river valleys. The existence of travel corridors between seasonal habitats is therefore an important factor determining the overall habitat suitability of a region. Although detailed studies are lacking, travel corridors at least 500 m in width will permit migrations between seasonal habitats.

· The presence of roads and trails in an area may cause a reduction in moose densities because of the occurrence of moose/vehicle collisions and increased hunting or predation pressures. Generally, hunting pressures are greatest within 1 km of a road or trail.

· Currently, the literature contains little or no information on the minimum area requirements of moose. However, even if such information existed minimum area requirements would probably vary widely from area to area and from year to year based on the availability and dispersion of food and cover over time.

J3.2 BLACK BEAR

J3.2.1 Food Requirements

· Black bears are omnivorous and consume a wide variety of foods; however, in the boreal forest a major portion of the diet consists of herbaceous plants and berries.

· During the early spring, following the denning period, black bears select open areas where they feed on newly emergent vegetation, particularly grasses, sedges and horsetails.

· With the progression of summer, dietary preference switches to berries, nuts, insects and a variety of herbs.

· Berries are the preferred late summer/fall food because of their high sugar content and digestibility. Guidelines for Reclamation of Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX J Page J-12 Prepared By: The Oils Sands Vegetation Reclamation Committee

· Habitat selection is closely tied to food availability, particularly the availability of berries.

J3.2.2 Cover Requirements

· For much of the year, habitats selected by bears for their food producing capabilities also provide adequate escape cover.

· Dense shrub or tree cover is used to escape from predators (e.g., other bears and wolves). Black bears also commonly bed in dense shrub communities.

· Black bears will also climb trees to escape predators. Trees with diameters at breast height of at least 15-20 cm are required to support the weight of a climbing bear.

· In the spring, black bears often feed in open areas with little or no cover. However, the use of open areas decreases at distances greater than 200 m from escape cover.

J3.2.3 Special Habitat Requirements

· Typically, black bears select dens located on steep slopes with north or east aspects in mature forests, however, they will also den in level areas of deciduous forests.

J3.2.4 Landscape Components

· Habitat interspersion increases habitat quality in an area. Areas including open, early successional areas (where newly emergent spring vegetation is available) adjacent to forested areas (providing escape cover and berry-producing plants) represent ideal habitat conditions.

· Currently, the literature contains little or no information on the minimum area requirements of black bear. However, even if such information existed minimum area requirements would probably vary widely from area to area and from year to year based on the availability and dispersion of food and cover over time.

J3.3 SNOWSHOE HARE

J3.3.1 Food Requirements

· The diet of the snowshoe hare varies seasonally. In late fall and winter, hares forage primarily on buds, twigs, bark, conifer needles and the evergreen leaves of woody plants.

· In the summer, the diet consists of a variety of leaves, herbs and green plant material. Guidelines for Reclamation of Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX J Page J-13 Prepared By: The Oils Sands Vegetation Reclamation Committee

· Several species of plants are known to be unpalatable, limiting habitat use in areas dominated by such species. Unpalatable species include black spruce, Labrador tea, low-bush cranberry, bracted honeysuckle and snowberry.

J3.3.2 Cover Requirements

· Snowshoe hare occur in pure coniferous, pure deciduous, or mixedwood forests.

· The most important aspects of habitat is a dense understory cover (50-60%) at a height of 1-3 metres to serve as escape cover and as a winter food supply.

· Deciduous stands dominated by aspen, balsam poplar and paper birch comprise the best habitat for snowshoe hares, although coniferous habitat provides better thermal protection during winter.

· Areas with 16 320+ stems/ha (shrubs and trees) provide adequate cover (and food).

· Open areas are considered poor habitat because of increased exposure to predation, decreased forage availability and reduced thermal cover.

· In the boreal forest, high quality early successional habitats regenerate following clearcutting or fire.

J3.3.3 Landscape Components

· Habitat quality increases with increased habitat dispersion. A patchy habitat mosaic, providing dense thickets for winter use and more open summer range, allows hares to shift range use seasonally to take advantage of changing environmental conditions.

J3.4 RED SQUIRREL

J3.4.1 Food Requirements

· Food is the single most important factor influencing the distribution of the red squirrel.

· Red squirrels are omnivores, eating a variety of seeds, nuts, berries, tree bark, fungi, insects, birds eggs, young birds and mice.

· The seeds of white and black spruce are the most important dietary item and influence reproduction and territory size. White spruce seed crops are variable from year to year, but provide a high quality food source. Black spruce seeds are of lower nutritive value, but crop production is relatively stable on an annual basis.

· During late summer and fall, red squirrels store cones in caches, often around the base of large spruce trees, under fallen logs, or in cavities of large snags. Guidelines for Reclamation of Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX J Page J-14 Prepared By: The Oils Sands Vegetation Reclamation Committee

J3.4.2 Cover Requirements

· Optimum cover for red squirrels occurs in mature, dense coniferous forests with greater than 50% canopy closure which provide winter food, nesting sites, thermal cover, moisture, and shade for food storage.

· Mature white spruce forests, followed by black spruce dominated forests, provide the most suitable habitat for red squirrels in the oil sands area.

J3.4.3 Special Habitat Requirements

· Red squirrel territorial requirements are highly variable, being heavily dependent on food availability. Reported densities range from 1 squirrel/0.5 ha to 1 squirrel/4.8 ha.

J3.5 RUFFED GROUSE

J3.5.1 Food Requirements

· Ruffed grouse are omnivores, feeding largely on buds, twigs, forbs, fruits, berries, and insects. Diet varies throughout the year based on the availability of foods.

· As berries, fruits and seeds become available in summer, preference is shown towards strawberries, raspberries, cherries, juneberries and sedges.

· The fall diet is comprised mainly of berries, including: cranberries, red-oiser dogwoods and roses.

· Buds, twigs and catkins of aspen and willow, supplemented with overwintering berries, are winter food staples. The use of these foods persists into early spring.

· Insects comprise only a small portion of adult food, however, the diet of chicks consists of about 90% invertebrates until about eight weeks of age.

J3.5.2 Cover Requirements

· Ruffed grouse occupy a variety of climax and successional forest community types in North America. However, the most important factor in habitat selection is the presence of a substantial deciduous tree component, particularly aspen and birch, in the tree canopy.

· In Alberta, ruffed grouse primarily occupy aspen forests with dense understories of berry-producing shrubs.

· Young aspen stands first become occupied by ruffed grouse about 4 to 12 years after regeneration following logging or fire, when trees are 8-10 m tall and stem densities are less than 14 800/ha. Grouse continued to use the habitat throughout the year for the next 10 - 15 years, until stem densities drop below 800/ha. Guidelines for Reclamation of Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX J Page J-15 Prepared By: The Oils Sands Vegetation Reclamation Committee

J3.5.3 Special Habitat Requirements

· Females typically nest in dense stands of older aspen with an open canopy and understory. Nests are usually located within 15 m of a forest opening.

· Females with broods show a preference for brushy habitat with dense escape cover nearby. Small clearings in deciduous forest, 1/10 to 1/2 hectare in size, are important brood rearing areas.

· Male ruffed grouse select territories where they display to females from raised structures on the forest floor, usually on trunks and fallen trees. Drumming sites are associated with areas of forests where woody stems are denser, the canopy cover is predominantly deciduous, the coniferous cover is very young white spruce and the shrub canopy cover is located well above the ground.

J3.6 FISHER

J3.6.1 Food Requirements

· Fisher will consume a wide variety of food, including small mammals, invertebrates, large mammal carrion, and a variety of birds.

· Snowshoe hare are a primary prey species which may, in conjunction with cover requirements, determine fisher abundance. Fishers will switch primary prey species in years of low hare availability/abundance.

· The fisher is also a specialized predator of the porcupine.

J3.6.2 Cover Requirements

· Fishers require mature to old forests with dense canopy closure (80-100%).

· Coniferous and mixedwood stands, with 50-80+% coniferous species composition, are the most suitable habitats for fishers.

· Pure deciduous stands and stands with more than 70% deciduous cover are avoided. Clearcuts and open forests are also avoided.

· Important stand level characteristics associated with old forests include: hollow logs, snag cavities, brush piles and snow dens which are used as resting areas.

J3.6.3 Special Habitat Requirements

· Maternity dens are almost always located in tree cavities, and minimum diameter at breast height for maternity den trees is 51 cm. Guidelines for Reclamation of Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX J Page J-16 Prepared By: The Oils Sands Vegetation Reclamation Committee

· Movements may be restricted by soft, deep (> 20 cm) snow.

J3.6.4 Landscape Components

· Annual home range size vary considerably. Values from 6.6 to 78.2 km2 have been reported.

· Male fishers occupy larger territories than females.

· Currently there are no estimates in the literature regarding minimum area requirements of fishers. Guidelines for Reclamation of Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX J Page J-17 Prepared By: The Oils Sands Vegetation Reclamation Committee

J3.7 GREAT GREY OWL

J3.7.1 Food Requirements

· Great grey owls primarily prey upon small mammals, particularly rodents.

· Meadow voles dominate the diet of the great grey owl over most of the owl’s range, particularly in the northern boreal forest. Nest site productivity and habitat selection are closely tied to the availability of meadow voles.

J3.7.2 Cover Requirements

· Great grey owl breeding habitat generally consists of extensive forest interspersed with sphagnum bogs, muskeg and other open areas.

· Nesting commonly occurs in stands of mature poplar adjacent to muskegs. Islands of poplars amid stands of spruce or pine are common breeding locations, as are groves or marginal strips of often-stunted tamaracks in wetter sites.

· Foraging habitat closely corresponds to areas occupied by meadow voles, which includes moist grass/sedge openings and open herbaceous forest habitats. The primary cover requirement of the meadow vole is the availability of dense, grassy vegetation. In the oil sands area, meadow voles are most abundant in areas with dense ground cover of either successional herbaceous plants or graminoids. Meadow voles also prefer wet to moist soil conditions (soil moisture exceeding 30%).

J3.7.3 Special Habitat Requirements

· Perches must be available in foraging areas as hunting from the ground or during flight is rare.

· Great grey owls do not build their own nests and thus must rely on abandoned hawk and raven stick nests or natural depressions on broken-topped snags or stumps for nest sites.

J3.7.4 Landscape Components

· Typically, great grey owls are reliant on interspersion of open, grassy areas for foraging, and mature forested areas for nesting. These two habitat conditions must therefore occur adjacent to one another to create suitable nesting habitat.

· In almost all cases nests are within 500 m of preferred foraging areas. Guidelines for Reclamation of Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX J Page J-18 Prepared By: The Oils Sands Vegetation Reclamation Committee

J3.8 MICROTINES: RED-BACKED VOLE

J3.8.1 Food Requirements

· Red-backed voles are omnivorous, feeding largely on forbs, shrubs, berries lichens, fungi and insects.

· Diet is seasonally variable.

· During the winter, overwintering fruits, small twigs, buds and lichens are staples of the diet.

· Overwintering berries are important dietary items in early spring; newly emerged berries, leaves of trees and shrubs, mosses, lichens, fungi, and horsetails are important foods from late spring to fall. In some areas, fungi can make up a large part of the summer diet.

J3.8.2 Cover Requirements

· Although this species occupies a wide range of plant communities, it is most common in moist, mature forest with relatively dense shrub canopies and abundant litter, moss and deadfall.

· Clearings and other unforested habitats are avoided unless an abundance of protective ground litter and shrub cover is present.

J3.8.3 Special Habitat Requirements

· Red-backed voles require a high daily intake of water. Consequently, the species is often restricted to low, wet areas or to areas where abundant succulent food is available.

· High litter abundance (i.e., leaves, needles, organic mulch) is commonly associated with vole habitat suitability.

J3.8.4 Landscape Components

· A minimum of 2 ha of suitable habitat may be required before an area will be occupied. Red-backed voles have limited mobility and poor dispersal abilities through open habitats to new areas, hence, the need for continuous habitat. Guidelines for Reclamation of Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX J Page J-19 Prepared By: The Oils Sands Vegetation Reclamation Committee

J3.9 MICROTINES: DEER MOUSE

J3.9.1 Food Requirements

· Food habits of the deer mouse can best be described as opportunistic and omnivorous. Regular seasonal shifts in diet are typical.

· Arthropods typically make up a large portion of the diet, particularly in the early spring.

· Berries, fruits or seeds of grasses, shrubs and trees are used as they become available and gradually comprise more of the diet in the late summer and fall.

· When available, fungi may also form a large part of the summer diet.

· In the oil sands area, willow leaves are an important dietary item during the late summer and fall.

· During the fall and winter, the diet largely consists of the seeds and fruits of grasses, shrubs and trees. Seeds of conifers may also be consumed in the winter in forest habitats.

· In the oil sands area, willow bark also appears to be an important winter food.

· Conifer needles may be eaten during periods of deep snow.

J3.9.2 Cover Requirements

· Deer mice are widely distributed and show few restrictions in habitat use.

· The species occurs primarily in woodlands and brushlands, but it also occurs in open areas such as grasslands and early successional habitats.

· Preferred habitats consist of forested areas with dense shrub and ground cover.

· In the oil sands area, deer mice are most abundant in forests with a dense understory dominated by currant, dogwood, alder or raspberry; a dense ground cover of horsetail and a variety of herb species; moderate to thick accumulations of litter and deadfall; an absence of grass/sedge cover; and dense vertical plant cover up to a height of 1.5 m.

J3.9.3 Special Habitat Requirements

· Habitat use may be influenced by interspecific and intraspecific competition. In particular, deer mice may be excluded from grassland areas through competition with meadow voles. Guidelines for Reclamation of Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX J Page J-20 Prepared By: The Oils Sands Vegetation Reclamation Committee

J3.10 PASSERINES

Optimal habitat for many passerines has a high structural diversity both vertically and horizontally in the landscape. There are many species restricted to a single habitat type. These species are considered habitat specialists.

J3.10.1 Food Requirements

· The Cape May warbler is primarily an insectivore, although it may feed on a variety of invertebrates and other materials. Numbers of Cape May warblers have been positively correlated with outbreaks of spruce budworm.

· Ovenbirds are insectivorous ground foragers and feed on a variety of invertebrate species.

· Warbling vireos are insectivores and eat a number of invertebrates including hairy caterpillars, which are avoided by many other species. Feeding generally occurs in shrubs.

J3.10.2 Cover Requirements

· Cape May warblers breed and nest in mature white spruce stands (> 60 years old) within coniferous and mixedwood forests, preferring open stands and stand edges.

· During the breeding season, male Cape May warblers select tall conifers that rise above the rest of the forest for singing perches.

· Ovenbirds are primarily associated with mature, closed canopy deciduous and mixedwood forests with little or no shrub cover. Open areas in forests are avoided.

· Warbling vireos are associated with old deciduous (60+ years) forests or aspen thickets within mixedwood forests. Forests with an open canopy and well developed shrub understory are preferred.

J3.10.3 Landscape Component

· Ovenbirds are a habitat size dependent species that are restricted to forest patches > 4 ha in area.

· Habitat quality for the warbling vireo improves in edge areas where dense shrub/sapling areas (foraging areas) are located adjacent to mature deciduous forests (nesting areas). Guidelines for Reclamation of Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX J Page J-21 Prepared By: The Oils Sands Vegetation Reclamation Committee

J4. ECOSITES AND LANDSCAPE PATTERNS THAT WILL PROVIDE HABITAT FOR TARGET WILDLIFE SPECIES

Prepared by Greg Wagner, BOVAR Environmental

The ecosites and landscape patterns on the reclaimed oil sand leases that would meet the habitat requirements of the target wildlife species are outlined below:

The value of each of these ecosites, with their associated ecosite phases and plant communities, for the target wildlife species are discussed below. Habitat suitability index models have been developed for the oil sands region for some of the targeted species. For these species, habitat suitability indices have been determined for various ecosite phases and vegetation communities. Under this format, an HSI value of 1.0 would be assigned to the habitat conditions that are considered optimal, and 0.0 would be assigned to habitat conditions that are not suitable for a particular species. In the following text, habitat suitability is discussed only in relation to ecosite phases. A number of vegetation communities can be represented under an ecosite phase, and habitat suitabilities can vary for each vegetation community within an ecosite phase. As such, an ecosite phase may be shown as having both high and moderate suitability for a species based on habitat suitability rankings for various vegetation communities that are representative of a particular ecosite phase. For other target species, habitat suitability index models have not been developed for the oil sands region. For these species, a qualitative ranking has been assigned to various vegetation communities based on the habitat requirement of each species. These qualitative rankings are also presented in relation to ecosite phases. The palatability of the shrub and herbaceous plant species found within ecosites that can be supported on reclaimed sites, based on Beckingham and Archibald (1996), is presented in Table J.1.

J4.1 MOOSE

J4.1.1 Food and Cover Requirements

Food and cover for moose are provided by deciduous, mixed wood and coniferous forests and shrubland habitat types. Diet is seasonally variable (Table J.1). In the spring/summer season, herbaceous and aquatic vegetation, and deciduous browse are important. During the fall and winter the diet is made up almost exclusively of tall deciduous browse species. Within the oil sands region, reclamation activities can create a variety of vegetation communities important to moose (Table J.2). These vegetation communities are representative of the following ecosite phases: · High Suitability Habitats - aspen/low-bush cranberry (d1) and aspen/white spruce/low bush cranberry (d2) ecosite phase. · Moderate Suitability Habitats - jack pine-aspen/blueberry (b1); aspen/low-bush cranberry (d1); aspen-white spruce/low-bush cranberry (d2); white spruce/low-bush cranberry (d3); white spruce/dogwood (e3); and, to a lesser extent, aspen-white spruce/blueberry (b3) and white spruce-jack pine/ blueberry (b4) ecosite phases. Guidelines for Reclamation of Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX J Page J-22 Prepared By: The Oils Sands Vegetation Reclamation Committee

J4.1.2 Landscape Component

Moose occupy relatively large seasonal home ranges (5-10 km2) and on an annual basis are dependent on a mosaic of habitat types. Habitat interspersion, therefore, plays an important role in determining overall habitat quality and the following factors should be incorporated in the development of re-constructed landscapes for moose: · Moose show a preference for edge habitats, including areas with high browse yields (shrub and successional habitats with tall, preferred browse species) adjacent to forests that provide escape and thermal cover and reduced snow depths during the winter. Habitat quality for moose can be improved by locating successional or shrub dominated communities (associated with shrubby rich fen ecosite phases) adjacent to forest communities with high cover values. To maximize habitat quality, foraging areas should be developed as 100 m strips adjacent to 200 m forest strips. · Muskeg, riparian and marshy areas are seasonally important as calving areas. Reclamation habitats should be developed in proximity to such areas. In particular, habitat quality in calving areas can be improved by developing isolated forest patches or forest islands within or adjacent to muskeg and marshy habitats. · During the open-water season, lakes and ponds are used as foraging areas and provide relief from insect harassment and high temperatures. Lake and pond habitats are currently somewhat limited in the oil sands area, however, such habitats are being developed to a greater extent in reclaimed areas. To improve habitat quality for moose, extensive littoral zones (water depths of 0-3 m) should be developed on these wetlands to promote the establishment of aquatic vegetation. Vegetation communities with high habitat suitability should also be established adjacent to existing and reconstructed wetlands. · Habitat quality for moose is reduced in areas within 1 km of roads. · Within parts of the oil sands area, moose make seasonal migrations between summer habitats in upland areas and wintering habitats along major river valleys or in the Fort Hills. Travel corridors at least 500 m in width should be maintained between these areas. Guidelines for Reclamation of Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX J Page J-23 Prepared By: The Oils Sands Vegetation Reclamation Committee Table J.1

Palatability of Plant Species for Key Wildlife Indicator Species Based Upon Defined Community Types

Plant Species Palatabiltity For Key Wildlife Indicator Species Common Name Latin Name Black Bear Moose General Palatabilityd Ruffed Grouse Snowshoe Hare Red-backed Vole

Food Foodf Percent Food Percent Weighte Common Palatability Tolerance Food Percent Volumeb Percent Volumec* Food Foodf Food Food Mean Usea Frequencyg Usea, l Foragesj Usea Usea Usea Usek Consumptionh

Shrub Layer balsam fir Abies balsamea 2-3 16.92 3-4 m green alder Alnus crispa 1-3, 2 * Medium Medium + 1 m river alder Alnus tenuifolia 2-3, 2 * Medium Medium 3-4 3-4 m Saskatoon Amelanchier alnifolia 1 t 1, 1 0.88 * Medium Medium 2 5% 2.5% 1 Bearberry Arctostaphylos 38 Medium Medium * uva-ursi white birch Betula papyrifera 3-4 2-3, 2 * Medium Medium 1-2 9.2% 2 m Dogwood Cornus stolonifera + 1-2, 1 25.34 * Medium-High High 1-2+ 1.6% beaked hazelnut Corylus cornuta 1 0.81 * 1-3 4% 1.4% 2 m * Labrador tea Ledum groenlandicum 1 0.04 * twin-flower Linnaea borealis * bracted honeysuckle Lonicera involucrata t * u white spruce Picea glauca * High Low 1-2 3 m 0.19 black spruce Picea mariana * t m balsam poplar Populus balsamifera 5 2 6.85 * Medium-High High 1% m aspen Populus tremuloides 3, 3 7.47 * Medium-High Medium-High 3-4 35% 2-3 m * pin cherry Prunus pensylvanica 1-4 1+ * 1-2 10.6% + choke cherry Prunus virginiana 1-4 1+ 0.13 * Medium Medium 1-2 + * 3.1* currant Ribes spp. 24.4 * prickly rose Rosa acicularis 18.5 0.03 * High Medium 1+ 5% * wild red rasberry Rubus idaeus m 28 - 0.04 * Low-Medium Medium 1-2+ 8.8% m * Willow Salix spp. 8.5 2-3, 1 22.64 * High High 1-2 31% 2-3 m Canada buffalo-berry Sheperdia canadensis 1 40 * Low Medium-High 2% * Snowberry Symphoricarpos albus 2 1 u Blueberry Vaccinium myrtilloides 3-4 m 43* 1-2+ 3 bog cranberry Vaccinium vitis-idaea + m 43* + 1 - - * low-bush cranberry Viburnum edule 2-3 * 1-2 2.0% 3 m & u Guidelines for Reclamation of Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX J Page J-24 Prepared By: The Oils Sands Vegetation Reclamation Committee

Table J.1 (cont’d)

Plant Species Palatabiltity For Key Wildlife Indicator Species Common Name Latin Name Black Bear Moose General Palatabilityd Ruffed Grouse Snowshoe Hare Red-backed Vole

Food Foodf Percent Food Percent Weighte Common Palatability Tolerance Food Percent Volumeb Percent Volumec* Food Foodf Food Food Mean Usea Frequencyg Usea, l Foragesj Usea Usea Usea Usek Consumptionh

Forb Layer m m wild sarsaparilla Aralia nudicaulis showy aster Aster conspicuus 3* + lady fern Athyrium felix-femina 1 2 Bunchberry Cornus canadensis 2.4 shield fern Dryopteris carthusiana + 1.2% 2 Fireweed Epilobium angustifolium 3.5 1 Medium Low common horsetail Equisetum arvense + 50* 2 * meadow horsetail Equisetum pratense + 50* 2 * scouring rush Equisetum scirpoides + 50* 2 * woodland horsetail Equisetum sylvaticum + 2 * oak fern Gymnocarpium dryopteris 2 cream-coloured Lathyrus ochroleucus 73 * Medium Low + vetchling stiff club-moss Lycopodium annotinum wild lily-of-the-valley Maianthemum canadense + 1 3.0 tall lungwort Mertensia paniculata Common Pyrola asarifolia + pink wintergreen Dewberry Rubus pubescens 22 1-2 Grass Layer m 89.33* m marsh reed grass Calamagrostis canadensis Low-Medium Low Sedge Carex spp. + 2 * Medium Medium 2 2.3% m hairy wild rye Elymus innovatus Low & Medium Medium Guidelines for Reclamation of Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX J Page J-25 Prepared By: The Oils Sands Vegetation Reclamation Committee

Footnotes to Table J.1 a Martin, A.C., H.S. Zim and A.L. Nelson. 1951. American plants and wildlife: a guide to wildlife food habits: - = Use to an indeterminate extent; + = 0.5-2 % of diet; 1 = 2-5 % of diet; 2 = 5-10% of diet; 3 = 10-25% of diet; 4 = 25-50% of diet; 5 = 50% or more of diet. Multiple values reflect regional variations in species usage. b Doerr, P.D. 1973. Ruffed grouse ecology in central Alberta - demography, winter feeding activities, and the impact of fire. Doctoral Thesis, University of Wisconsin. c Bump, G, R.W. Darrow, F.C. Edminster and W.F. Crissey. 1947. The ruffed grouse. * includes volumetric percentages to genus level, t = trace. d Hardy BBT Limited 1989. Manual of plant species suitability for reclamation in Alberta - 2nd Edition. Alberta Land Conservation and Reclamation Council Report No. RRTAC 89-4. 436 pp. e Zach, R. and K.R. Mayoh. 1982. The transfer of fallout Cesium-137 from browse to moose part 1: moose food habits. f Chapman, J.A. and G.A. Feldhamer. 1982. Wild mammals of North America: m = major foods in Canada: u = unpalatable. g Holcroft, A. and S. Herrero. 1991. Black bear, Ursus americanus, food habits in southwestern Alberta. Canadian Field-Naturalist 105 (3):335-345. Note: genus level. h Vickery, W.L. 1979. Food consumption and preferences in wild populations of Clethrionomys gapperi and Napaeozapus insignis. Can. J. Zool. 57: 1536-1542. i Beckingham, J.D. and J.H. Archibald. 1996. Field guide to ecosites of northern Alberta. Special Report No. 5, Canadian Forest Service. j Stelfox, J. B. (ed.). 1993. Hoofed mammals of Alberta. Lone Pine Publisher, Edmonton, Alberta. k Green, J.E. 1979. The ecology of five major species of small mammals in the AOSERP study area: a review. Prep. for the Alberta Oil Sands Environmental Research Program by LGL Ltd. AOSERP Report 72. 104 pp: * = species. l BOVAR Environmental and Axys Environmental Consultants. 1996. Wildlife populations and habitat resources for the Syncrude local study area and the Syncrude/Suncor regional study area. Preferred species in north-eastern Alberta: 1 = most preferred. Guidelines for Reclamation of Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX J Page J-26 Prepared By: The Oils Sands Vegetation Reclamation Committee

Table J.2 Habitat Suitability Index Values For Key Wildlife Indicator Species For Ecosite Phases And Plant Community Types Within The Syncrude Aurora Mine Area.(A)(B)

Ecosite Ecosite Phase Plant Community Type HSI (Overall) Cape May Warbler Great Gray Owl Ruffed Grouse Snowshoe Hare Red-backed Vole Fisher Black Bear Moose

a lichen a1 jack pine/lichen 0.47 - 0.24 0.23 0.49 0.45 0.67

a1.1 jack pine/bearberry/lichen 0.45 0.23 0.24 0.2 0.49 0.39 0.67 0.3

a1.2 jack pine/blueberry/lichen 0.45 0.2 0.14 0.18 0.38 0.32 0.35 0.25

a1.3 jack pine/green alder/lichen 0.45 0.23 0.25 0.31 0.41 0.35 0.81 0.3

b blueberry b1 jack pine-aspen/blueberry 0.24 0.38 0.51 0.53 0.57 0.66 0.76 0.47

b1.1 jack pine-aspen/blueberry-bearberry 0.22 0.24 0.43 0.53 0.51 0.59 0.88 0.5

b1.2 jack pine-aspen/blueberry-green alder 0.22 0.19 0.49 0.67 0.69 0.71 0.94 0.5

b1.3 jack pine-aspen/blueberry-Labrador tea - 0.29 0.64 0.3 0.47 0.6 0.9 0.33

b2 aspen (white birch)/blueberry 0.02 - 0.64 0.1 0.31 0.42 0.1 -

b2.1 aspen (white birch)/blueberry-bearberry ------

b2.2 aspen (white birch)/blueberry-green alder ------

b2.3 aspen (white birch)/blueberry-Labrador tea ------

b3 aspen (white spruce)/blueberry 0.12 - 0.47 0.63 0.65 0.69 0.62 -

b3.1 aspen-white spruce/blueberry-bearberry 0.12 0.21 0.47 0.63 0.65 0.69 0.62 0.33

b3.2 aspen-white spruce/blueberry-green alder 0.4 0.37 0.31 0.29 0.35 - 0.62 0.22

b3.3 aspen-white spruce/blueberry-Labrador tea 0.45 0.25 0.43 0.4 0.66 0.57 0.32 0.31

b4 white spruce-jack pine/blueberry 0.27 - 0.44 0.7 0.79 0.8 0.8 -

b4.1 white spruce-jack pine/blueberry-bearberry 0.25 0.17 0.44 0.7 0.79 0.79 0.8 0.36

b4.2 white spruce-jack pine/blueberry-green alder 0.25 0.17 0.36 0.68 0.76 0.83 0.54 0.41

c Labrador Tea-mesic c1 jack pine-black spruce/Labrador tea-mesic 0.47 - 0.26 0.38 0.57 0.45 0.6 -

c1.1 jack pine-black spruce/Labrador tea/feather 0.45 0.25 0.26 0.38 0.61 0.46 0.6 0.3 moss c1.2 jack pine-black spruce/green alder/feather 0.45 0.25 0.26 0.38 0.61 0.46 0.6 0.3 moss c1.3 jack pine-black spruce/feather moss 0.45 0.25 0.26 0.38 0.61 0.46 0.6 0.3 Guidelines for Reclamation of Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX J Page J-27 Prepared By: The Oils Sands Vegetation Reclamation Committee

Table J.2 (cont’d)

Ecosite Ecosite Phase Plant Community Type HSI (Overall) Cape May Warbler Great Gray Owl Ruffed Grouse Snowshoe Hare Red-backed Vole Fisher Black Bear Moose

d low-bush cranberry d1 aspen/low-bush cranberry 0.02 - 0.7 0.91 0.76 0.68 0.96 -

d1.1 aspen/Canada buffalo-berry 0.03 0.21 0.73 0.78 0.77 0.58 0.92 0.56

d1.2 aspen/saskatoon-pin cherry 0.02 0.29 0.87 0.91 0.83 0.68 0.98 0.73

d1.3 aspen/beaked hazelnut ------

d1.4 aspen/green alder 0.02 0.25 0.8 1 0.79 0.72 0.98 0.73

d1.5 aspen/low-bush cranberry 0.02 0.27 0.67 0.68 0.9 0.73 0.94 0.49

d1.6 aspen/rose 0.02 0.21 0.64 0.78 0.74 0.57 0.92 0.59

d1.7 aspen/beaked willow 0.02 0.35 0.22 0.47 0.29 0.24 0.53 0.46

d1.8 aspen/forb ------

d1.9 aspen/balsam fir 0.02 0.23 0.7 0.91 0.76 0.64 0.96 0.7

d2 aspen-white spruce/low-bush cranberry 0.11 - 0.5 0.68 0.57 0.5 0.6 -

d2.1 aspen-white spruce/Canada buffalo-berry 0.11 0.47 0.53 0.51 0.67 0.43 0.88 0.41

d2.2 aspen-white spruce/beaked hazelnut 0.11 0.21 0.5 0.68 0.57 0.44 0.6 0.44

d2.3 aspen-white spruce/green alder 0.05 0.25 0.47 0.53 0.47 0.43 0.58 0.5

d2.4 aspen-white spruce/low-bush cranberry ------

d2.5 aspen-white spruce/rose 0.11 0.21 0.46 0.61 0.54 0.4 0.6 0.44

d2.6 aspen-white spruce/beaked willow ------

d2.7 aspen-white spruce/forb ------

d2.8 aspen-white spruce/balsam fir/feather moss ------

d2.9 aspen-white spruce/feather moss 0.11 0.21 0.5 0.68 0.57 0.44 0.6 0.44

d3 white spruce/low-bush cranberry 0.85 - 0.31 0.46 0.4 0.45 0.74 -

d3.1 white spruce/Canada buffalo-berry ------

d3.2 white spruce/green alder 0.85 0.22 0.4 0.61 0.57 0.46 0.88 0.6

d3.3 white spruce/low-bush cranberry 0.65 0.38 0.38 0.31 0.53 0.61 0.61 0.58

d3.4 white spruce/balsam fir/feather moss ------

d3.5 white spruce/feather moss 0.75 0.09 0.12 0.54 0.79 0.5 0.34 0.19 Guidelines for Reclamation of Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX J Page J-28 Prepared By: The Oils Sands Vegetation Reclamation Committee

Table J.2 (cont’d)

Ecosite Ecosite Phase Plant Community Type HSI (Overall) Cape May Warbler Great Gray Owl Ruffed Grouse Snowshoe Hare Red-backed Vole Fisher Black Bear Moose

e dogwood e3 white spruce/dogwood 0.65 0.41 0.33 0.22 0.48 0.7 0.41 0.52

e3.1 white spruce/dogwood/fern 0.65 0.32 0.33 0.22 0.57 0.69 0.55 0.52

e3.2 white spruce/green alder-river alder/fern 0.65 0.41 0.33 0.22 0.57 0.69 0.55 0.52

e3.3 white spruce/balsam fir/fern ------

e3.4 white spruce/fern/feather moss ------

g Labrador g1 black spruce-jack pine/Labrador 0.38 0.11 0.09 0.81 0.45 0.39 0.48 0.19 tea-subhygric tea-subhygric g1.1 black spruce-jack pine/Labrador 0.41 0.08 0.09 0.81 0.35 0.5 0.58 0.19 tea/feather moss g1.2 black spruce-jack pine/feather moss 0.38 0.11 0.09 0.81 0.47 0.5 0.62 0.19

h Labrador h1 white spruce-black spruce/Labrador 0.38 - 0.11 0.66 0.48 0.48 0.31 - tea/horsetail tea/horsetail h1.1 white spruce-black spruce/Labrador 0.18 0.08 0.06 0.64 0.38 0.37 0.14 0.19 tea/horsetail h1.2 white spruce-black spruce/ Labrador 0.38 0.12 0.14 0.66 0.4 0.41 0.59 0.23 tea/feather moss i bog i1 treed bog 0.53 - 0.1 0.59 0.45 0.39 0.3 - i2 shrubby bog i1.1 black spruce/Labrador tea/cloudberry/peat moss 0.53 0.11 0.1 0.48 0.45 0.33 0.44 0.19

0.53 0.15 0.1 0.59 0.51 0.39 0.16 0.19

i2.1 black spruce-Labrador tea/cloudberry/peat 0.53 0.15 0.22 0.59 0.65 0.43 0.16 0.19 moss j poor fen j1 treed poor fen 0.53 - 0.22 0.74 0.56 0.49 0.32 0.37 j2 shrubby poor fen j1.1 black spruce-tamarack-dwarf 0.53 0.16 0.22 0.74 0.65 0.46 0.32 0.37 birch-sedge/peat moss 0.06 - 0.06 0.89 0.38 0.53 0.14 0.18

j2.1 black spruce-tamarack-dwarf 0.15 0.12 0.06 0.89 0.38 0.49 0.14 0.18 birch/sedge/peat moss Guidelines for Reclamation of Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX J Page J-29 Prepared By: The Oils Sands Vegetation Reclamation Committee

Table J.2 (cont’d)

Ecosite Ecosite Phase Plant Community Type HSI (Overall) Cape May Warbler Great Gray Owl Ruffed Grouse Snowshoe Hare Red-backed Vole Fisher Black Bear Moose

k rich fen k1 treed rich fen 0.53 - 0.1 0.59 0.47 0.39 0.16 - k2 shrubby rich fen k3 graminoid rich fen k1.1 tamarack/dwarf birch/sedge/golden moss 0.53 0.15 0.1 0.59 0.57 0.39 0.16 0.19 0.05 - 0.18 0.72 0.38 0.42 0.14 -

k2.1 dwarf birch/sedge/golden moss 0.06 0.14 0.08 0.64 0.42 0.34 0.18 0.17

k2.2 willow/sedge/brown moss 0.02 0.17 0.31 0.67 0.3 0.35 0.11 0.59

k2.3 willow/marsh reed grass ------

0.03 0.72 0.03 0.04 0.05 0.11 0 0.17

k3.1 sedge fen ------

k3.2 marsh reed grass fen ------

l marsh l1 marsh 0.53 0.11 0.10 0.48 0.45 0.33 0.44 0.19

l1.1 cattail marsh 0.53 0.11 0.10 0.48 0.45 0.33 0.44 0.19

l1.2 reed grass marsh ------

l1.3 bulrush marsh ------

(A) Beckingham and Archibald 1996. Field Guide to Ecosites of Northern Alberta. (B) HSI values were not calculated for great grey owl and moose for ecosite phases because these involved buffer calculations. Guidelines for Reclamation of Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX J Page J-30 Prepared By: The Oils Sands Vegetation Reclamation Committee

J4.2 BLACK BEAR

J4.2.1 Food and Cover Requirements

For much of the year, habitats selected by black bears for their food producing abilities (Table J.1) also provide adequate escape cover. The highest quality habitat occurs in forested communities with the following characteristics: a dense berry-producing shrub stratum; dense tree or shrub cover for predator avoidance and bedding; and, trees suitable for climbing to escape predators. Reclamation habitats important to black bears in the oil sands area include vegetation communities representative of the following ecosite phases (Table J.2):

· High Suitability Habitats - jack pine-aspen/blueberry (b1); aspen/low-bush cranberry (d1) and, to a lesser degree, aspen-white spruce/low-bush cranberry (d2); white spruce/low-bush cranberry (d3); and white spruce-jack pine/blueberry (b4) ecosite phases.

· Moderate Suitability Habitats - jack pine/lichen (a1); aspen-white spruce/blueberry (b3); jack pine-black spruce/Labrador tea-mesic (c1); aspen-white spruce low-bush cranberry (d2); white spruce/dogwood (e3); black spruce-jack pine/Labrador tea-subhygric (g1); and, to a lesser degree, white spruce/low-bush cranberry (d3).

J4.2.2 Landscape Component

Habitat selection varies seasonally based on food availability. Habitat interspersion is therefore an important factor determining habitat quality and the following factors should be considered in the development of reclaimed landscapes for black bears:

· During the early spring following denning, black bears select open areas where they feed on newly emergent vegetation. The extent to which open areas are utilized decreases with increasing distance from cover. Reclaimed forest communities with high cover values should be developed within 200 m of open areas, such as successional habitats or bog ecosites that serve as early spring feeding areas for black bears.

· For denning habit, reclaimed forest communities should be developed on steep north and east-facing slopes. Guidelines for Reclamation of Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX J Page J-31 Prepared By: The Oils Sands Vegetation Reclamation Committee

J4.3 SNOWSHOE HARE

J4.3.1 Food and Cover Requirements

The snowshoe hare is common and widely distributed resident of the boreal forest, inhabiting forests, swamps and riverside thickets. Although the apparent overstory preferences of snowshoe hare can be highly variable from region to region, it is widely accepted that all habitats supporting hares contain a low, dense brushy (deciduous or coniferous) understory. Within the oil sands area vegetation communities representative of the following reclamation ecosite phases provide suitable habitat for the snowshoe hare:

· High Suitability Habitats: white spruce-jack pine/blueberry (b4); aspen/low-bush cranberry (d1); aspen-white spruce/low-bush cranberry (d2); and black spruce-jack pine/Labrador tea-subhygric (g1) ecosite phases.

· Moderate Suitability Habitats: jack pine-aspen/blueberry (b1); aspen-white spruce/blueberry (b3); aspen/low-bush cranberry (d1); aspen-white spruce/low-bush cranberry (d2); and white spruce-black spruce/Labrador tea/horsetail (h1).

J4.3.2 Landscape Component

In general, habitat quality increases with increased vegetation community interspersion and the following factor should be considered when developing reclaimed areas for the snowshoe hare:

· Reclaimed landscapes should include a patchy habitat mosaic including dense thickets for winter use and more open summer range.

J4.4 RED SQUIRREL

J4.4.1 Food and Cover Requirements

The red squirrel is a resident of mature, spruce dominated forests in montane and boreal regions of North America. Coniferous trees provide escape and thermal cover. However, food appears to be the single most important factor influencing the distribution of the red squirrel. The seeds of white and black spruce are the most important dietary item and influence reproduction and territory size. White spruce seeds are favoured over black spruce seeds. Although detailed habitat modelling has not been conducted, it is believed that mature, reclaimed forest communities representative of the following ecosite phases will provide habitat for the red squirrel:

· High Suitability Habitats - jack pine/lichen (a1) white spruce-jack pine/ blueberry (b4); white spruce/low-bush cranberry (d3); white spruce/dogwood (e3); black spruce-jack pine/Labrador tea-subhygric (g1); and white spruce-black spruce/Labrador tea/horsetail (h1). Guidelines for Reclamation of Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX J Page J-32 Prepared By: The Oils Sands Vegetation Reclamation Committee

· Moderate Suitability Habitats - aspen-white spruce/blueberry (b3); jack pine-black spruce/Labrador tea-mesic (c1); aspen-white spruce/low-bush cranberry (d2); and balsam poplar-white spruce/dogwood (e2).

J4.4.2 Landscape Requirements

The following factor should be considered in the development of reclaimed forests for the red squirrel:

· Currently, there are no estimates of required block sizes for red squirrels. However, estimates of territory size range from 0.2 to 4 ha. Habitat blocks developed for red squirrel should therefore be a minimum of 4 ha and preferably much larger, probably an order of magnitude larger.

J4.5 RUFFED GROUSE

J4.5.1 Food and Cover Requirements

The ruffed grouse is a non-migratory, widely distributed species, occurring in a broad band of deciduous and mixed deciduous-coniferous forest habitats across North America. The overriding factor in habitat selection is the presence of a substantial deciduous tree component, particularly aspen and birch, in the tree canopy. Vegetation communities representative of the following ecosite phases have been ranked as important habitats for ruffed grouse in the oil sands area:

· High Suitability Habitats: aspen/low-bush cranberry (d1) ecosite phase.

· Moderate Suitability Habitats: jack pine-aspen/blueberry (b1); aspen (white spruce) blueberry (b2); aspen-white spruce/blueberry (b3); white spruce-jack pine/ blueberry (b4); aspen/low-bush cranberry (d1); aspen-white spruce/low-bush cranberry (d2); and balsam poplar-aspen/dogwood (e1) and balsam poplar-white spruce dogwood (e2) ecosite phases.

J4.5.2 Landscape and Special Habitat Requirements

The following factors should be considered in the development of reclaimed forests for ruffed grouse:

· Initial stem densities of < 14 800/ha should be established in habitat areas developed for ruffed grouse.

· Forests developed for ruffed grouse should include small forest openings 0.1 to 0.5 ha in size to provide brood rearing habitat.

· Drumming logs are important for territorial displays by male grouse, and can be added to moderate aged successional habitats to improve habitat quality. Guidelines for Reclamation of Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX J Page J-33 Prepared By: The Oils Sands Vegetation Reclamation Committee

J4.6 FISHER

J4.6.1 Food and Cover Requirements

The food and cover requirements of the fisher are provided by mature to old forest with a strong coniferous component and dense canopy closures. Within the oil sands area, suitable habitats for the fisher are provided in vegetation communities representative of the following ecosite phases:

· High Suitability Habitats: aspen-white spruce/blueberry (b3); white spruce-jack pine/blueberry (b4); and white spruce/low-bush cranberry (d3) ecosite phases.

· Moderate Suitability Habitats: jack pine-aspen/blueberry (b1); aspen-white spruce/blueberry (b3); jack pine-black spruce/Labrador tea-mesic (c1); aspen/low- bush cranberry (d1); aspen-white spruce/low-bush cranberry (d2); white spruce/ low- bush cranberry (d3); and black spruce-jack pine/Labrador tea-subhygric (g1).

J4.7 GREAT GREY OWL

J4.7.1 Food and Cover Requirements

Great grey owl breeding habitat consists of extensive forest interspersed with sphagnum bogs, muskeg and other open spaces. Nesting commonly occurs in stands of mature poplar adjacent to muskegs. Islands of poplar amid stands of spruce or pine are common nesting locations, as are groves of marginal strips of often-stunted tamaracks in wetter sites. Great grey owls primarily prey on small mammals, especially rodents. Meadow voles dominate the diet of the great grey owl over most of the owl’s range, particularly in the northern boreal forest. Because of its dependence on meadow voles, habitat utilization of great grey owls is closely linked to that of the meadow vole. Typically meadow voles occur in wet areas dominated by dense, grassy vegetation.

Within the oil sands region, great grey owls are largely dependent on one vegetation community representative of the graminoid rich fen ecosite phase. Unfortunately, this habitat cannot be developed on reclaimed landscapes. Nonetheless, the habitat requirements of the great grey owl can be incorporated in the development of reclamation landscapes adjacent to undisturbed areas dominated by this plant community (see below). Only five vegetation communities with moderate habitat quality for great grey owls can be developed on reclaimed landscapes (Table J.2). These communities are representative of the aspen-white spruce/blueberry (b3); aspen/low-bush cranberry (d1); aspen-white spruce/low-bush cranberry (d2); white spruce/low-bush cranberry (d3); and white spruce/dogwood (e3) ecosite phases. Guidelines for Reclamation of Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX J Page J-34 Prepared By: The Oils Sands Vegetation Reclamation Committee

J4.7.2 Landscape Component

Habitat interspersion is an important aspect of habitat selection and the following factors should be incorporated into the development of reclaimed landscapes for great grey owls:

· Typically this species is reliant on open grassy areas for foraging, and mature forested areas for nesting. These two habitat conditions must therefore occur to adjacent to one another to create suitable nesting areas. Ideally mature forest communities, with an aspen component, should be established within 500 m of graminoid rich fen ecosite phases left in undeveloped area adjacent to reclaimed landscapes. · The habitat suitability of foraging areas could be increased by establishing perches in foraging areas. This could be done by planting individual or small isolated patches of trees, or by placing poles in graminoid rich fen ecosite phases. · Nest site availability could be increased by establishing artificial nest sites (nest baskets or nest poles) in forested areas located within 500 m of foraging habitat.

J4.8 CAPE MAY WARBLER

J4.8.1 Food and Cover Requirements

The Cape May warbler is a neotropical migrant breeding in the boreal forest of North America. Cape May warblers breed in mature white spruce stands within coniferous and mixed wood forests, preferring open stands and stand edges. Within the oil sands area, vegetation communities representative of the following ecosite phases will provide breeding habitat for the Cape May warbler:

· High Suitability Habitat: white spruce/low-bush cranberry (d3) and white spruce-jack pine blueberry (b4) ecosite phase. · Moderate Suitability Habitat: jack pine/lichen (a1); jack pine-aspen/blueberry (b1); aspen-white spruce/blueberry (b3); jack pine-black spruce/Labrador tea-mesic (c1); white spruce/low-bush cranberry (d3) ecosite phase; white spruce/dogwood (e3); and black spruce-jack pine/Labrador tea-subhygric (g1).

J4.8.2 Landscape Component

The following factors should also be considered in the development of re-constructed landscapes for Cape May warblers:

· Edge area should be maximized in vegetation communities planted for Cape May warblers. · Males select tall conifers that rise above the rest of the canopy for singing perches. This condition should be mimicked in reclaimed forests established for Cape May Guidelines for Reclamation of Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX J Page J-35 Prepared By: The Oils Sands Vegetation Reclamation Committee

warblers. This could possibly be accomplished by planting low densities of white spruce (25-50 m spacing) in suitable habitat areas and allowing the rest of the forest to develop around these trees.

J4.9 OVENBIRD

J4.9.1 Food and Cover Requirements

The food and cover requirements of the oven bird are met in mature, closed deciduous and mixed wood forests with little ground cover. Mature, reclaimed vegetation communities representative of the following ecosite phases will most likely provide breeding habitat for ovenbirds: jack pine-aspen/blueberry (b1); aspen (white birch)/blueberry (b2); and aspen-white spruce/blueberry (b3).

J4.9.2 Landscape Component

The following factor should be incorporated into the development of reclaimed landscapes for ovenbirds:

· Habitat blocks that are a minimum of 4 ha in size are required before breeding ovenbirds will occupy an area.

J4.10 WARBLING VIREO

J4.10.1 Food and Cover Requirement

Warbling vireos are associated with old (60+ years) deciduous and mixed wood forests with an open canopy and dense shrub understory. Mature forest communities representative of the following ecosite phases will provide breeding habitat for the warbling vireo: jack pine- aspen/blueberry (b1); aspen (white birch)/blueberry (b2); aspen-white spruce/blueberry (b3); aspen/low-bush cranberry (d1); and aspen-white spruce/low-bush cranberry (d2).

J4.10.2 Landscape Component

Habitat interspersion can improve habitat quality and the following factor should be considered in the development of reclaimed landscapes for the warbling vireo:

· Early successional and other shrub dominated communities should be established adjacent to open, mature deciduous dominated forests to increase the amount of foraging habitat available for warbling vireos. Guidelines for Reclamation of Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX J Page J-36 Prepared By: The Oils Sands Vegetation Reclamation Committee

J4.11 RED-BACKED VOLE

J4.11.1 Food and Cover Requirements

Although the red-backed vole occupies a wide variety of plant communities, it is generally most abundant in moist, mature forests with relatively dense shrub canopies and abundant litter, moss and deadfall. Clearings and other unforested habitats are generally avoided unless an abundance of protective ground litter and shrub cover is present. Within the oil sands region, reclaimed forest communities representative of the following ecosite phases will provide habitat for the red-backed vole:

· High Suitability Habitats - jack pine-aspen/blueberry (b1); aspen-white spruce/ blueberry (b3); white spruce-jack pine/blueberry (b4); and aspen/low-bush cranberry (d1) ecosite phases.

· Moderate Suitability Habitats - jack pine/lichen (a1); jack pine-aspen/blueberry (b1); aspen-white spruce/blueberry (b3); jack pine-black spruce/Labrador tea-mesic (c1); aspen-white spruce/low-bush cranberry (d2); white spruce/dogwood (e3); and black spruce-jack pine/Labrador tea- subhygric (g1).

J4.12 DEER MOUSE

J4.12.1 Food and Cover Requirements

The deer mouse is a widely distributed species, which shows few restrictions in habitat use. Preferred habitats include deciduous, mixed and coniferous forests and shrublands with dense ground and litter cover. Early successional habitats are also occupied by the species. A two- year study conducted in the region revealed that deer mice preferred balsam poplar and mixed wood forests and early successional habitats. In contrast, willow-birch and tamarack habitats were avoided and jack pine and black spruce forests received marginal use. Some differences in habitat use were, however, noted between the two years of the study. Based on the results of this study, habitat preference in the oil sands area will probably be shown towards reclaimed vegetation communities representative of the following ecosite phases:

· High Suitability Habitat: aspen (white birch)/blueberry (b2); aspen-white spruce/ blueberry (b3); aspen/low-bush cranberry (d1); aspen-white spruce/low-bush cranberry (d2); balsam poplar-aspen/dogwood (e1); balsam poplar-white spruce/ dogwood (e2); white spruce/dogwood (e3); and white spruce-black spruce/Labrador tea/horsetail (h1).

· Moderate Suitability Habitat: jack pine/lichen (a1); jack pine-aspen/blueberry (b1); white spruce-jack pine/blueberry (b4); jack pine-black spruce/Labrador tea- mesic (c1); white spruce/low-bush cranberry (d3); and black spruce-jack pine/ Labrador tea-subhygric (g1). Guidelines for Reclamation of Forest Vegetation October, 1998 In the Alberta Oil Sands Region – APPENDIX J Page J-37 Prepared By: The Oils Sands Vegetation Reclamation Committee

J5. LITERATURE CITED

Wildlife Management Division. 1996. The Status of Alberta Wildlife. Wildlife Management Division, Natural Resources Service, Alberta Environmental Protection. Publication No. I/620. 44 pp.