Pinyon Juniper Cover Across the Southern Rockies Landscape Conservation Cooperative; Data from LANDFIRE Existing Vegetation Type (EVT)

Spatially Explicit Vulnerability Assessment of Piñon-Juniper Ecosystems in the Four Corners and Upper Rio Grande Landscapes

Prepared by Megan Friggens, Stephanie Mueller, and Mary Williams

September 30th, 2018

Picture Credit: Pinus edulis (Pinyon pine) from Uniprot (http://www.naturesongs.com/vvplants/pinyon1.jpg); P. edulis branch with Pinyon Jay, By Sally King, National Park Service; Juniperus occidentalis from Lava Beds, NM. By Walter Seigmund 2015; Juniper berries, National Park Service

Megan Friggens is a Research Ecologist with the USDA Forest Service, Rocky Mountain Research Station.

Stephanie Mueller is a graduate student at Northern Arizona University.

Mary Williams is an Ecologist with the Nez Perce Tribe, Lapwai, Idaho.

This analysis was prepared in partial fulfillment of agreements between Rocky Mountain Research Station (RMRS) and the Southern Rockies Landscape Conservation Cooperative (SRLCC: FWS #15-IA- 011221632-164 and BOR #15-IA-11221632-163). This is one in a series of four assessments produced for focal landscapes within the SRLCC. Data associated with this assessment can be found at the SRLCC Conservation Planning Atlas web portal: https://srlcc.databasin.org. Alternatively, spatial datasets and reports are available on the ScienceBase Website here: https://www.sciencebase.gov/catalog/item/569414a2e4b039675d00472e and here https://www.sciencebase.gov/catalog/item/5693e5a9e4b09c7f9a21a428 and here https://www.sciencebase.gov/catalog/item/5693e5c7e4b09c7f9a21a42d

Acknowledgements: We thank the participants of the 2016 and 2017 Adaptation Forums held in Durango, CO, and Albuquerque and Taos, NM. We thank Jack Triepke for sharing data and preliminary reports from the USFS Region 3 Vulnerability Assessment. Logistical support was provided by David Hawksworth.

For questions or comments please contact Megan Friggens, [email protected]

Abbreviations: AS Adaptation Strategy BLM Bureau of Land Management CNHP Colorado Natural Heritage Program FS Forest Service IAP Intermountain Adaptation Partners LCD Landscape Conservation Design NCCSC North Central Climate Science Center PJ Piñon Juniper SRLCC Southern Rockies Landscape Conservation Cooperative VA Vulnerability Assessment WWA Western Waters Association

Contents Introduction ...... 5 Background ...... 5 Vulnerability Assessments ...... 7 I. Literature Review of Piñon-Juniper Ecosystems in the four Corners and Upper Rio Grande Focal Areas 9 Geographic Focus ...... 9 Introduction ...... 9 Composition and Structure ...... 11 Threats and Issues for Piñon-Juniper Ecosystems in the Four Corners and Upper Rio Grande Focal Areas ...... 13 Disturbances ...... 13 Climate ...... 18 Drought ...... 19 Current Status ...... 22 Expansion ...... 23 Declines ...... 24 Future Status ...... 25 II. Vulnerability Assessment ...... 30 Method and Approach ...... 30 Background ...... 30 Structure ...... 31 Measuring Vulnerability ...... 31 Spatial Units ...... 32 Indicator Variables ...... 32 Results ...... 38 Exposure (Table 2.4, Fig. 2.5) ...... 38 Sensitivity (Table 2.5, Fig. 2.6) ...... 39 Adaptive Capacity (Table 2.6, Fig. 2.7) ...... 41 Impact (Table 2.7, Fig. 2.8) ...... 42 Vulnerability (Table 2.8, Fig. 2.9) ...... 43 Discussion...... 44 Overview ...... 44 Comparison of Upper Rio Grande and Four Corners Areas ...... 44

Using This Data ...... 45 Missing Data and Other Uncertainties ...... 45 Management Considerations ...... 47 Literature ...... 49 Appendix 1. Additional Resources ...... 65 Appendix 2. Indicator Datasets ...... 67 Exposure...... 67 1. Temperature ...... 67 2. Decline in Winter Precipitation ...... 68 3. Change in Development ...... 68 4. Road Density ...... 69 5. High to Very High Fire Potential ...... 69 6. Insect and Disease Threats ...... 70 7. Loss of Climate Suitability (Great Basin Conifer Woodlands) ...... 71 Sensitivity ...... 71 1. Presence of Energy Development ...... 71 2. Current Development ...... 72 3. Mechanically Disturbed Forests ...... 73 4. Piñon-Juniper Habitat ...... 73 5. Wildlife Diversity ...... 74 6. Available Water Storage Capacity ...... 75 Adaptive Capacity ...... 76 1. Increase in Suitable Climate ...... 76 2. Protected Land Designation ...... 76 3. Piñon-Juniper Habitat ...... 77 4. Low Fire Potential ...... 78 References ...... 79

Introduction A major challenge of managing rapidly changing landscapes is translating broad concepts into specific, tangible actions. Vulnerability assessments are a key step toward identifying effective management strategies and prioritizing specific management actions to conserve and restore natural resources. As part of the effort to utilize landscape conservation planning and design for the Southern Rockies Landscape Conservation Cooperative (SRLCC), the RMRS developed a vulnerability assessment approach that: 1) covers a diverse set of focal resource targets; 2) incorporates input from science and working group members and other stakeholders; 3) compliments the overarching principles guiding science and management planning within the LCCs (Box 1). Our approach borrows from other successful assessments strategies and used Adaptation Forums to facilitate stakeholder BOX 1. COORDINATING WITH THE LCC STRATEGIC input. As part of this work, RMRS has PLAN produced a series of state of knowledge The vulnerability assessment process and its products syntheses and developed spatially explicit contribute to multiple objectives outlined within the LCC vulnerability assessment products including Network Strategic Plan. maps, datasets and training materials. It is 1. Vulnerability Assessments identify the relative the objective of this project that these vulnerability of focal targets to climate change products provide the SRLCC stakeholders thereby providing mechanisms for prioritizing with the capacity to identify shared science and management needs. conservation priorities and goals. All data 2. When conducted for multiple time periods or and associated webinar and workshop climate scenarios, measures of vulnerability may also identify urgency. information is available online to facilitate 3. Vulnerability Assessments identify how and why additional analyses and assessments resources may be vulnerable to climate change identified by users. The following report thereby providing starting points for management contains the synthesis and analysis planning, conservation, and monitoring. component of this project for piñon-juniper 4. Vulnerability Assessments consider measures of (PJ) ecosystems. In the first half of this uncertainty and identify areas needing additional report, you will find a review of literature research. for piñon-juniper (PJ) ecosystems within two SRLCC focal areas: The Four Corners and Upper Rio Grande (see https://southernrockieslcc.org/geographical-areas for more information). The second half of this report presents the methods and results of our vulnerability assessment.

Background In 2016, the Rocky Mountain Research Station began to develop vulnerability assessment products for SRLCC focal resources. Prior to the involvement of RMRS, SRLCC staff and steering committee members identified five focal resources, native fish, stream flow, elk and mule deer, sagebrush ecosystems, and cultural resources, and three focal landscapes, the Green River Basin, the Four Corners, and the Upper Rio Grande (https://southernrockieslcc.org/geographical-areas). The assessments conducted by RMRS regard focal resources in the Four Corners and Upper Rio Grande (Fig. 1.1). After receiving feedback from a series of Adaptation Forums in 2016, we refined the focal resources targets to include riparian

Figure 1.1 Southern Rockies Landscape Cooperative (SRLCC) boundaries and focal areas addressed this assessment. areas and piñon-juniper habitats. We also dropped cultural resources from the current effort (now addressed through another project: https: //southernrockieslcc.org/issue /cultural-resources). The objectives of the vulnerability assessment project were multiple: 1. Summarize current status, potential climate change impacts, and other issues facing SRLCC Focal Resources in the Upper Rio Grande and Four Corners Geographic Focus Areas. 2. Identify climate change impacts and other issues facing natural and cultural resources in the southern tier of SRLCC (includes the Upper Rio Grande and Four Corners Geographic Focus Areas).

3. Develop a Vulnerability Assessments approach to allow integration with other assessments and Landscape Conservation Design (LCD) efforts of BLM, FS, CNHP and other partners. 4. Apply Vulnerability Assessments approach to quantify vulnerability of SRLCC Focal Resources and Geographic Focus Area select targets.

Vulnerability Assessments Vulnerability assessments are a critical component in adaptive management planning and risk analysis. An assessment of vulnerability can identify relative impacts from disturbance and the source of those impacts, thereby facilitating the identification and prioritization of management strategies. Vulnerability is a key concept for assessing climate impacts to natural resources and has been adopted by the Forest Service and other agencies as a primary mechanism for developing effective adaptation options to manage natural resources under climate change (e.g. USDA Forest Service 2011). Climate change and related vulnerability assessments (hereafter VA) are initiated during the planning phase of conservation projects to provide information on relative impacts and priority needs within a target system. VA are amendable to multiple data sources and incorporate a variety of analysis methods thereby providing a framework for creating diverse data products that might be useful for planning efforts. Thus, the utility of VA can extend beyond initial efforts to identify climate change impacts. We developed a vulnerability assessment framework based on the core principles common to many climate change vulnerability assessments (Box 2).

As commonly applied to climate change issues, VA provide a structure for organizing complex BOX 2. PRIMARY ELEMENTS OF A VULNERABILITY information and addressing uncertainty (IPCC ASSESSMENT 2007). Although there are various definitions, Exposure: The magnitude of climatic or ecological vulnerability is generally thought of as the changes within target landscape Sensitivity: The response of targets to exposure susceptibility of a target to negative impacts Adaptive Capacity: The potential of target to cope with from some disturbance (Füssel 2007; Hinkel exposure 2011). Assessment of climate change vulnerability typically considers three elements: exposure, sensitivity, and adaptive capacity (Glick et al. 2011). Exposure is the magnitude of climate and climate-related phenomena (e.g., fire, floods) whereas sensitivity (i.e., response to exposure) and adaptive capacity (i.e., ability to cope with negative impact) are traits or conditions that predict how a target will respond to that disturbance. These definitions can vary according the goals and the target of an analysis. For instance, sensitivity may represent the innate traits or qualities of a target that increase the likelihood it will experience a negative response. Alternatively, sensitivity may represent the potential cost of a disturbance (e.g. watershed values - Furniss et al. 2013). Adaptive capacity can be identified as the intrinsic and/or externally driven mechanisms that represent the potential for a target or system to withstand a disturbance.

The assessment approach used in this analysis was designed to complement LCC Network Strategic Plan objectives and produce products and outputs with multiple uses (Table 1.1; Box 1). Assessment products

provide information on the overall vulnerability of the focal resource in each landscape and identify the most sensitive or resilient features within those systems. To implement the assessment process, we identified relevant spatial data that can be used to assess areas of high risk or resilience within the landscapes (see Section 2.Methods and Appendix 2 for more Information). These spatial datasets form the core of our assessment scores and can be used to construct maps that classify areas according to the presence of vulnerable components.

Table 1.1. Planned inputs and outputs generated as part of the vulnerability assessment (VA) process have a variety of uses. VA element Types of Inputs/Outputs Additional uses Exposure (degree change or Maps depicting % change in temp, Identify areas falling outside of known disturbance) prcp, cover. range of variability (losers) Predicted land use change Identify areas that maintain natural Spatial data on distribution of cultural range of variability (winners, “Refugia”) and natural resources Identify spatial distribution and overlap of multiple resources Analysis of fragmentation, etc. Risk Assessments (likelihood and magnitude of change)

Measures/indicators of Maps showing: Identify changes to biodiversity sensitivity or adaptive capacity Biodiversity (turnover) Socio-economic vulnerability Biodiversity hotspots current and Shifts in distribution of species and future vegetation Identify monitoring targets Water quality change Identify management targets Percent change in distribution of Identify resilient landscapes habitat or habitat features Identify sensitive landscapes Various estimates of magnitude of Identify critical landscapes or change landscape features Risk Assessments (likelihood and magnitude of change) Identify threshold indicators, likelihood of type conversion

Identification of critical Ranked vulnerability across Incorporate results into agency vulnerabilities elements/locations adaptation plans/strategies Information on key sensitivities and Identify future management and adaptive capacity monitoring priorities Identify actions addressing specific threats and issues Identification of best bet and ROI actions

I. Literature Review of Piñon-Juniper Ecosystems in the four Corners and Upper Rio Grande Focal Areas Geographic Focus The Southern Rockies Landscape Conservation Cooperative (SRLCC), encompassing over 127 million acres, covers portions of Arizona, Colorado, New Mexico, Utah, and Wyoming (Fig. 1.1). The focal areas of interest within the SRLCC, Four Corners (FC) which covers parts of Arizona, Colorado, New Mexico, and Utah, and Upper Rio Grande (URG) which covers Colorado and New Mexico are located at the southern end of the SRLCC. Land in the focal areas is primarily owned and managed by Tribes, States, Bureau of Land Management (BLM), Forest Service (FS), and National Park Service (NPS). Tribal governments, including the Navajo Nation and The Hopi Tribe, manage approximately 28% of the land in Arizona (AGFD 2012; Schussman and Smith 2006). Other management agencies include the Bureau of Reclamation (BOR), Department of Defense (DOD), Fish and Wildlife Service (FWS), and Department of Energy (DOE).

Introduction Piñon-Juniper (PJ) ecosystems include a range of forested, woodland and savanna landscapes within the SRLCC focal area (Fig. 1.2). Piñon-Juniper woodlands, the focus of many studies and this assessment, comprise nearly 40 million hectares in the western U.S. (Romme et al. 2009). Occurring in the transitional zone between lowland grassland and shrubland ecosystems and mountain conifer ecosystems (1370-2290 m), piñon-juniper woodlands consist of evergreen pines and junipers with other shrubs, grasses, and forbs (Gori and Bate 2007; Nielson 2009). The distribution and abundance of characteristic tree species in PJ systems depends on topography, climate, and soil. Within the SRLCC, the most common piñon tree, Colorado or two-needle piñon pine (Pinus edulis, PIED) is found with one- seed juniper (Juniperus monosperma), Utah juniper (Juniperus osteosperma), Rocky Mountain juniper (J. scopulorum), and alligator juniper (J. deppeana). Singleleaf piñon (P. monophylla) is less common than PIED with scattered locations in southwestern Utah, where it typically co-occurs with Utah juniper and Rocky Mountain juniper (Little 1971). Piñon species dominate in stands at upper elevations or in more mesic areas and Juniper tends to dominate lower and drier elevation sites (Gottfried et al. 1995). Both piñon and juniper species are components of mixed conifer forests (Romme et al. 2009).

Piñon-Juniper ecosystems have cultural and ecological significance in western landscapes. PJ historically provided human inhabitants with an important source of food, medicine, fire wood, building materials and fibers (Janetski 1999; Flint-Lacey 2003; McCabe et al. 2005; Tilford 1997; Schlanger and Larralde 2008). Piñon nuts are an excellent source of food and were harvested extensively by Native American populations within the southwest. Fresh or dried juniper berries were also consumed and used as a flavoring agent in soups, tea and bread. Juniper was also important for providing fuel and materials to build shelters (Janetski 1999). Currently, PJ ecosystems are important for livestock grazing, recreation, and firewood.

From an ecosystem perspective, PJ plays an important role for both plant and animal communities. Within PJ ecosystems, trees create islands of high fertility soils by accumulating soil moisture, organic

Figure 1.2 Pinyon Juniper cover across the Southern Rockies Landscape Conservation Cooperative; data from LANDFIRE Existing Vegetation Type (EVT). material and nutrients beneath their canopies (Barth 1980; Nielson 2009) and may increase water retention in some areas (Madsen et al. 2008). Piñon-juniper woodlands also provide valuable cover, food, and nesting sites for many wildlife species (Box 3; Gallo et al. 2016; Bombaci and Pejchar 2016). Piñon juniper woodlands support high wildlife diversity comparable to that found in riparian and aspen forests (Bombaci and Pejchar 2016). In Colorado alone PJ supports 67 species of greatest conservation need and 39% of species that use PJ are obligate or semi-obligate (CO SWAP 2014; Paulin et al. 1999). Not all PJ stands are equally important, however, and many wildlife species are associated with more mature and dense woodland types (Johnson et al. 2016; Gallo et al. 2016). Indeed, breeding bird diversity and abundance has been positively correlated with piñon juniper woodland tree density and size (Rosenstock and van Riper 2001 and references therein). In turn, PIED regeneration is dependent on several species, including Scrub jay, piñon jay, Steller’s jay, and Clark’s nutcracker (Evans 1988; Hall and Balda 1988; Ronco 1990; Zouhar 2001) and juniper species also rely on birds for seed dispersal.

Composition and Structure Across its range, piñon-juniper ecosystems vary in extent, successive stage and condition. Romme et al. 2009 describes three distinct piñon-juniper BOX 3. WILDLIFE AND PIÑON-JUNIPER ECOSYSTEMS ecoystem types, savannas, wooded PJ associated species including the piñon mouse, woodrats shrublands, and persistent woodlands, (Stephen’s woodrat), piñon jay, gray flycatcher, screech owl, each with unique climate associations scrub jay, plain titmouse, and gray vireo (Short and (Table 1.2). Piñon-Juniper savannas are McCulloch 1977; Balda and Masters 1980; Meeuwig et al. common in basins, foothills and areas with 1990; Morrison and Hall 1999; Bosworth 2003; Bombaci and Pejchar 2016) – some of which are important prey mostly monsoonal moisture trends. Warm populations for large mammals and raptors, such as season rains favor growth of warm season ferruginous hawks (Zouhar 2001). PJ woodlands provide grasses typically found in these savannas. important wintering habitat for Clark’s nutcracker (Vander PJ wooded shrublands are more common Wall et al. 1981), mule deer (Evans 1988), and mountain where winter precipitation predominates lions (Laing 1988; Laundre and Hernandez 2003). Elk, mule including areas in northern Arizona and deer, bighorn sheep, and pronghorn rely on forage and cover New Mexico. PJ wooded shrublands are found in these woodlands (Zouhar 2001; Anderson 2002). often found in areas with recent woodland Many lizards and snakes find food and shelter on and in expansion and/or contraction. Persistent trees and within downed woody debris in PJ forests PJ woodlands are most common in areas (Bosworth 2003; Corkran and Wind 2008; Oliver and Tuhy 2010). Water resources located near PJ woodlands are dominated by winter or bimodal essential to Great Basin spadefoot, tiger salamander, many- precipitation patterns. These woodlands lined skink, ornate tree lizard, ring-necked snake, common are sometimes described as unproductive kingsnake, and terrestrial gartersnake (Pilliod and Wind and containing a sparse understory cover 2008). PJ located near cliffs, caves, and riparian areas, though old-growth piñon stands are provide habitat for peregrine falcons (Craig and Enderson considered important for a diverse range of 2004), Allen’s big-eared bat, long-eared myotis, little brown fauna and flora (Floyd et al. 2015). bat, Yuma myotis, fringed myotis (tree rooster), hoary bat, silver-haired bat (tree rooster), western pipistrelle, and The SRLCC landscape bridges two spotted bat (Oliver 2000; Bosworth 2003; Valdez and Cryan climatically distinct areas: the Colorado 2009; Rhea et al. 2013). plateau, which typically experiences precipitation peaks in winter and summer months, and the Southwest, where summer monsoons dominate. Thus, though covering a relatively small geographic portion of the range of PJ, this assessment considers the interface of several PJ types. Board et al. (2018) identifies four PJ types within our SRLCC study area: Colorado Plateau Pinyon-Juniper Shrubland (NW corner), Colorado Plateau Pinon-Juniper Woodland (predominate type), Southern Rocky Mountain PJ woodland, and, Southern Rocky Mountain Juniper woodland and Savanna (USGS 2011 National Gap Analysis Program Land Cover Data). Of these types, this assessment primarily focuses on PJ woodlands. In particular, mature PJ forests are associated with a high degree of biological complexity and appear most at risk of declines due to a variety of stressors (Floyd et al. 2015). Importantly, mature piñon forests represent a late successional ecosystem that takes 200+ years to establishment in previously uninhabited areas (Cole et al. 2008). Therefore, potential declines have long term implication. Once lost, the mature PJ woodlands critical to many wildlife species are unlikely to return in our lifetime.

Compositional changes in PJ ecosystems are driven by differential effects of climate and disturbance on tree, shrub and grass species (Jacobs et al. 2008; Margolis 2014). Piñon and juniper trees have unique reproductive and metabolic processes that predispose their response to

Table 1.2 Piñon-juniper woodland types. Adapted from the National Park Service Series: Piñon-Juniper Woodlands, Chapter 2: Species Composition and Classification (https://www.nps.gov/articles/series.htm?id=0216D798-933C-2108-EB4384D97499E89A)

Savanna Wooded Shrubland Persistent Woodland

Precipitation Receives mainly summer Most often occurring (but Receives mainly winter precipitation. Bi-modal not restricted to) areas of precipitation. precipitation in some areas. winter precipitation.

Soils Deeper, fine-textured soils Wide variety of substrates. Shallow, often rocky soils do support a range of vegetation not support continuous types. vegetative cover.

Terrain Gentle terrain- valleys, Wide variety of topography- Rugged upland sites and basins, and foothills. plains, valleys, and lower steep, rocky terrain. montane.

Stand Structure Open savanna-like stand Areas of woodland Multi-aged stand structure. structure, supports low expansion and contraction. Range of tree densities and density of trees and shrubs, Shifts from herbaceous, to canopy cover, depending on and dense herbaceous shrub, to tree dominance site conditions. growth: grasses, forbs, and over time, in the absence of annuals. fire. Often shrub- dominated, with trees colonizing when growing conditions are favorable.

Historical Fire Regime Frequent, lower-intensity Infrequent, high-severity Infrequent high-severity surface fires maintain grasses fires and patchy, mixed- fire. Significant barriers to and open stand structure. severity fires. fire spread: cliffs, canyons, Few barriers to fire spread. exposed rock, topographic isolation. precipitation change and disturbances like fire (Table 1.3). In general, piñon species are more susceptible to declines following disturbance and drought, and are more susceptible to insect related mortality (Redmond et al. 2013). In recent decades, piñon pine has experienced widespread mortality due to the interactive effects of drought and insect outbreaks (Meddens et al. 2015). Greater than 95% of one seed piñon pine trees died in a northern New Mexico site surveyed during a recent (2002-2003) drought period (Allen 2007). Juniper is also affected by drought and insects but appears more resilient to both. In addition, juniper, in particular J. osteosperma, is an early successional species that has higher establishment following disturbances compared to P. edulis and P. monophylla (Chambers et al. 1999; Redmond et al. 2013). As a result, some persistent PJ woodlands have experienced a shift to juniper

dominated ecosystems in recent decades (Romme et al. 2009; Koepke et al. 2010; Redmond and Barger 2013). In some areas, PJ woodlands have experienced shrub encroachment due to drought and, in other areas, increasing establishment of invasive grasses has altered fire regimes and encouraged stand replacing fire to the detriment of both piñon and juniper. Threats and Issues for Piñon-Juniper Ecosystems in the Four Corners and Upper Rio Grande Focal Areas Assessments conducted for PJ have identified energy and mineral development, fire regime alteration, habitat loss and fragmentation, vehicles/roads/recreation, improper grazing, exotic/invasive plant species, motorized recreation, and vegetation treatments as threats to Piñon-Juniper woodlands (Tuhy et al. 2002; CO SWAP 2015; Appendix 1). Exposure to these disturbances varies geographically across the range of PJ and individual stands may be more or less resilient to specific issues. Local disturbance, climate and fire regimes all influence piñon and juniper recruitment and survival. Research is relatively limited for the current study area with the greatest body of information focused on PJ in the Great Basin area of the United States, where climates support different types of PJ species and systems. The following discussion focuses on the primary issues facing PJ woodlands that are likely to hold the most relevance for the Four Corners (FC) and Upper Rio Grande (URG) landscapes.

Disturbances Fire Piñon-juniper woodlands are considered a shade intolerant climax community and generally not well adapted to wildfire (Tuhy et al. 2002). Piñon pine and juniper trees have thin bark, low crowns and are easily killed by wildfire (Nielson 2009; Ryan and Reinhardt 1988; Margolis 2014) with younger trees more likely to experience fire-related mortality (Margolis 2014). Once disturbed by wildfire, it takes several decades for PJ to become re-established (Nielson 2009). As such, wildfire is thought to have limited the historic extent of PJ woodlands in some areas (e.g. Colorado, CO SWAP 2014). Alligator juniper is perhaps the only fire tolerant species and is capable of resprouting after BOX 4. RECENT TRENDS FOR WILDFIRE IN PJ a fire if exposed when it is relatively young. Board et al. (2018) provide a comprehensive analysis of recent fire regimes within PJ systems and note a significant Fire suppression has been implicated as increase in annual area burned, number of fires per year, driving infill in PJ savannas and wooded fire season and fire size from 1984 through 2013 in the shrublands and is listed as a cause of recent Southern Rocky Mountain geographic regions. Within the increases in tree density and cover in Southwest region, Arizona and New Mexico mountains persistent woodlands (Margolis 2014; experience the largest increases in annual area burned. Board et al. 2018). Margolis (2014) suggests Importantly, the Southern Rocky Mountain geography was that fire exclusion, over other potential the only region to experience significant increases in fire mechanisms, is responsible for increased season length and fire size (Board et al. 2018). Observed fire tree density in PJ savannas, especially in return intervals over this same period ranged from 105 to areas with monsoon dominated 7000 years with an average of 702 years. precipitation patterns. This may or may not be true for mature PJ woodlands where the

role of wildfire and fire suppression likely varies with local conditions, land use history, and climate conditions (Board et al. 2018). Additionally, suppression may increase risk of moisture stress during drought where it leads to an increase in high density/small diameter stands, which, in turn supports high severity fires that lead to high tree mortality (Clark et al. 2016).

In most systems, piñon-juniper communities are associated with infrequent high-severity fire regimes occurring at intervals around 400-480 years (Baker and Shinneman 2004). Studies on PJ communities close to or within the current focal areas report similar or somewhat shorter fire return intervals:

 Multiple studies describe fire intervals of 300-500 years for PJ on western slopes of the Rocky Mountains in Colorado (CO SWAP 2014; Eisenhart 2004; Romme et al. 2009).  Huffman et al. (2008) describes widespread fire across high-elevation piñon-juniper stands in Arizona and New Mexico, with fire return intervals of 290-340 years.  Studies indicate PJ woodlands in the Colorado Plateau likely had infrequent (>250 year interval) high severity fires (Floyd et al. 2004; Huffman et al. 2008; Shinneman and Baker 2009).

Long fire return intervals have been attributed to fuel patchiness and rugged topography (e.g., canyons, rocky soils) that prevents fire spread beyond a single or a few trees (Romme et al. 2003). Evidence suggests the occurrence of stand replacing fires has increased with increasing tree expansion and density since the 1800’s (Box 4; Romme 2009; Board et al. 2018)

Though wildfire is widely recognized as having an important role in structuring piñon-juniper ecosystems, specific fire histories have been debated (Brown et al. 2001; Floyd et al. 2000). Given variation in piñon-juniper vegetation and precipitation regimes across its range, it is not surprising that PJ ecosystems exhibit a range of responses when exposed to wildfire (Table 1.2; Jacobs 2008; Romme 2009). Persistent woodlands are considered relatively resilient to low surface fires where, historically, stand replacing fires are rare (Carroll et al. 2016). Frequent low-severity surface fires were likely important for maintaining piñon-juniper stand structure (Shinneman 2004), especially within PJ savannas (Romme et al. 2009; Margolis 2014). Margolis (2014) suggests low severity fires are most likely in systems like the PJ savanna sites in Northwestern New Mexico (at upper end of elevational gradient) because they have enough warm season precipitation to support sufficient grass cover for fire spread. However, evidence for frequent low severity fire has not been found for all PJ types and locations (Floyd et al. 2004; Shinneman and Baker 2009).

Fire regimes, as well as tree response to fire, are influenced by fuels, tree density, biomass, canopy cover and fire weather (Romme 2009; Board et al. 2018). In areas with minimal or native grass cover, individual trees are more like to survive fire where tree density is low (Romme 2009; Board et al. 2018). Higher levels of grass cover support better fire spread and greater tree mortality under extreme fire weather conditions (Romme et al. 2009). In areas with shrub co-dominants, fire intensity is greater and corresponds to increased risk of torching and tree mortality under both typical and extreme fire weather. For mature stands, high severity stand replacing fire is more likely due to increased tree densities (Board et al. 2018).

Large scale projections of future conditions show an increase in wildfire in woodland habitats and, as a result, shifts from woodland to grassland and shrubland ecosystem types (Moritz et al. 2012). For piñon- juniper woodlands in the Intermountain West, short-term projections favor an increase in wildfire (Rocca et al. 2014) and an increase in tree death due to wildfire (Baker and Shinneman 2004). Decker et al. (2015) considered PJ ecosystems in Colorado highly vulnerable to climate change due to increasing issues with fire and insect outbreaks. Within the Southern Rocky Mountain Region, elevated CO2 that increases woody biomass and invasive annual grasses will likely contribute to this risk (Ramsey and West 2009; Romero-Lankao et al. 2014; Board et al. 2018). Board et al. (2018) note that projected changes in precipitation regimes are likely to increase fire season length and tree mortality in PJ in the Southern Rockies Region. Changes in the timing of fire could also be problematic when it kills freshly dropped seeds or seedlings (Chambers et al. 1999). Areas that experience increased fire are likely to transition to grasslands or shrublands because post-fire recovery is very slow, particularly where there is a lack of precipitation (Romme et al. 2003; Rocca et al. 2015).

Grazing Livestock grazing in combination with fire suppression is cited a major driver of increased PJ tree density and extent in Western landscapes (Gottfried et al. 1995; Gori and Bate 2007; Margolis 2014). Grazing can benefit PJ where it reduces herbaceous cover and, thereby, competition for nutrients and moisture and fine fuels that carry fire between trees. These benefits vary according to site characteristics and PJ composition, however. The greatest benefit is likely for savannah habitats or at grassland/woodland interfaces where grass cover currently supports fire regimes that keep tree recruitment in check (Bradley and Fleishman 2008). In Utah, grazing did not appear to favor PIED growth or recruitment in persistent PJ woodlands in part because herbaceous understory growth was already limited when grazing was introduced (Barger et al. 2009). The authors suggest that climate is a more important driver of PJ dynamics than land use practices in these types of PJ woodlands. However, heavy grazing has been associated with increases in PIED (Anderson 2002) and Jacobs (2011) makes a strong case for the role of grazing in PJ expansion across the west.

In other assessments, PJ has been described as having low resistance and resilience to improper grazing management (Gori and Bate, 2007; Decker et al. 2017) because PJ systems are highly susceptible to erosion. Grazing reduces herbaceous vegetation cover and soil infiltration rates thus contributing to erosion potentials at a site (Davinport et al. 1998). Grazing can also compact soil and reduce the presence of biological soil crust leading to increased surface runoff and erosion. In particular, biological, or cryptogamic, crusts are an important component of arid land systems that protect soil from erosion, increase water infiltration, and facilitate seed germination (Baymer and Kloipatek 1992; Belnap and Eldridge 2001). Mechanical disturbance (vehicle, livestock, and hiking) and grazing (Baymer and Kloipatek 1992) damages important biological soil crusts (Gottfried et al. 1995; Nielson 2009). Grazing may also facilitate the expansion of nonnative invasive weeds that compete with PJ species and support undesirable fire regimes.

PJ removal treatments/ Timber Harvest/Mechanical disturbance Timber harvest historically had a big impact on PJ woodlands but today harvests are mostly limited to domestic heating and cooking needs. Expansion of piñon juniper woodlands have led to efforts to

reduce PJ overstory in some areas in order to protect sagebrush associated species or improve forage and habitat for rare species, hunted species and livestock (Bombaci and Pejchar 2016). Woodland removal is often implemented by mechanical methods such as chaining or bulldozing or by thinning, prescribed fire, herbicide or some combination of all of these (Bombaci and Pejchar 2016). Seeding is commonly done after tree removal to prevent soil erosion and facilitate forb recovery. In some areas, these practices are increasing in response to perceptions of increased risk of species sensitive to conifer encroachment (e.g. BLM 2011; DOI 2013; Schoennagle and Nelson 2011; Bombaci and Pejchar 2016). Studies identify positive effects of these treatments: increased habitat quality for mule deer (Bender et al. 2013); increased mule deer fawn survival on cleared sites (Bergman et al. 2014); increased water yields and cover of grass and forage plants (Clary et al. 1974; Redmond et al. 2013; Roundy et al. 2014); and, decreased soil erosion in some areas (Davinport et al. 1998).

However, it is not clear that these treatments provide lasting reduction in woody species (Tausch and West ; Anderson 2008; Bombaci and Pejchar 2016) and treatments effects for increasing grass cover (Rippel et al. 1983), water quality, or livestock carrying capacity varies considerably among sites (Clary et al. 1974). In some areas, mechanical disturbance increases nonnative species cover and decreases important biological crusts (Redmond et al. 2015). Redmond et al. (2015) observed that 25 years of tree reduction treatment in Utah’s Grand Staircase-Escalante National Monument resulted in substantial increases in surface fuels and reduced the effectiveness of treatments to reduce wildfire. In some areas, mechanical treatments favor juniper species, which are better able than PIED to reestablish after disturbance (Redmond et al. 2015). Tree removal is sometimes implemented to reduce soil erosion rates but in sites with high site erosion potential, removal of trees to increase herbaceous cover may be insufficient to reduce erosion (Davinport et al. 1998).

Wildlife, particularly piñon and juniper associated species, generally do not respond well to PJ removal treatments. Birds that favor high tree densities respond negatively to tree removal activities. Mechanical, but not fire, treatment of PJ led to decreased bird and mammal species diversity immediately following treatment of sites in northwestern Colorado (Gallo et al. 2017) and these impacts were seen up to 15 years after chain treatment (O’Meara et al. 1977). In general, it appears bird diversity is higher on undisturbed woodlands or within woodland habitats that border treatment areas than within treated areas themselves (Rosenstock and van Riper 2001). Notably, though treatments can increase forage productions, elk and mule deer appear to not favor recently bulldozed sites, presumably due to the loss of cover (Short et al. 1997; Bombaci and Pejchar 2016). In their review of 19 papers, Bombaci and Pejchar (2016) note most species had non-significant or negative responses to mechanical treatments. Importantly, chaining and burning treatments are known to reduced PJ habitat quality compared to historic norms (Colorado SWAP 2015) and are considered a threat to current PJ stands (Gori and Bate, 2007).

Insects and disease Piñon pine are susceptible to numerous fungal agents including the pathogen Lemptographium wageneri var. wageneri, which causes black stain root disease (Kearns and Jacobi 2005), insects including bark beetles (Ips confuses) and twig beetle (Pityophthorus opaculus), piñon needle scale, piñon sawfly (Neodiprion edulicolus), and piñon dwarf mistletoe, (Arceuthobium divaricatum) (Negrón and

Wilson 2003; Zouhar 2001; Anderson 2002). Juniper is affected by insects and pathogens as well including twig beetle (Phloeosinus spp.) twig pruners (Stylox bicolor), wester cedar borer (Trachykele blondeli) and branch-girdling rusts (Gymnosporangium spp.) and juniper mistletoe (Phoradendrong juniperinum)

BOX 5. INTERACTIVE EFFECTS OF BEETLE OUTBREAKS AND DROUGHT

There is a clear link between drought stress and insect attacks (Clark et al. 2016; Gustafson et al. 2015; Gaylord et al. 2015) with episodic drought leading to large-scale beetle-kill in PJ communities (Weisberg et al. 2007). In particular, drought is associated with more intense outbreaks of the Ips bark beetle (Ips confuses) in piñon pine (Breshears et al. 2005; Shaw et al. 2005; Floyd et al. 2009; Clifford et al. 2011; Gustafson et al. 2015). Widespread dieback of piñon pine has been noted during extreme drought periods in the 1950s, 1980s, and early 2000s presumably because drought predisposes stands to beetle related mortality by reducing metabolic defense functions (Falco 2014). Ips beetles typically infest stressed, damaged or dying trees that are unable to produce enough resin or “pitch” to overcome beetles. Drought conditions reduce moisture availability, which limits pitch production effectively reducing tree defense against infestation. In turn, insect defoliators and bark beetle outbreaks may reduce drought tolerance in trees. In addition, during drought trees may become more susceptible to dwarf mistletoe, juniper mistletoe, piñon needle scale, piñon pine sawfly (Zlatnik 1999; Zouhar 2001; Meddens et al. 2014), predisposing them to co-infection by numerous agents.

Pinon Ips beetles (Ips confuses) have a demonstrated negative impact on PIED (Falco 2014). A combination of drought induced stress and Ips beetle infestations has been identified as the primary cause of extensive tree mortality in pinon in the Southern Rockies and Southwest (Box 5; Breshears 2005; Shaw et al. 2005). Recent large scale mortality events across the southwest regions have been associated with a combination of drought, high temperature, and increased tree density all of which increase moisture stress in piñon (Negrón and Wilson 2003; Kleinman et al. 2012). In addition, warmer temperatures increase beetle survival and developmental rates and a longer growing season increases the number of beetle generations leading to more severe outbreaks (Waring et al. 2009; Bentz et al. 2010).

The relationship between tree mortality, insect and drought appears to be mediated in part by stand characteristics including tree age and density (Floyd et al. 2009). Younger or recently disturbed stands sometimes fair better under infections, presumably because there is less canopy leaf area and, therefore, competition for water resources. Older stands (trees > 100yrs) may be more susceptible to blue stain fungus introduced by bark beetles because photosynthesis and therefore production of resin needed to halt spread of fungus is reduced in older trees (Christiansen et al. 1987; Peterman et al. 2012). Tree density appears to be a risk factor for beetle related mortality in PIED stands at upper elevations (Kleinman et al. 2012, Four Corners region). At lower elevations, abiotic factors predominate

and PIED appear at greater risk of Ips damage when associated with southern aspects, faster draining soils, and genetic phenotypes associated with reduced lifetime growth vigor (Ogle et al. 2000, Gitlin et al. 2006, Clifford et al. 2008, Santos and Whitman 2010). Previous studies have also noted potential increased susceptibility for stands in ecotones and in areas where trees are already infected by other agents such as piñon dwarf mistletoe, Arceuthobium divaricatum, (Meddens et al. 2014) and black stain root disease, Leptographium wageneri (see review in Falco 2014).

Invasive plants Cheatgrass (Bromus tectorum), knapweeds (Centaurea spp.), thistles (Cirsium spp. Carduus spp.), Onopordum spp.), mustards (sisymbuim spp. And Brassica spp.), and Dalmatian toadflax are commonly present with PJ ecosystems (Neilson 2009). Piñon-juniper is considered susceptible to adverse effects from exotic/invasive species like cheatgrass and red brome, which are widespread in lower and middle elevation sites in the Colorado Plateau where they support more frequent fire regimes (Gori and Bates 2007; Balch et al. 2013). Cheatgrass tends to invade PJ ecosystems after fire, especially where native forbs and grasses have been depleted (Chambers et al. 2014).

The more frequent and larger fires supported by invasive grass support do not favor PJ establishment and persistence (Board et al. 2018). Grass supported fire is more problematic in areas with warmer drier soils or areas that receive little summer precipitation. Invasive species may also reduce likelihood of recovery after fire. For example, areas burned within Mesa Verde National Park have shown no recovery by piñon or juniper but have been invaded by cheatgrass and smooth brome (Colorado SWAP 2014).

Climate Temperature and precipitation regimes have a large impact on PJ composition and structure (Clifford et al. 2011; Barger et al. 2009). Previous expansions of PJ were primarily driven by wet cool conditions that favored PIED recruitment and growth (Barber et al. 2009). While disturbances influence local patterns of recruitment and die-off in PJ, large scale trends are driven primarily by weather patterns and their interactions with disturbance agents. Specifically, weather conditions determine the degree to which trees are susceptible to pathogens and insects, influence fire behavior, and determine how well PJ trees are able to regenerate following disturbance. Recent assessments and reports have indicated climate as a major threat to PJ ecosystems (San Juan Basin, Rondeau et al. 2017; Colorado, Decker and Fink 2014; Northern New Mexico, NM SWAP 2016, Triepke 2017).

Increased air and soil temperature increases evapotranspiration and soil evaporation (Settele et al. 2014) and causes earlier snowmelt, earlier thaw dates for soil, and changes in timing and amount of streamflow and snowpack (Houle et al. 2012; Romero-Lankao et al. 2014; Diffenbaugh and Giorgi 2012; Williams et al. 2012). In turn, changes in hydrology and soil moisture have substantial impacts on tree growth, plant community composition, and disturbances like fire. Water stress can trigger dieback in juniper and increase susceptibility to insects and pathogens, especially among piñon pine populations growing in the Southwest (Meddens et al. 2015). Moisture stress also increases piñon pine susceptibility to parasitism by the piñon dwarf mistletoe (Arceurthobium divaricatum) (Neilson 2009) and contributes to large scale defoliation by the Piñon tip moth (Dioryctria albovitella), piñon cone moth (Eucosma bobana), piñon need scale (Matsucoccus acalyptus) and piñon needle miner (Coleotechnites edulicola).

Temperature influences tree survival during drought (McDowell et al. 2015). Though both growing season vapor pressure deficit (VPD-driven by temperature) and winter precipitation have strong influences on mortality within Pinus edulis (PIED) (Williams et al. 2013; Allen et al. 2015), VPD has been reported as a more important predictor of drought sensitivity (Williams 2012). VPD increases nonlinearly with hotter temperatures, and especially during drought when VPD cannot be offset by increases in humidity (Allen et al. 2015). Therefore, trees are observed to die faster under hotter temperatures during drought (Allen et al. 2015).

Changes in climate (Box 6), including drought and warmer temperatures are likely to reduce the capacity of PJ woodlands to recover from recent die-offs and eliminate any gains in range experienced over the last century. Mortality in PIED during the 2002-2004 drought was associated with increased growing season temperature and, in particular, with the magnitude change rather than absolute high temperature value (Redmond et al. 2012). Redmond et al. 2012 noted in their study of PJ in New Mexico and Oklahoma that areas that experience the greatest increase in growing season temperatures (high elevation sites) also show the greatest decline (40% reduction) in seed cone production. Interestingly, this may reflect local adaptation where populations in warmer climates were better able to maintain cone production under warming conditions than those in cooler climates.

BOX 6. REVIEW OF CLIMATE PROJECTIONS FOR SW U.S.  Increase in mean annual temperature, warming in all seasons (Diffenbaugh and Giorgi 2012; Romero-Lankao and others 2014)  Increase in occurrence of extremely hot seasons, warmer summers (Diffenbaugh and Giorgi 2012; Romero-Lankao and others 2014)  Decrease in precipitation for some areas, particularly winter precipitation for southwest region (Seager and others 2007; Seager and Vecchi 2010)  Increase in drought frequency and warm, dry summers (Drake and others 2005; Sheffield and Wood 2008; Allen and others 2010; Gutzler and Robbins 2011; Romero- Lankao and others 2014)  Decrease snowfall and snowpack; winter precipitation falling as rain and not snow (Diffenbaugh and Giorgi 2012)  Shift in precipitation events and amounts (Doeskin and others 2003)

Drought Periodic droughts, common in the southwest, represent a major disturbance (Weiss et al. 2009) and an important factor mediating woodland dynamics (Shinneman and Baker 2009; Clifford et al. 2011). Drought-induced declines in PJ arise where there is a lack of precipitation or increased evaporative demand associated with higher temperatures (Williams et al. 2012). Severe multi-year droughts in the western U.S. have been related to widespread tree mortality, especially piñon, in PJ systems (Clark et al. 2016). Drought affects PJ stand structure and distribution through several mechanisms: increased mortality and dieback due to moisture stress; reduced tree growth and recruitment (in PIED, Ronco 1990; Breshears et al. 2005); and increased tree susceptibility to beetle and fungus outbreaks (Box 5;

Breshears et al. 2005; Gustafson et al. 2015). Specific studies in the FC and URG focal areas note the following:  Widespread stand replacement in PJ has been particularly notable during hotter drought periods (Shaw et al. 2005; Cole et al. 2008; Clark et al. 2016; Allen et al. 2015).  Recent drought related mortality was greatest for PJ stands in the Four Corner states, specifically central Arizona, New Mexico, and southwestern Colorado (Shaw et al. 2005).  Mortality in the Colorado Plateau (Coconino National Forest) due to the 2002-2004 drought exceeded any gains in canopy cover from the preceding decades and drought was also associated with a much greater proportion of tree loss over fire or land use (e.g. grazing)(Clifford et al. 2011).

Drought impacts are ubiquitous and often exacerbate impacts from other disturbances (Box 5; Breshears and others 2005). Changes in hydrology and soil moisture also have substantial impacts to disturbances like fire with consequences for PJ persistence.

Response of Piñon and Juniper to Drought Drought influences PJ stand structure when more tolerant species outcompete weaker species (Clark et al. 2016). Piñon appear less tolerant to drought than juniper species. Under wetter conditions, piñon has higher photosynthetic rates than juniper and a competitive advantage. Multi-year periods of cool-wet climate favors PEID recruitment and winter and early spring precipitation has the strongest influence on PEID growth (Barger et al. 2009). However, piñon pine experiences reduced photosynthesis more quickly than juniper under water stress conditions. Water scarcity can kill trees in two ways: 1) cavitation, or the collapse of xylem, associated with hydraulic failure as a result of water scarcity, and 2) carbon starvation as trees close off stomata (and reduce photosynthesis) to reduce water loss and cavitation (Meddens et al, 2015; Gaylord et al. 2015). Piñon pine are isohydric, which means they reduce stomatal conductance but not leaf water potential under dry conditions. This leads to carbon starvation under conditions of moisture stress via the second mechanism noted above. Piñon pine appears relatively resilient to short drought periods but experiences greater mortality when drought exceeds eight months because trees become more susceptible to damage by Ips beetle infestations and fungi due to reduced carbon uptake and, thereby, reduced metabolically derived defense functions (Breshears et al. 2009; Meddens et al, 2015; Koepke et al. 2010; Peterman et al. 2012). Recent research has indicated that piñon pines also experience cavitation under prolonged drought (McDowell et al. 2008; Gaylord et al. 2015). Juniper are anisohydric and reduce leaf water potential but not stomatal conductance in sync with reduced soil moisture. Under prolonged drought, this leads to cavitation and tree death. Over shorter time frames, anisohydric trees are more tolerant than isohydric trees, which begin to reduce photosynthesis and carbon uptake under relatively mild drought conditions (Sade et al. 2012). Even under prolonged drought, juniper tends to experience cavitation on individual shoots and thus dieback rather that mortality due to hydraulic failure within a single main stem (Koepke et al. 2010). This tendency for dieback can be seen as an adaptation to drought and ultimately improves water and nutrient availability to surviving tree tissue. Juniper species also tend to have better recruitment following drought (Redmond et al. 2015) and greater seed longevity that allows it to reestablish more quickly following disturbance (Clifford et al. 2011; Table 1.3).

Table 1.3. Characteristics of piñon and juniper tree species. Data compiled from Gori and Bate 2007 and Nielson 2009. Characteristic Piñon Juniper Mast 3 years 2-5 years (alligator juniper can resprout)

Seed dispersal Cached by birds Fruit dispersed by birds

Seed survival 1 year 3 years

Establishment Favored by nurse plants Shade intolerant, prefers open areas

Time to establish Several decades; grows more Several decades quickly than juniper under moist conditions

Moisture stress Anhydrous; drought causes carbon Isohydrous; experiences cavitation starvation and reduces metabolic in response to prolonged drought; defense against insects. Cavitation, cavitation occurs mainly in terminal when it occurs affects main stem branches causing dieback branch resulting in tree death

Influence of Site Characteristics Vegetation response to drought is influenced by local soils, topography, and elevation (Herrmann et al, 2016; Allen and Breshears 1998) where site characteristics effect water availability during drought (Meddens et al. 2015). Soil texture and depth directly influence soil water content and water availability for trees (Meddens et al. 2015). Drought related mortality has been related to lower soil water storage capacity (Peterman et al. 2012). Piñon regeneration was greatest where soil available water capacity was highest in sites in both Colorado (Redmond and Barger, 2013) and Arizona (Redmond et al. 2015). However, not all studies of piñon mortality events have found a relationship between soil properties and the severity of drought related mortality events (e.g. Koepke et al. 2010; Clifford et al. 2013).

Temperature varies with elevation and it is generally thought that trees at lower elevations will be exposed to greater overall temperature increase (i.e. be hotter), whereas trees at upper elevations are likely to experience a greater magnitude change from current temperatures. As discussed earlier, hotter temperatures, particularly in the spring do not favor PIED growth, recruitment or survival during drought. However, a recent review failed to find consensus between mortality events and elevation (Meddens et al. 2015). Clifford et al. (2011) note greater drought mortality in low elevation stands in Arizona. Other studies covering our focal area were unable to relate elevation to recruitment (Redmond et al. 2015) or mortality (Clifford et al. 2013) of PIED in New Mexico and Arizona, respectively. Though drought stress may be greater at lower elevations, issues related to beetles, fire, and competitive impacts are greater at higher elevation sites (Meddens et al. 2015). Mid-elevation sites appeared to be buffered the best from either disturbance. Further, upper elevation sites have shown an increase in

productivity under dry warm conditions whereas low elevation sites appear to show declines in productivity (as measure through NDVI) (Herrmann et al. 2016). These observations are driven by a delayed onset of moisture stress at upper elevations and an increase in upper elevation growing season under warmer conditions. However, this benefit is likely to be short-lived as decreasing moisture supplies eventually affect higher elevation locations. Similarly, local benefits relating to decreased snow cover in some areas (that allow tree establishment) are likely to be offset with far greater declines in trees downhill, where lack of runoff effectively reduces water availability (Trujillo et al. 2012; Herrmann et al. 2016). Ultimately, local gains in vegetation productivity in warm drought years are likely to be overshadowed by the larger losses at the regional scale.

Influence of Stand Density/Demographics At local scales, drought impacts can vary depending on structure and density of vegetation (Peterman et al. 2012) though debate exists around specific stand characteristics that influence dieback events (Floyd et al. 2009; Klienman et al. 2012). During the recent droughts that preceded widespread mortalities in piñon pine in the Southwest, mortality was general associated with larger trees (Meddens et al. 2015; Redmond et al. 2015). At sites with lower mortality, mostly younger trees were killed (Looney et al, 2012), whereas in dense stands that experienced high mortality, older trees were killed. Floyd et al. (2009) reported the greatest mortality in older, reproductive piñon trees in their tristate analysis of stand die-offs during the regional drought period that began in 1996. Importantly, juniper mortality has been much lower than piñon mortality under this prolonged drought period. Meddens et al. (2015), note 3 primary mechanisms for the trend of increased mortality in older trees: 1) though presumed to have a competitive advantage (due to greater access to water and greater storage capacity), older trees also have higher metabolic demands and vulnerability to hydraulic failure due to higher leaf area, structure and reproductive maturity; 2) bark beetles like older trees; and 3) generally shallow roots of piñon pines.

Conflicting evidence exists for the relationship of stand density and drought related mortality (Peterman et al. 2012). Above average stand densities increase competition for water and have been considered a factor that intensified piñon susceptibility to Ips attack and increased PIED die-offs (Negrón and Wilson 2003 (Arizona); Greenwood and Wiesberg 2008 (Nevada); Weisberg et al. 2007 (Nevada); Klienman et al. 2012 (Southwest)). However, others have found either a negative relationship between tree density and tree die-offs of PIED (central New Mexico, Clifford et al. 2013) or no relationship (New Mexico, Arizona and southwestern Colorado, Floyd et al. 2009; Arizona, Clifford et al. 2011). In contrast, Negron and Wilson (2009) report a strong positive relationship between tree density and piñon mortality. Floyd et al. (2009) suggest the extreme severity of the 2002 drought cycle likely masked density dependent relationships in PIED stands in their study (Floyd et al. 2009). Current Status Variation in individual species’ tolerance to drought, insect and fire has influenced the status of PJ in many areas within its range, with piñon species typically considered at greater risk than juniper. To date, no comprehensive assessment exists that describes the status of PJ across its range. However, State Wildlife Actions Plans (SWAP) are available that assess the current status of wildlife habitat as part of a process to develop management practices for species of greatest conservation need. Arizona, Colorado,

and New Mexico consider PJ woodlands important habitat for multiple species and, therefore, describe its current status within state boundaries. In Arizona, PJ (i.e. Great Basin Conifer Woodlands) is considered significantly affected by changes in fire regime, livestock grazing, and mechanical and chemical treatments (AZ SWAP 2017). Additionally, increased density of trees and invasive Bromus has increased the likelihood of crown fires to the detriment of some PJ habitats (Tausch 1999; Gruell 1999 in AZ SWAP 2017). In Colorado, most PJ is considered in fair to good condition with some excellent conditions in more remote and undisturbed areas (CO SWAP 2014). PJ in poor condition is associated with areas of intensive grazing that has reduced native bunch grasses and encouraged establishment of invasive cheatgrass. In New Mexico, PJ woodlands have expanded into upper elevation Ponderosa Pine forest and lower elevation grasslands but are considered at risk of widespread declines under climate changes that increase drought conditions and insect and fire disturbances (NM SWAP 2016). In Utah, PJ was not considered an important habitat and, given PJ encroachment into lower elevation sagebrush and grassland communities, is generally the target of removal strategies (UT SWAP 2016).

Expansion Increased abundance and density of woody plants into what were formerly grassland and shrublands is a major concern in some areas (e.g. Coates et al. 2017; Carrol et al. 2016) and has led to many years of treatments to reduce piñon and juniper trees. Increases in PJ manifest as expansion where trees colonize new sites or as increases in tree density due to infill of existing stands. Expansion may be driven by natural or human derived factors (Box 6). Increased PJ has been attributed to recovery from disturbance, natural range expansion, livestock grazing, climatic variability, elevated CO2 and fire suppression (Neely et al. 2001; Romme et al. 2009; Swetnam et al. 2016). PJ expansion and increased tree density are associated with declines in understory cover of shrubs and native perennial grasses, which can leave landscapes prone to changes in wildfires, soil stability, and species composition (Carroll et al. 2016, Jacobs and Gatewood 1997; Jacobs 2015). Accelerated rates of soil erosion have been attributed to increases in woody species, specifically where such increases have led to decreased ground cover (Davenport et al. 1988). Hydrological changes are also present where deeply rooted piñon and juniper vegetation can reduce available subsurface water (Carroll et al. 2016, though see Madsen et al. 2008). However, there is increasing push back on the notion that PJ ecosystems are undesirable or uncontrollably overrunning other vegetation communities. For instance, soil loss and erosion associated with increased PJ are most commonly found within transitional habitats (e.g. recent conversion to PJ from some other ecosystem type), which are by their nature less stable and, therefore, prone to further degradation (Davinport et al. 1998; Jacobs 2015). Recent studies have also shown that infiltration of runoff is greater under PJ canopies than areas outside PJ canopy (Madsen et al. 2008) and soil moisture is typically greater under canopy versus intercanopy locations (Davinport et al. 1998).

Historic trends show substantial expansion of BOX 6. ENCROACHMENT VS. SUCCESSION PJ into many shrublands in the Great Basin Debate exists on the nature of PJ expansion in the West. (Neely et al. 2001): PJ is estimated to have Where drivers of expansion are mainly human related increased by 20% in the Colorado Plateau (e.g. grazing, fire suppression), it is generally considered since 1850 (Tuhy et al. 2002) and by 10% in encroachment. If, on the other hand, drivers are mainly the lower southern Rockies (Neely et al. natural (e.g. climate), PJ expansion may be considered part of a natural succession process (Clifford et al. 2011). 2001). In general, the rate of woody species For example, the establishment of transitional woodlands encroachment (all types) appears to have in what was historically ponderosa pine dominated been slower within the Southern Rocky savannas and persistent woodlands in Northern New Mountains (0.5 % change/year) than in the Mexico was facilitated by tree die-offs in the 1950s’ due Southern and Central Great Plains (~0.75 and to beetle related mortality followed by moist conditions 1.5 % increase/yr, respectively) (Barger et al. (Jacobs 2015). 2011). Within existing ranges, many PJ woodlands have experienced infilling, increased tree density, and increased canopy cover (Romme et al. 2008; Carroll et al. 2016). Piñon juniper savannahs have experienced infill though it is unclear whether this is natural expansion or recovery of persistent woodlands in areas that have been disturbed (Box 6). Tree density increases in wooded shrublands have been most pronounced in the Great Basin shrublands and in parts of Arizona and New Mexico but are much more limited in Western Colorado (Weisberg et al. 2006; Romme et al. 2008).

To some degree, infilling and increased tree density may be overestimated. Archer et al. (1996) note that research to measure woody species encroachment typically happens where such encroachment is known to have occurred thus potentially biasing results. Bias may also result where studies using remote sensing overestimate PJ cover. For instance, one study estimated an average rate of PJ increase of 11% over 30 years (Weisberg et al. 2007). However, this rate was based on an analysis that included a misclassification error of nearly 35% for PJ (1 in 3 sites classified as PJ were not PJ). This level of error suggests the possibility that the analysis overestimated actual PJ presence and, thus, expansion. Still, numerous studies employing a variety of techniques agree that woody species including pinon and juniper have increased and expanded into many shrublands and grasslands in the Western U.S.

Declines Recent trends show significant declines in PJ and, in particular, piñon trees across the Southwest. Marked contractions of PJ have been noted for areas in Colorado, New Mexico, and Arizona (Manier et al. 2005; Shaw et al. 2005; Clifford et al. 2011; Arendt et al. 2013; Arendt and Baker 2013). A combination of drought induced stress and piñon Ips bark beetle (Ips confuses) infestations has been identified as causing extensive tree mortality in PJ ecosystems (Breshears 2005). Mortality in the Colorado Plateau (Coconino National Forest) due to the 2002-2004 drought exceeded any gains in canopy cover from the preceding decades (Clifford et al. 2011). Mortality rates of piñon has been particular high (30-65%) in the Four Corners region of the Southwest (Shaw et al. 2005). Shorter fire

return intervals as a result of cheatgrass invasion were considered the primary reason for piñon and juniper declines in the Colorado Plateau (Arendt and Baker 2013). It is possible that observed contractions are related to increased density as a result of infill, which has left piñon stands more prone to drought related die-offs (Floyd et al. 2015). However, recent evidence does not support an entirely density dependent mechanism. Meddens et al. (2014) reviewed several studies of piñon die-offs within the SW and failed to find a consistent relationship between higher mortality and stem density. Many studies point towards a continued trend for reduced PJ within the study area especially under warmer and dryer climate conditions (see next section). Additionally, shifts to more juniper dominated systems are possible and juniper expansion has been noted in grasslands in the Colorado plateau (Millar and Rose 1995).

Future Status Declines or increases in climates suitable for PJ depend on both future precipitation and future temperature regimes (Rehfeldt et al. 2006; Keane et al. 2008; Romme et al. 2008). Climate has direct implications for piñon and juniper survival and regeneration and indirect implications via its influence on fire regimes and beetle attacks. Moisture availability and temperature influence the biogeographic and physiographic range of PJ habitats. Piñon-juniper ecosystems are found in areas with a mean temperature range 40-61⁰C and 7-25 in of rain (Gori and Bate 2007; Ronco 1990). Historically, piñon- juniper woodlands moved up and down environmental and geographical gradients in response to changes in precipitation regimes (Kyllo 2016). Periods of water surplus encourages expansion and wet climatic conditions have been association with increased PJ seed recruitment and growth (Neilson

2009). Warming, increased precipitation, and increased CO2 may encourage further expansion of PJ woodlands (Miller and Wigand 1994; Miller and others 2008; Allen et al. 2015). Additionally, earlier spring greening and flowering, longer growing season may increase productivity within these ecosystems (Settele et al. 2014). Some point out that conditions that favor expansion of these woodlands will also have the negative impact of encouraging cheatgrass recruitment at lower elevation sites (Ziska et al. 2005; Poiani et al. 2010).

Alternatively, widespread loss of trees is predicted under warmer and dryer conditions (Allen and Breshears 1998; McDowell et el. 2008; Williams et al. 2012; Anderegg et al. 2013; Allen et al. 2015; Clark et al. 2016) and climate models generally project warmer and more arid conditions in the SW (Seager et al. 2007; Williams et al. 2012; Box 6). Though precipitation projections are less certain, models indicate that without significant increases in precipitation that can compensate for increased water usage associated with higher temperatures, there is likely to be a net reduction to surface water, soil moisture and water availability (Dai 2013). In turn, this influences how woodland habitats respond to changes in climate. Climate projections for warmer and more arid conditions in the SW form the basis of a number of predictions for declining PJ under the interaction of warmer temperatures, drought, insect outbreaks and fire disturbance. Studies that regard species and locations relevant to the current assessment make the following conclusions:  An increase in hotter droughts will continue to reduce tree densities and survivorship (Allen et al. 2015).  Climate-related changes to fire regimes will negatively impact PJ in Colorado (CO SWAP 2014).

 Colorado piñon pine are predicted to experience the greatest declines under warmer and dryer conditions (Thompson et al. 1998; Cole et al. 2008).  Drought and heat induced mortality is most likely for trees in areas with lower than normal winter precipitation particularly when followed by a dry, hot, and long growing season (Settele et al. 2014).  Declines in PJ woodlands will result from declining piñon recruitment following loss of overstory trees due to drought (Floyd et al. 2015).  Increased temperature will likely reduce seed cone production (Ronco 1990; Redmond et al. 2012; Clark et al. 2016). Importantly, these trends may be more dramatic at upper elevations where climate is expected to change to a larger degree (Giorgi et al. 1997) thus reducing potential high elevation refugia (Pearson et al. 2002).  Future conditions that include altered climate and fire regimes may not allow the natural ebb and flow of conifer density (infill and mortality) that have given rise to important old growth piñon juniper forests (Floyd et al. 2015).

BOX 7. CONSEQUENCES OF CLIMATE RELATED DECLINES IN PINON-JUNIPER HABITATS

Declines in PJ could hold short term benefits for browsers where sufficient understory vegetation is present. However, loss of trees and/or conversion to grass/shrub due to increasing drought and fire will reduce food, cover and nest site availability for a wide range of PJ obligate species. In particular, declines of PJ will result in loss of food (juniper berries and pine seeds) and breeding/nesting sites for small mammals (chipmunks, jack rabbits, squirrels, woodrats) (Zlatnik 1999; Zouhar 2001; Anderson 2002), ferruginous hawks (Holechek 1981; Bosworth 2003), and birds including gray flycatchers, which are already showing population decline (Sauer et al. 2008). Commensal relationship between piñon and seed eaters are likely to accelerate declines as Scrub jay, piñon jay, Steller’s jay, and Clark’s nutcracker caches are important for PIED regeneration (Evans 1988; Hall and Balda 1988; Ronco 1990; Zouhar 2001). Declines in PJ would also be detrimental to obligate species such as piñon mouse, woodrats (Stephen’s woodrat), piñon jay, screech owl, scrub jay, plain titmouse, and gray vireo (Short and McCulloch 1977; Balda and Masters 1980; Meeuwig et al. 1990; Morrison and Hall 1999; Bosworth 2003) – some of which are important prey populations for large mammals and raptors (Zouhar 2001). Reduced PJ could also result in loss of resources for insects like, for example, bee species that use piñon pitch for building nests (Lanner 1981).

Recent efforts to model climate suitability for PJ in the West also provide clues as to how well these ecosystems will persist into the future. Cole et al. 2008, using historical migration patterns to estimate PIED distribution changes in response to climate changes and related influences on fire and beetle infestations, show increasingly poor conditions for PIED in Arizona and Utah, but some expansion to higher elevations in Colorado and New Mexico. However, realized expansion rates of piñon means that only small portions of these new areas are likely to be populated by PIED in the next 100 years. Clark et al. (2016) note there is limited evidence that species are migrating in response to climate changes and cast doubt on the idea that species will able to keep up with the velocity of climate change. Still, Clark et al. (2016) expect both positive and negative changes throughout the range of PJ with impacts to growth and recruitment within as well as along trailing edges of its range. McDowell et al. (2016) used experimentally derived thresholds of piñon and juniper tree leaf water potential and soil water potential

to modeled response under business as usual greenhouse gas emissions (RCP8.5) and found evidence for widespread die-offs by 2100. Mortality events were greatest for areas at the southern extent of the SRLCC and, importantly, their models indicate high susceptibility of juniper as well as piñon. Rehfeldt et al. (2012) projections for individual species shows widespread declines for all piñon and juniper species (Figs. 1.3 and 1.4) under CMIP3 generated climate scenarios. Similarly, Allen et al. 2015 predict declines in both piñon and juniper under CMIP5 scenarios and estimate a near 100% mortality in SW forests by year 2100.

Figure 1.3 (preceding page). Change in three juniper species within the SRLCC boundary under three GCMs (CGCM3, GFDLCM21, and HADCM3) and four SRES emission scenarios (A1B, A2, B1, B2) for three time periods. For each time period, we show the consensus (>2/3 of models) of projections for suitable climate as compared to current time periods (year 2005 baseline). Black areas indicate no change from current distribution, blue indicates loss from current, and red indicates potential range expansion.

Figure 1.4. Change in two piñon species within SRLCC boundary under three GCMs (CGCM3, GFDLCM21, and HADCM3) and four SRES emission scenarios (A1B, A2, B1, B2) for three time periods. Processing is as described for Fig 1.3.

For both figures, Data downloaded from http://charcoal.cnre.vt.edu/climate/species/index.php. These maps as well as additional maps of individual climate model results can be found with other project documents at: https://www.sciencebase.gov/catalog/item/569414a2e4b039675d00472e

II. Vulnerability Assessment Method and Approach Background As commonly applied to climate change issues, VA provide a structure for organizing complex information and addressing uncertainty (IPCC 2007). Although there are various definitions, vulnerability is generally thought of as the susceptibility of a target to negative impacts from some disturbance (Füssel 2007, Hinkel 2011). Assessment of climate change vulnerability typically considers three elements: exposure, sensitivity, and adaptive capacity (Glick et al. 2011). Exposure is the magnitude of climate and climate-related phenomena (e.g., fire, floods) whereas sensitivity (i.e., response to exposure) and adaptive capacity (i.e., ability to cope with negative impact) are traits or conditions that predict how a target will respond to that disturbance. These definitions can vary according the goals and the target of an analysis. For instance, sensitivity may represent the innate traits or qualities of a target that increase the likelihood it will experience a negative response. Alternatively, sensitivity may represent the potential cost of a disturbance (e.g. watershed values - Furniss et al. 2013). Adaptive capacity can be identified as the intrinsic and/or externally driven mechanisms that represent the potential for a target or system to withstand a disturbance.

Table 2.1. Framework for assessing vulnerability of focal resources in the Southern Rockies Landscape Conservation Design Element Definition Examples Indicators Exposure External threat to the • Human Impacts • Urbanization target species, system, • Natural disturbances • Wildfire potential or place • Climate change • Magnitude departure in temperature

Sensitivity Qualities that make the • Traits associated with • Narrow physiological threshold target more susceptible increased negative • Current departure from to negative impacts response reference condition from disturbance or • Indicators of potential • Presence of T&E species threat cost of disturbance • High value watersheds

Adaptive Capacity The ability of the target • Traits/conditions • Wide physiological tolerance to cope with associated with resilience • Diverse prey base disturbance or threat • Potential for successful • Capacity to implement management intervention conservation action (e.g. land ownership profile) • Effective management options available (e.g. thinning can alter wildfire outcome) Many species-specific assessments define adaptive capacity through the identification of intrinsic traits, whereas landscape assessments often include externally driven sources of adaptive capacity such as landscape context and potential for management intervention. For the SRLCC assessments we specify

definitions and criteria that are inclusive and adaptable to multiple scales of assessment and uses (Table 5).

Structure Vulnerability assessments take a wide range of forms and approaches (Glick et al. 2011). The most effective vulnerability assessments are tailored to address specific objectives of resource managers or others who will use the information for management decisions (Friggens et al. 2013). For the assessments of focal resources with the Southern Rockies Landscape Conservation Cooperative, we identified a Vulnerability Assessment (VA) framework that could be used to identify relevant information and assign it to the appropriate measure of vulnerability. The VA framework contains an inclusive list of potential measure of vulnerability that relates to both species specific and landscape considerations. We use this framework to identify datasets and analyses that could inform our assessment as an indicator of one of the vulnerability elements. Once we compiled relevant and meaningful indicators, we estimate vulnerability as the collective impact of exposure and sensitivity weighted against adaptive capacity (Fig. 2.1). Vulnerability is then visualized by comparing the impact scores with adaptive capacity scores using a matrix (Table 2.2). This system provides considerable flexibility so that assessments can identify vulnerability across diverse focal resources and be quickly tailored to user needs.

Figure 2.1. Structure underlying the estimation of vulnerability; exposure and sensitivity collectively represent impact to a resource of interest.

Adaptive capacity modulates this impact resulting

in more or less vulnerability. In the system used in this assessment, we create scores representing the cumulative impact of disturbances and sensitivities (impact) and adaptive capacity and then compare relative impact and adaptive capacity values to generate vulnerability classes.

Measuring Vulnerability The framework identified in Table 2.1 and Fig 2.1, encompasses the overarching structure to the assessment process. Several steps are involved in the use of this framework to generate vulnerability assessment projects. 1) We identified relevant data that relate to potential threats or issues, state of the focal resource, and traits or conditions that influence how that resource will respond to disturbance. We call these data indicators. For each focal area, we considered a diverse set of data and analysis and selected those that had some capacity to represent the potential disturbance or response of a resource to a disturbance. We collected some information on potential stressors and threats during conversations during Adaptation Forums. Additional threats or stressors were identified from other assessments or primary literature. 2) For each indicator, we determine whether it most appropriately measures exposure, sensitivity or adaptive capacity. Depending on the focal resource, some indicators maybe used to measure more than one vulnerability element. For instance, road density, a common measure of human

disturbance and activity might be considered under exposure. Alternatively, roads can also represent a barrier to movement and contribute to a focal resources’ sensitivity to disturbance. The assignment of a particular dataset to a particular element was made based on the relationship of the focal resource to that data and where it was determined the measure would provide the most meaningful output. 3) Once identified and assigned to a vulnerability element, a threshold of effect was determined for each indicator. This threshold was the cutoff value based on the original range of data values that would determine whether an area was considered effected or not. For exposure values that represented a meaningful impact were given a score of 1. Similarly, values of data within datasets contributing to sensitivity were assigned a value of 1 where they were considered to represent a condition of increased potential negative response or cost. Adaptive capacity represents resilience and data values that could be inferred to represent greater resilience were assigned a 1. For each element, scores were added to create cumulative indices representing Exposure, Sensitivity, and Adaptive Capacity. Exposure and Sensitivity scores were combined to create an impact score and this was compared to Adaptive Capacity to generate vulnerability scores (Table 2.2).

Table 2.2. Matrix of Impact versus Adaptive Capacity Scores. Indicators are summed to give total scores for Exposure (E), Sensitivity (S), and Adaptive Capacity (AC). Exposure and Sensitivity are added together to represent impact and all values are rescaled to a 1-5 range. Impact increases as values increase along the horizontal and Adaptive Capacity increases as values increase along the vertical. Vulnerability is determined by the relative Impact versus Adaptive Capacity of the Focal Resource according to these potential combinations.

Vulnerability Impact (Exposure + Sensitivity) Value 1 2 3 4 5 Adaptive 1 Low Intermediate High Very High Highest Capacity 2 Low Intermediate High Very High Very High Value 3 Very Low Low Intermediate High Very High 4 Very Low Very Low Intermediate High High 5 Lowest Very Low Intermediate Intermediate High Spatial Units This assessment is focused on two geographic areas: the Four Corners region containing portions of AZ, NM, UT and CO and the Upper Rio Grande region covering the Rio Grande headwaters in Colorado and Middle Rio Grande Valley in New Mexico (Fig. 1.1). All spatial data has been processed and summarized at the HUC12 subwatershed scale.

Indicator Variables We selected indicator variables based on results of our literature review, variables used in previous assessments of PJ ecosystems, and limiting factors discussed by managers at the 2017 Adaptation Forum

workshops (Appendix 2; Tables 2.3; Figs. 2.2-2.4). All data and associated webinar and workshop information is available online (https://www.sciencebase.gov/catalog/item/5693e56ee4b09c7f9a21a41d) to facilitate additional analyses and assessments identified by users.

We focused on spatial datasets covering the full extent of the study region. Exposure considers the proximity or percent area of known sources of disturbance or threats as well as the degree to which a watershed might be affected by a disturbance agent. Measures include wildfire hazard, infect and disease risk, change in critical climate variables, presence of energy development, and road density (Fig. 2.2, Table 2.3). Sensitivity considers the presence of disturbances, associated species richness, current cover of PJ, and soil available water storage capacity (Fig. 2.3, Table 2.3). Those areas with the highest value are considered most sensitive to disturbances that may cause declines in PJ or PJ related attributes. Adaptive capacity considers measures that indicate increased coping capacity (Fig. 2.4, Table 2.3). We include potential management capacity (conservation potential), low fire risk, increased climate suitability, and current cover of PJ.

To calculate vulnerability scores from individual indicators, we transformed each indicator into a 1/0 (presence/absence) score and individual indicators were then summed for each watershed to give cumulative scores for exposures, sensitivity, and adaptive capacity. Impact was calculated by adding together sensitivity and exposure scores and a rescaled (1-10) value was then compared to similarly scaled adaptive capacity scores to assign a final vulnerability classification (Table 2.2). This simply additive approach has been found to perform well in vulnerability assessments by minimizing user bias and increasing flexibility. However, it also treats each indicator as equally important for determining individual vulnerability classes. As well, since each element of vulnerability (Exposure, Sensitivity, and Adaptive Capacity) had unique sets and numbers of indicators and resulting values are then transformed to a single scale, the relative impact of individual indicators vary (see discussion for more information).

Table 2.3. Indicator variables used to measure vulnerability for Piñon-Juniper Ecosystems in current assessment. Data sources, justification and further information available in Appendix 2. Name Description How Used % FC % URG Exposure Mean maximum temperature in >1.8 degrees C° Max Summer Temperature 82 19 the warmest month in 2030 = 1

Percent of total normal winter <100 percent = Change in Winter Precipitation 82 31 precipitation (Nov-Feb) in 2030 1

Change in Development Change in Development, 2040 Increase = 1 46 42 >25th Road Density Road Density 16 38 percentile = 1 Percent HUC with High to Very High or Very Risk of Stand Replacing Wildfire 20 26 High Fire Potential High = 1 Percent HUC with High Risk of Loss due to Insects and Disease >0 = 1 34 39 Insect and Disease Percent Cover Suitable Climate for Loss due to Climate Change >30 percent = 1 24 3 Great Basin Conifer Woodlands Sensitivity Presence of Energy (oil & gas Energy Activities Present = 1 20 6 wells, solar) Presence of Medium or High Medium or High Intensity Development 38 36 Intensity Development High = 1 Presence of Mechanically Current Disturbance Present = 1 9 20 Disturbed Forest Low Cover of Piñon-Juniper Percent Cover of Piñon-Juniper <30 percent = 1 48 64 Ecosystems Woodlands/Savannas Wildlife Diversity Gap Species Data, n = 17 >12 species = 1 6 49 Soil Available Water Low Available Water >0 km2 = 1 28 9 Storage Capacity Storage Capacity (<0.10)

Adaptive Capacity Increase in climates suitable for Area Change in Suitable Climate >30 percent = 1 31 23 PJ for Great Basin Conifer Woodland Lands designated as protected by Percent HUC area that is >30 percent = 1 37 51 GAP analysis (=1 or 2) Protected High Cover of Piñon-Juniper Percent Cover of Piñon-Juniper >60 percent = 1 14 8 Ecosystems Woodlands/Savannas Areas with Low to Very Low Fire Percent Low to Very Low Fire Low or Very 8 5 Potential Potential Low = 1

Figure 2.2. HUC12 within two focal areas that have increased exposure (blue). Criteria for determination is indicated in legend.

Figure 2.3. HUC12 within two focal areas that have increased sensitivity (blue). Criteria for determination is indicated in legend.

Figure 2.4. HUC12 within two focal areas that have increased adaptive capacity (blue). Criteria for determination is indicated in legend.

Results Exposure (Table 2.4, Fig. 2.5) The Upper Rio Grande contained a greater proportion of HUC12 with low exposure scores. Areas around Trinidad, Colorado and the foothills of the Sangre de Cristo Mountains have high exposure scores. Further north a few watersheds draining into Wet Mountain and San Luis valleys appear at high risk. To the south, the highest scoring watersheds were found around Thoreau, NM and areas around Mount Taylor and the foothills of the Zuni Mountains. Areas around the Jicarilla Apache Nation Reservation and nearby Nacimiento, Jemez, and San Pedro mountains in the Santa Fe National Forest, the Rio Grande del Norte Monument, and watersheds within the Cibola National Forest and between Santa Rosa and Moriarity, New Mexico appear to be at low risk of exposure to stressors indicated in this analysis.

Table 2.4. Percent HUC12 within each Exposure category. % FC % URG None 1.1 8.8 Minimal 5.7 25.8 Low 58.7 58.6 Moderate 26.0 3.9 High 7.6 2.8 Very High 0.9 0.0

Four Corners generally had more expected high exposure and held the only HUC12 that received a very high exposure score in this analysis. Areas with high exposure scores included areas around Gallup, New Mexico, Flagstaff, Arizona, and Black Mesa, Red Rock Valley and Shonto Plateau all within the Navajo Nation Reservation in Arizona. Three watersheds associated with the Sitgreaves National Forest around Springerville, AZ also had all exposure variables. Areas within the Jicarilla Apache Nation Reservation and Carson National Forest had relatively low exposure.

Figure 2.5. Exposure Scores for HUC12 within the Four Corners (FC) and Upper Rio Grande (URG) Focal Areas.

Sensitivity (Table 2.5, Fig. 2.6) Within the upper Rio Grande focal area, watersheds in the northern extent typically had higher sensitivity scores than those to the south. The highest sensitivity was found for watersheds on the eastern slope of the Jemez Mountains. Many of the watersheds to the south and near El Malpais National Monument and the Zuni Mountains received a score of 0 indicating no presence of sensitivity factors.

Within Four Corners, the highest scoring watersheds were located on the Jicarilla Apache Nation Reservation and near the city of Farmington in New Mexico. Numerous watersheds received a score of 0 indicating no relevant sensitivity measures.

Table 2.5. Percent HUC12 within each Sensitivity category. % FC % URG None 19.4 8.1 Minimal 33.8 31.8 Low 28.3 33.9 Moderate 15.5 19.8 High 2.9 6.1 Very High 0.1 0.2

Figure 2.6. Sensitivity Scores for HUC12 within the Four Corners (FC) and Upper Rio Grande (URG) Focal Areas.

Adaptive Capacity (Table 2.6, Fig. 2.7) The majority of watersheds within the upper Rio Grande had no or a single adaptive capacity factor. A watershed on Glorieta Mesa (south of Glorieta), New Mexico, an area near Petaca Mesa in Carson National Forests, and a couple of watersheds within the El Malpais National Conservation area received the highest scores.

Table 2.6. Percent HUC12 within each Adaptive Capacity The Four Corners showed similar category. patterns with the majority of % FC % URG watersheds receiving a “0” or very low None 28.6 31.8 adaptive capacity score. Watersheds Minimal 55.0 50.1 showing high adaptive capacity were Low 14.7 17.1 northeast of Chaco Culture National Moderate 0.0 0.0 Historical Park in Arizona, an isolated High 1.6 1.0 watershed west of Gallup, New Mexico Very High 0.0 0.0 and another south of Quemado, Arizona, watersheds within an area northeast of Show Low, Arizona and a watershed at the western extent of our study area near Grand Canyon Village, Arizona.

Figure 2.7. Adaptive Capacity Scores for HUC12 within the Four Corners (FC) and Upper Rio Grande (URG) Focal Areas. 41

Impact (Table 2.7, Fig. 2.8) Areas with the greatest cumulative exposure and sensitivity scores are found on the eastern slopes of the Jemez Mountains in New Mexico and the area to the north of El Malpais National Monument and near Mt. Taylor. Watersheds with Moderate impact scores were found throughout both study areas. A single watershed outside of Flagstaff received a score of very high. In general, URG watersheds scored as having lower expected impact (exposure + sensitivity) than watersheds within the FC. Low impact areas are found throughout the study areas.

Table 2.7. Percent HUC12 within each Impact category. % FC % URG None 0.4 4.4 Minimal 21.3 40.1 Low 54.9 42.0 Moderate 22.5 12.4 High 0.9 1.1 Very High 0.1 0.0

Figure 2.8. Impact Scores for HUC12 within the Four Corners (FC) and Upper Rio Grande (URG) Focal Areas.

Vulnerability (Table 2.8, Fig. 2.9) Across the collective study area, a single watershed outside of Flagstaff, AZ received the highest score indicating high impact and low adaptive capacity. Other highly vulnerable watersheds in the URG are present on the east flanking slopes of the Jemez Mountains and outside of Mt Taylor. Within the FC, highly vulnerable watersheds are located near Crownpoint and Show Low, AZ and the Chuska Mountain range, AZ. Sites with low vulnerability scores are scattered throughout the area though a number are Table 2.7. Percent HUC12 within each present along the border between URG and FC Vulnerability category. focal areas. Results are fairly localized to % FC % URG individual watersheds and it is not uncommon to None 0.2 1.5 find adjacent watersheds with different Lowest 0.0 0.0 vulnerability classes. For instance, areas near Very Low 1.2 0.9 Show Low, Arizona contain sites that received Low 21.4 42.6 both very high and very low vulnerability scores. Intermediate 54.1 41.7 High 22.1 12.2 Similarly, the watersheds near El Malpais National Very High 0.9 1.1 Conservation Area have some of the highest and Highest 0.1 0.0 lowest scores in the URG.

Figure 2.8. Vulnerability Scores for HUC12 within the Four Corners (FC) and Upper Rio Grande (URG) Focal Areas. 43

Discussion Overview Piñon-Juniper ecosystems have benefited from human related activities in the western U.S. The degree to which these benefits have resulted in the establishment of new persistent woodlands varies according to the geographic region considered. However, PJ ecosystems are also highly susceptible to widespread die-off due to the collective stress of drought and insect infestations. Though these impacts are not felt equally by the piñon and juniper species comprising these ecosystems, recent research and this assessment indicate that the collective piñon-juniper woodland ecosystem is at risk under continued drought and climate warming. Importantly, this assessment considers PJ ecosystems in the heart of an area that has experienced dramatic die-offs in recent years. The results of this assessment and our literature review show that PJ ecosystems will continue to face numerous hurdles that are likely to contribute to further declines in important woodlands in at least some watersheds in the study area. This assessment was also able to identify several watersheds that are likely to continue to support PJ into the future.

Within SRLCC boundary, PJ ecosystems vary in tree and understory composition and structure, and are found in sites that contain a wide range of climates, soils, and land use histories. Therefore, research conducted for PJ in one site is not necessarily relevant to PJ that exist under different environmental conditions. Several recent studies attempt to tease out the relative impacts of climate and disturbance on PJ and the interaction of PJ with other ecosystems. However, information gaps identified in this assessment and during discussions held the 2017 Adaptation Forums indicate a need for far more research. Similarly, management practices are needed that consider site and stand specific conditions. The maps and data produced as part of this assessment take an important first step to distinguish areas by the specific threats and issues that are likely to impact the persistence of PJ woodlands.

Comparison of Upper Rio Grande and Four Corners Areas The URG focal area appears to have more refugia and less highly vulnerable HUC12 than the FC focal area (Fig 2.8). Areas around Mt Taylor, outside of Gallup and Flagstaff, and on Tribal lands held the greatest risk of decline under ongoing and future threats. Overall, the FC appears at a higher risk of seeing PJ declines under climate warming but also contains a larger number of watersheds that are likely to become more suitable for PJ under climate change.

Watersheds in the FC were far more likely to receive a score for exposure indicators relating to climate (loss of suitable climate, increase in max summer temperature and decline in winter precipitation) than watersheds in the URG (Fig. 2.2; Table 2.3). Concurrently, a greater number watersheds within the FC were expected to see increases in climates suitable for PJ than in the UGR (Fig 2.4). Declines in suitable climates were mostly concentrated in northern and central areas of the FC (Fig 2.2), whereas increases were more common in southern areas (Fig 2.5). Within the URG, increase in suitable habitat tend to be located along the Rio Grande corridor in the north and losses along the Rio Grande corridor to the south (Figs 2.2 and 2.5). Energy development is a far greater threat to HUC12 within the northern portion of the FC than elsewhere (Fig.2.3). For both areas, watersheds to the north appear more sensitive to disturbances, likely because they tend to harbor more species and have concentrations of high intensity development (Fig. 2.4). FC HUC12 were more likely to contain soils with low AWS and low percent cover

of PJ woodlands (Fig. 2.3), whereas a greater proportion of watersheds in the URG were affected by mechanical disturbance and contained a greater diversity of wildlife. Using This Data The results of the current assessment reflects a compilation of data representing threats and issues and potential resiliencies of PJ woodlands within two focal landscapes. Within the study landscape, this assessment found several areas where impacts from identified threats and issues will exceed potential adaptive capacity and result in high vulnerability. We also found many areas within the SRLCC boundary that appear to be at relatively low risk of future decline. Figure 2.9 shows several areas with a high risk of PJ decline as well as potential refugia or areas that may hold conservation opportunities. The maps produced from this analysis provide a backdrop that can be combined with maps of other important resources to identify areas that may need actions to reduce current or expected stressors and areas that are most likely to persist and present opportunities for conservation. These maps and associated spatial data are available on the ScienceBase website: https://www.sciencebase.gov/catalog/item/5693e5a9e4b09c7f9a21a428

Spatially explicit vulnerability assessments provide a rapid method for viewing relative impacts to a resource across large landscapes. Maps allow managers to find the locations where PJ is likely to persist into the future, as well as areas that are vulnerable to type change. This assessment provides the first step toward developing management options for PJ within two focal landscapes but, in some situations, the best course of action following this assessment will be to ‘ground truth’ low/high vulnerability watersheds with the land managers of those watersheds.

In addition to identifying sources of vulnerability, each map and its associated data provides a mechanism for identifying potential management intervention points. By considering the indicators that contribute to a vulnerability score for a particular watershed, resource managers can identify actions to reduce current stressors (e.g. energy development and wildfire risk) or increase resiliency (e.g. increasing protected areas or recruitment of PJ). Opportunities for conservation or potential intervention may focus on the individual indicators (e.g. road density, wildlife diversity) identified as an issue for a given watershed, may consider the collective impact of multiple indicators, or may identify issues common to a group of watersheds.

It is also important to consider these results in context of each vulnerability element: exposure, sensitivity and adaptive capacity. For instance, lower vulnerability scores may be generated for areas where both impact and adaptive capacity are low or may arise when impact and adaptive capacity are high (thus balancing each other out). The need for management intervention is likely to be quite different depending on which of these situations apply to a particular management unit.

Missing Data and Other Uncertainties We provide details on individual datasets, including uncertainties, in Appendix 2. This information needs to be considered before data from this assessment are adapted to new efforts or extrapolated to sites outside the current focal areas. In addition, several caveats exist related to the use of the data presented in the current assessment. First, we treat PJ as a unit of ecological importance and do not distinguish between relative impacts of climate and related disturbances for piñon and juniper species

despite the fact that several indicators (e.g. Low Available Soil Water) relate to specific attributes of PIED. In addition, indicators were sometimes drawn from studies conducted outside the current study area. Therefore, the results of these assessments may reflect more expectations for PIED than juniper components within PJ ecosystems and some indicators may be more or less relevant to the current focal area.

Second, this assessment and the quantified scores upon which it is based are generated based on an uneven distribution of indicators. Specifically, we have four indicators of adaptive capacity, six indicators of sensitivity, and seven indicators of exposure. Since the data contain a greater number of exposure and sensitivity measures, and each indicator contributes 1 point to the final exposure or sensitivity score, which is then scaled to 1-10, each exposure and sensitivity indicator contributes less to the final vulnerability scores than each adaptive capacity indicator. As a specific example, each exposure 1/ indicator contributes 7 to the final exposure score, whereas each adaptive capacity indicator contributes ¼ to the final adaptive capacity score. In addition, our measures of adaptive capacity were often not found within the watersheds of the focal areas, which may reflect a true lack of adaptive capacity in these watersheds or a lack of relationship between the indicators chosen for the assessment and factors that relate to PJ adaptive capacity. Perhaps as a result, many of the suggestions for further research derived from this assessment and discussions during the 2017 Adaptation Forums (see below), focus on elements that may improve our capacity to identify and measure adaptive capacity in PJ ecosystems.

Third, several indicators of exposure, sensitivity and adaptive capacity were not available at the time of this assessment and are not represented in these assessment maps. In particular, our literature review noted a large impact from grazing pressure and drought stress. Grazing data is available for BLM but is not consistently available for other lands, which prevented us from creating a reliable representation of grazing pressure across the study area. Measures of drought are available but we lack research that provides a definitive response between these indices and PJ woodland response that could be used in this type of assessment (though Clifford et al. 2011 does identify precipitation thresholds). Future research that can better identify drought impacts in PJ could facilitate a method for conducting to real- time vulnerability assessments (Medden et al. 2015).

Several other indicators were identified by the 2017 Adaptation Forum participants (Workshop report available here: https://www.sciencebase.gov/catalog/item/56a79ee6e4b0b28f1184d937). Among those identified that would improve future assessment efforts are:

 The occurrence of piñon jay nests or populations, which could be used as proxies to measure of PJ health and regeneration capacity. Piñon jays are relatively understudied and exhibit irruptive movement patterns which challenges analysis of habitat suitability.  Additional measures of PJ adaptive capacity relating to piñon and juniper dispersal rate and distance. These measures would be useful to identify areas for assisted migration, etc.  Estimates or measures of wildfire and mechanical disturbances that relate to habitat qualities of PJ. In general, there is a lack of information on species use of PJ.  More definitive information on the relationship between specific fire regimes and various PJ ecosystem types. Research is needed specific to the ecosystems present in the SW.

 Additional climate variables that link to the likelihood of masting events. Piñon crops require 3 years to develop during which certain precipitation and temperature conditions must be present. Research is needed to identify specific climate parameters that could represent these necessary conditions. Additionally, methods are needed to translate future climate projections to more local annual weather events that influence to masting events.  Metrics on age and stand structure of PJ forests. This assessment pertains to a broad swath of diverse habitats that are likely to benefit from different management strategies. To provide site specific information, assessments need to be able to stratify the landscape according to PJ structure, age and composition.  Metrics relating to Natural Range of Variability as an additional indicator for adaptive capacity (within natural range) or Sensitivity (departed from natural range). This information, used in other landscape assessments of aquatic and riparian systems, is not currently available for PJ ecosystems within the study area.

Management Considerations Woody plant encroachment has long been a concern for ranchers who want to maximize forage production and conservationists who want to maximize grassland diversity. Historical management practices for rangelands includes deliberate removal of trees by mechanical, chemical, or fire means. However, it is not clear that these methods fully achieve management goals and discrepancies between short- and long-term flora and fauna responses to treatment call for additional research. In particular, research needs to consider how soil, topographic and land use practices mediate plant response to such treatments (Archer et al. 1996).

Other recent assessments have compiled management suggestions for PJ that are relevant to the systems present within FC and URG focal areas (EcoAdapt 2016; Rondeau et al. 2017). Adaptive strategies for PJ can center around three types of actions (Rondeau et al. 2017): 1) identify and protect persistent ecosystems as refugia; 2) proactively manage for resilience; and 3) accept, assist and allow for transformation in non-climate refugia sites. The maps produced as part of this assessment provide a first step towards identifying stands with high productivity that can be managed to sustain wildlife and cultural needs. From discussions during the Adaptation Forums and through the results of this assessment including the literature review, we identify several key management strategies for increasing PJ resilience in the FC and URG focal areas.

1. Reduce stressors relating to human disturbances where possible (e.g. energy activities, development, urbanization). Several HUC12 within the focal areas are impacted by multiple threats and issues leading to high vulnerability to decline under future climate change. 2. Improve thinning practices to reduce negative impacts for PJ and PJ dependent communities. Thinning woodlands can lead to a more resilient understory, reduce sediment production and runoff, increase post fire recovery, reduce fire risk and competition for water, and improve intercanopy site conditions (e.g. soil moisture)(Chambers et al. 2013; Jacobs 2015; Ecoadapt 2016). However, Shinneman and Baker (2009) note that recent fluctuations in tree populations as a result of drought indicate efforts to reduce current stands may be misdirected. Dense stands can be thinned, but complete removal of PJ

stands provides only short-term benefits to wildlife (Cokran and Wind 2008). Mechanical treatments to thin sites should consider soil conditions and climate projections to determine whether PJ could or should continue to persist at a site (Davidson 1998). Thinning PJ to any degree may be counterproductive in landscapes that are likely to experience PJ declines as a result of climate and disturbance. Scale is also important to consider when manipulating PJ stand structure as several wildlife studies note a preference of species for a mosaic of different vegetation types (Bender et al. 2013; Bergman et al. 2015). 3. Manage for wildfire. PJ is intolerant of fire (Baker and Shinneman 2004; Floyd and others 2004) and fire suppression combined with fuel mangement treatments may be the best strategy for maintaining PJ woodlands especially where stand replacing fires are likely (Rocca and others 2015). Ecoadapt (2016) in their assessment of California PJ stands, identify several management actions for conserving PJ in the face of wildfire threats: prevent stand replacing fire by reducing fine fuels, especially exotic grass species; aggressively manage wildfire; and, ensure survival of seedlings by creating barriers to sources of human ignition. Short- and long-term projections for PJ are complicated by the presence of invasive grasses (cheatgrass), which can lead to more frequent wildfire and further encourage the conversion of PJ to annual, invasive grasslands (Shinneman and Baker 2009; Rocca and others 2015). As a result, invasive species management is also an important component of any management plan that seeks to reduce wildfire risk in existing PJ stands. 4. Facilitate opportunities for PJ expansion where models indicate suitable climate. Models project both increasing and decreasing climate suitability for PJ in the study location and a reduction of co-occurring piñon and juniper species in some areas (Rehfeldt and others 2006; Keane and others 2008; Romme and others 2008). Managers can prepare for inevitable range shifts and maintain ecologically relevant PJ ecosytems by collecting seed from low elevation bands and increasing genetic diversity of seed stocks (Ecoadapt 2016). Surveys that map ongoing transitions to shrublands or other conversions and identify current stand structure will also be important for identifying and implementing appropriate and effective management strategies. As well, reducing fragmentation of current PJ stands and maintaining potential corridors for migration will facilitate natural range shifts (Ecoadapt 2016).

Finally, participants of the 2017 Adaptation Forums noted three specific research and management needs that will support the success of efforts to preserve important PJ woodlands within the focal areas. First, more research or synthesis on fire regimes in relation to the various forest types present within the FC and URG focal areas. Currently it is unclear to what degree fuels treatments can benefit different types of PJ or PJ woodlands that exist in different environments. Second, the development of Ecological Site Descriptions, which currently do not exist for this region, would help managers identify a common and necessary baseline to compare various management strategies. Also, additional information linking current PJ ecosystem type to specific soil and moisture regimes could be useful for additional vulnerability assessment efforts. Finally, management strategies should strive to consider future conditions. For example, future projections of suitable habitat could be used to reduce mechanical disturbance in critical areas (refugia) or help in efforts to triage actions within a landscape by allowing managers to focus on preserving PJ stands where they are most likely to remain viable into the future.

Literature 1. Allen C.D., Macalady A.K., Chenchouni H., Bachelet D., McDowell N., Vennetier M., Kitzberger T., Rigling A., Breshears D.D., Hogg E.H., Gonzalez P., Fensham R., Zhang Z., Castro J., Demidova N., Lim J-H., Allard G., Running S.W., Semerci A., and N. Cobb, 2010. A global overview of drought and heat- induced tree mortality reveals emerging climate change risks for forests. Forest Ecology and Management 259: 660-684.

2. Allen, C.D. 2007. Interactions across spatial scales among forest dieback, fire, and erosion in northern New Mexico landscapes. Ecosystems 10: 797-808.

3. Allen, C.D., Anderson, R.S., Jass, R.B., et al. 2008. Paired charcoal and tree-ring records of high frequency Holocene fire from two New Mexico bog sites. International Journal of Wildland Fire. 17(1): 115–130.

4. Allen, C.D., Breshears, D.D., and N.G. McDowell. 2015. On underestimation of global vulnerability to tree mortality and forest die-off from hotter drought in the Anthropocene. Ecosphere 6: 129.

5. Allen, C.D., and D.D. Breshears. 1998. Drought-induced shift of a forest-woodland ecotone: Rapid landscape response to climate variation. Proceedings of the National Academy of Sciences USA. 95(25): 14839–14842.

6. Anderegg, W.R., Kane, J.M. and L.D. Anderegg. 2013. Consequences of widespread tree mortality triggered by drought and temperature stress. Nature Climate Change, 3(1): 30.

7. Anderson, M.D. 2002. Pinus edulis. In: Fire Effects Information System, [Online]. U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station, Fire Sciences Laboratory (Producer). Available: https://www.fs.fed.us /database/feis/plants/tree/pinedu/all.html [ 2017, August 30].

8. Archer, S. 1996. Assessing and interpreting grass-woody plant dynamics. The ecology and management of grazing systems, pp.101-134.

9. Arendt, P.A., and W.L. Baker. 2013. Northern Colorado Plateau piñon-juniper woodland decline over the past century. Ecosphere. 4(8): 1–30.

10. Arizona Game and Fish Department. 2012. Arizona’s State Wildlife Action Plan: 2012-2022. Arizona Game and Fish Department, Phoenix, Arizona.

11. Arnold, J.F.; Jameson, D.A.; Reid, E.H. 1964. The piñon-juniper type of Arizona: Effects of grazing, fire, and tree control. Product Research Report No. 84. Washington, DC: Department of Agriculture, Forest Service. 28 p.

12. Baker, W.L., and D.J. Shinneman. 2004. Fire and restoration of piñon-juniper woodlands in the western United States: a review. Forest Ecology and Management 189: 1- 21.

13. Balch, Jennifer K.; Bradley, Bethany A.; D’Antonio, Carla M.; [et al.]. 2013. Introduced annual grass increases regional fire activity across the arid western USA (1980–2009). Global Change Biology. 19(1): 173–183.

14. Balda, R.P. 1987. Avian impacts on pinyon-juniper woodlands. p. 525-533. In Proceedings-Pinyon- Juniper Conference. Richard L. Everett, camp. USDA Forest Service, General Technical Report INT- 215. Intermountain Forest and Range Experiment Station, Ogden, UT. 581 p.

15. Balda, R.P., and N. Masters,. 1980. Avian communities in the pinyon-juniper woodland: a descriptive analysis. In: DeGraaf, R.M., technical coordinator. Management of western forests and grasslands for nongame birds: Workshop proceedings; 1980 February 11-14; Salt Lake City, UT. Gen. Tech. Rep. INT-86. Ogden, UT: U.S. Department of Agriculture, Forest Service, Intermountain Forest and Range Experiment Station: 146-169.

16. Barger N.N., S. Archer, J. Campbell, C. Huang, J. Morton, and A.K. Knapp. 2011. Woody plant proliferation in North American drylands: A synthesis of impacts on ecosystem carbon balance. Journal of Geophysical Research: Biogeosciences 116: G00K07. doi: 10.1029/2010JG001506.

17. Barger, N.N., Adams, H.D., and C. Woodhouse, [et al.]. 2009. Influence of livestock grazing and climate on piñon pine (Pinus edulis) dynamics. Rangeland Ecology and Management. 62(6): 531–539.

18. Barney, M.A. and N.C. Frischknecht. 1974. Vegetation changes following fire in the piñon-juniper type of west-central Utah. Journal of Range Management, 91-96.

19. Barth, R.C. 1980. Influence of Piñon Pine Trees on Soil Chemical and Physical Properties. Soil Science Society of America Journal 44: 112-114.

20. Bell, G., J. Baumgartner, J. Humke, A. Laurenzi, P. McCarthy, P. Mehlhop, K. Rich, M. Silbert, E. Smith, B. Spicer, T. Sullivan, and S. Yanoff 1999. Ecoregional Conservation Analysis of the Arizona - New Mexico Mountains Prepared by The Nature Conservancy’s Arizona - New Mexico Mountains Ecoregional Conservation Team Published by The Nature Conservancy, Santa Fe, New Mexico. 1999 The Nature Conservancy 73 pages +appendices

21. Belnap, J., and D. Eldridge. 2001. Disturbance and Recovery of Biological Soil Crusts. Pages 363-383 in Belnap, J., and O. L. Lange, editors. Biological Soil Crusts: Structure, Function, and Management. Ecological Studies, Vol. 150. Springer-Verlag Berlin Heidelberg New York. 503 pp.

22. Bender, L.C., Boren, J.C., Halbritter, H. and S. Cox. 2013. Effects of site characteristics, piñon-juniper management, and precipitation on habitat quality for mule deer in New Mexico. Human–Wildlife Interactions 7: 5.

23. Bentz B.J., Régnière J., Fettig C.F., Hansen E.M., Hayes J.L., Hicke J.A., Kelsey R.G., Negrón J.F., and S.J. Seybold. 2010. Climate change and bark beetles of the western United States and Canada: direct and indirect effects. BioScience 60(8): 602-613.

24. Bergman, E. J., P.F. Doherty, G.C. White, and D.J. Freddy. 2015. Habitat and herbivore density: response of mule deer to habitat management. Journal of Wildlife Management 79: 60-68.

25. Bergman, E.J., C.J. Bishop, D.J. Freddy, G.C. White, and P.F. Doherty. 2014. Habitat management influences overwinter survival of mule deer fawns in Colorado. Journal of Wildlife Management 78: 448-455.

26. Betancourt, J.L., E.A. Pierson, K.A. Rylander, J.A. Fairchild- Parks, and J. S. Dean. 1993. Influence of history and climate on New Mexico piñon-juniper woodlands. Pages 42-62 in E. F. Aldon and D. W. Shaw, editors. Managing P-J Ecosystems for Sustainability and Social Needs: Symposium Proceedings. General Technical Report RM-236, USDA Forest Service, Rocky Mountain and Range Experiment Station.

27. Beymer, R.J. and J.M. Klopatek. 1992. Effects of Grazing on Cryptogamic Crusts in Piñon-juniper Woodlands in Grand Canyon National Park Source: The American Midland Naturalist 127: 139-148. Stable URL: https://www.jstor.org/stable/2426329

28. Board, D.I., Chambers, J.C., Miller, R.F., and P.J. Weisberg. 2018. Fire patterns in piñon and juniper land cover types in the Semiarid Western United States from 1984 through 2013. RMRS-GTR-372. Fort Collins, CO: U.S Department of Agriculture, Forest Service, Rocky Mountain Research Station. 57 p.

29. Bombaci, S. and L. Pejchar. 2016. Consequences of piñon and juniper woodland reduction for wildlife in North America. Forest Ecology and Management 365: 34-50.

30. Bosworth W.R. III. 2003. Vertebrate information compiled by the Utah Natural Hertiage Program, a progress report. Salt Lake City, UT: State of Utah, Department of Natural Resources, Division of Wildlife Resources. Publication Number 03-45. 336 p.

31. Bradford D.F., Franson S.E., Neale A.C., Heggem D.T., Miller G.R., and G.E. Canterbury. 1998. Bird species assemblages as indicators of biological integrity in Great Basin rangeland. Environmental Monitoring and Assessment 49: 1-22.

32. Bradley, B.A. 2009. Regional analysis of the impacts of climate change on cheatgrass invasion shows potential risk and opportunity. Global Change Biology 15: 196-208.

33. Bradley, B.A. and E. Fleishman. 2008. Relationships between expanding piñon–juniper cover and topography in the central Great Basin, Nevada. Journal of Biogeography 35: 951-964.

34. Breshears, D.D., Cobb, N.S., Rich, P.M., Price, K.P., Allen, C.D., Balice, R.G., Romme, W.H., Kastens, J.H., Floyd, M.L., Belnap, J., Anderson, J.J.,Myers, O.B., and C.W.Meyer. 2005. Regional vegetation die-off in response to global-change-type drought. Proceedings of the National Academy of Sciences of the United States of America 102: 15144–15148.

35. Brown P.M., Kaye M.W., Huckaby L.S., and C.H Baisan. 2001. Fire history along environmental gradients in the Sacramento Mountains, New Mexico: influences of local patterns and regional processes. Ecoscience 8: 115–126.

36. Cartledge, T.R. and J.G. Propper. 1993. Piñon-juniper ecosystems through time: Information and insights from the past. From Aldon, E. C. and D. W. Shaw (Tech. Coords.). Managing piñon-juniper ecosystems for sustainability and social needs. RM-GTR-236. USDA Forest Service General Technical Report.

37. Chambers J.C., and M. Pellant 2008. Climate Change Impacts on Northwestern and Intermountain United States Rangelands. Rangelands June: 29-33.

38. Chambers J.C., Pendleton B.K., Sada D.W., Ostoja S.M., and M.L. Brooks. 2013. Maintaining and Restoring Sustainable Ecosystems in Southern Nevada. 7. In: editors. RMRS-GTR-303. USDA Forest Service, Rocky Mountain Research Service. p 125-154.

39. Chambers, J.C., Bradley, B.A., Brown, C.A., [et al.]. 2014. Resilience to stress and disturbance, and resistance to Bromus tectorum L. invasion in the cold desert shrublands of western North America. Ecosystems 17: 360–375.

40. Chambers, J.C., Vander Wall, S.B., and E.W. Schupp, 1999. Seed and Seedling ecology of Piñon and Juniper Species in the Pygmy Woodlands of Western North America. The Botanical Review 65: 1-8.

41. Christensen, J. H., B. Hewitson, A. Busuioc, A. Chen, X. Gao, I. Held, R. Jones, R. K. Kolli, W.-T. Kwon, R. Laprise, V. Magaña Rueda, L. Mearns, C. G. Menéndez, J. Räisänen, A. Rinke, A. Sarr and P. Whetton. 2007. Chapter 11, Regional Climate Projections. Pages 847- 940 in Solomon, S., D. Qin, M. Manning, Z. Chen, M. Marquis, K. B. Averyt, M. Tignor and H. L. Miller, editors. Climate Change 2007: The Physical Science Basis. Contribution of Working Group 1 to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA. URL: http:// www.ipcc.ch/pdf/assessment-report/ar4/wg1/ar4- wg1- chapter11.pdf. Accessed December 2008.

42. Clark, J.S., L. Iverson, C.W. Woodall, C.D. Allen, D.M. Bell, D.C. Bragg , A.W. D’amato, F.W. Davis , M.H. Hersh, I. Ibanez , S.T . Jackson, S. Matthews, N. Pederson, M. Peters, M.W. Schwartz, K. Sten, M. Waring, and N.E. Zimmermann. 2016. The impacts of increasing drought on forest dynamics, structure, and biodiversity in the United States. Global Change Biology doi: 10.1111/gcb/13160

43. Clary, W.P., Baker Jr, M.B., O'Connell, P.F., Johnson Jr, T.N. and R.E. Campbell. 1974. Effects of Piñon- Juniper removal on natural resource products and uses in Arizona. (RM-128)

44. Clifford, M.J., N.S. Cobb, and M. Buenemann. 2011. Long-term tree cover dynamics in a piñon- juniper woodland: climate-change-type drought resets successional clock. Ecosystems 14.6: 949- 962.

45. Clifford, M.J., Royer, P.D., Cobb, N.S., Breshears, D.D. and P.L. Ford. 2013. Precipitation thresholds and drought‐induced tree die‐off: insights from patterns of Pinus edulis mortality along an environmental stress gradient. New Phytology 200: 413-421. doi:10.1111/nph.12362

46. Coates, P.S., B.G. Prochazka, M.A. Ricca, K.B. Gustafson, P. Ziegler, and M.L. Casazza. 2017. Piñon and Juniper Encroachment into Sagebrush Ecosystems Impacts Distribution and Survival of Greater Sage-Grouse. Rangeland Ecology & Management 70: 25-38.

47. Coates, P.S., Gustafson, K.B., Roth, C.L., Chenaille, M.P., Ricca, M.A., Mauch, Kimberly, Sanchez- Chopitea, Erika, Kroger, T.J., Perry, W.M., and M.L. Casazza. 2017. Geospatial data for object-based high-resolution classification of conifers within greater sage-grouse habitat across Nevada and a portion of northeastern California: U.S. Geological Survey data release, https://doi.org/10.5066/F7348HVC.

48. Cole, K.L., Fisher, J., Arundel, S.T., Cannella, J. and S. Swift. 2008. Geographical and climatic limits of needle types of one‐and two‐needled piñon pines. Journal of Biogeography 35: 257-269.

49. Cole, K.L., Ironside, K.E., Arundel, S.T., Duffy, P. and J. Shaw. 2008. Modeling future plant distributions on the Colorado Plateau: an example using Pinus edulis. In: van Riper, C. III; Sogge, MK, Eds. The Colorado Plateau III. Integrating Research and Resources Management for Effective Conservation. Tucson, AZ: The University of Arizona Press. p. 319-330.

50. Colorado Parks and Wildlife. 2015. Colorado’s 2015 State Wildlife Action Plan. Cpw.state.co.us

51. Corkran C, and E. Wind. 2008. Juniper and pinyon-juniper woodlands. In: Pilliod DS, Wind E., editors. Habitat management guidelines for amphibians and reptiles of the Northwestern United States and Western Canada. Partners in Amphibian and Reptile Conservation, Technical Publication HMG-4, Birmingham, AL. 65-67.

52. Craig, G.R. and Enderson, J.H. 2004. Peregrine Falcon Biology and Management in Colorado. Colorado Division of Wildlife, Technical Publication No. 43. Available: http://wildlife.state.co.us/SiteCollectionImages/Photos/Birds/PEREGRINE.pdf

53. Cryan, P.M. 2003. Seasonal Distribution of Migratory Tree Bats (Lasiurus and Lasionycteris) in North America. Journal of Mammalogy 84(2): 579–593.

54. Dai, A. 2013. Increasing drought under global warming in observations and models. Nature Climate Change 3: 52–58.

55. Davenport, D.W., D. D. Breshears, B. P. Wilcox, and C. D. Allen. 1998. Viewpoint: Sustainability of piñon-juniper ecosystems- a unifying perspective of soil erosion thresholds. Journal of Range Management 51: 231-240.

56. Decker, K. and Fink, M., 2014. Colorado wildlife action plan enhancement: climate change vulnerability assessment. Colorado Natural Heritage Program, Colorado State University, Fort Collins, USA.

57. Decker, K., L. Grunau, J. Handwerk, and J. Siemers, editors. 2015. Climate Change Vulnerability Assessment for Colorado Bureau of Land Management. Colorado Natural Heritage Program, Colorado State University, Fort Collins, Colorado.

58. Diffenbaugh, N.S. and F. Giorgi. 2012. Climate change hotspots in the CMIP5 global climate model ensemble. Climatic Change 114(3-4): 813-822.

59. Dillon, G.K., Holden, Z.A., and P. Morgan, [et al.]. 2011. Both topography and climate affected forest and woodland burn severity in two regions of the western US, 1984 to 2006. Ecosphere. 2(12): Article 130. http://onlinelibrary.wiley.com/doi/10.1890/ES11-00271.1/epdf [Accessed January 16, 2017].

60. Doeskin NJ, Pielke RA, Bliss OAP. 2003. Climate of Colorado. Climatography of the United States No. 60. Colorado Climate Center. http://ccc.atmos.colostate.edu/climateofcolorado.php

61. DOI, 2013. Endangered and threatened wildlife and plants: endangered status for Gunnison Sage- grouse Part II. Federal Register, Department of the Interior, p. 81.

62. Drake, B.G., L. Hughes, E.A. Johnson, B.A. Seibel, M.A. Cochrane, V.J. Fabry, D. Rasse, and L. Hannah, 2005: Synergistic effects. In: Climate Change and Biodiversity [Lovejoy, T.E. and L.J. Hannah (ed.)]. Yale University Press, New Haven, CT, USA, pp. 296-316.

63. EcoAdapt. 2016. Climate change vulnerability assessment for the southern California climate adaptation project. Available at:

64. Falco III, Gennaro. 2014. Community classification of piñon-juniper vegetation in the four corner states. Thesis. Northern Arizona University.

65. Flint-Lacey, P. R. 2003. The Ancestral Puebloans and their piñon-juniper woodlands. Pages 309-319 in L. M. Floyd, D. D. Hanna, W. H. Romme, M. Colyer, editors. Ancient Piñon-Juniper Woodlands: A Natural History of Mesa Verde Country. University Press of Colorado. 389 pp.

66. Floyd, L.M., Clifford, M., Cobb, N.S., Hanna, D., Delph, R., Ford, P. and D. Turner. 2009. Relationship of stand characteristics to drought-induced mortality in three southwestern piñon-juniper woodlands. Ecological Applications 19: 1223-1230.

67. Floyd, M. L., and D.D. Hanna. 1990. Cone indehiscence in a peripheral population of pinon pines (Pinus edulis). The Southwestern Naturalist 35(2): 146-150.

68. Floyd, M.L., D.D. Hanna, and W.H. Romme. 2004. Historical and recent fire regimes in piñon-juniper woodlands on Mesa Verde, Colorado, USA. Forest Ecology and Management 198: 269-289.

69. Floyd, M.L., Romme, W.H., and D.D. Hanna, [et al.]. 2008. Fire history of piñon-juniper woodlands on Navajo Point, Glen Canyon National Recreation Area. Natural Areas Journal. 28(1): 26–36.

70. Floyd, M.L., W.H. Romme, and D.D. Hanna. 2000. Fire history and vegetation pattern in Mesa Verde National Park, Colorado, USA. Ecological Applications 10(6): 1666-1680.

71. Frontiers in Ecology and the Environment 9, 271–277. doi:10.1890/090199

72. Furniss, M.J., K.B. Roby, D. Cenderelli, J. Chatel, C.F. Clifton, A. Clingenpeel, P.E. Hays, D. Higgins, K. Hodges, K.,C. Howe, L. Jungst, J. Louie, C. Mai, R. Martinez, K. Overton, B.P. Staab, R. Steinke, and M. Weinhold. 2013. Assessing the vulnerability of watersheds to climate change: results of national forest watershed vulnerability pilot assessments. Gen. Tech. Rep. PNW-GTR-884. Portland, OR: U.S. Department of Agriculture, Forest Service, Pacific Northwest Research Station. 32 p. plus appendix

73. Füssel, H.M., 2007. Vulnerability: A generally applicable conceptual framework for climate change research. Global Environmental Change 17: 155-167.

74. Gallo, T., Stinson, L.T. and L. Pejchar. 2016. Pinyon-juniper removal has long-term effects on mammals. Forest Ecology and Management 377: 93-100.

75. Garfin, G., G. Franco, H. Blanco, A. Comrie, P. Gonzalez, T. Piechota, R. Smyth, and R. Waskom. 2014. Ch. 20: Southwest. Climate Change Impacts in the United States: The Third National Climate Assessment, J.M. Melillo, T.C. Richmond, and G.W. Yohe, Eds., U.S. Global Change Research Program, 462-486. doi: 10.7930/J08G8HMN.

76. Gaylord, M.L., Kolb, T.E., and N.G. McDowell. Mechanisms of piñon pine mortality after severe drought: a retrospective study of mature trees. Tree Physiology 35: 806-816.

77. Glick, P., Stein, B.A. and N.A. Edelson. 2011. Scanning the conservation horizon: a guide to climate change vulnerability assessment. National Wildlife Federation, Washington DC.

78. Goguen, C.B., and N.E. Mathews. 1998. Songbird community composition and nesting success in grazed and ungrazed piñon-juniper woodlands. Journal of Wildlife Management 62: 474-484.

79. Gori, D., and J. Bate. 2007. Historical range of variation and state and transition modeling of historical and current landscape conditions for montane grassland for the southwestern U.S. Prepared for the USDA Forest Service, Southwestern Region by The Nature Conservancy, Tucson, AZ. 21pp. URL: http://azconservation.org/projects/ southwest_forest_assessment. Accessed September 2008.

80. Gottfried, G.J., T.J. Swetnam, C.D. Allen, J.L. Betancourt, and A.L. Chung-MacCoubrey. 1995. Piñon- juniper woodlands. Pages 95-132 in D. M. Finch and J. A. Tainter, editors. Ecology, diversity, and sustainability of the middle Rio Grande basin. GTR-RM-268. USDA Forest Service, Fort Collins, Colorado, USA.

81. Greenwood, D.L., and P.J. Weisberg. 2008. Density-dependent tree mortality in piñon-juniper woodlands. Forest Ecology and Management. 255(7): 2129–2137.

82. Gruell, G. E. 1999. Historical and modern roles of fire in piñon-juniper. Pages 24-28 in S.B. Monsen and R. Stevens (Eds.). Ecology and management of piñon-juniper communities within the Interior West. Proceedings RMRS-P-9. U.S. Dept. of Agriculture, Forest Service, Rocky Mountain Research Station, Ogden, Utah.

83. Gutzler, D.S., and T.O. Robbins. 2011. Climate variability and projected change in the western United States: Regional downscaling and drought statistics. Climate Dynamics 37: 835–849.

84. Hall, L., and R.P. Balda. 1988. The role of scrub jays in pinyon regeneration. Final report: Cooperative agreement number 28-(X-387, Northern Arizona University). USDA Forest Service, Rocky Mountain Forest and Range Experiment Station, Fort Collins, CO. 26 p

85. Herrmann, S.M., Didan, K. Barreto-Munos, A. and M.A. Crimmins. 2016. Divergent responses of vegetation cover in Southwestern U.S. ecosystems to dry and wet years at different elevations. Environmental Research Letters 11: https://doi.org/10.1088/1748-9326/11/12/124005

86. Hinkel, J. 2011. Indicators of vulnerability and adaptive capacity: towards a clarification for the science-policy interface. Global Environmental Change 21: 198-208.

87. Holechek, J.L. 1981. Brush control impacts on rangeland wildlife. Journal of Soil and Water Conservation 36(5): 265-269.

88. Houle, D., A. Bouffard, L. Duchesne, T. Logan, and R. Harvey. 2012. Projections of future soil temperature and water content for three southern Quebec forested sites. Journal of Climate 25(21): 7690-7701.

89. http://dx.doi.org/10.1890/ ES15-00203.1

90. Huffman, D.W., Fulé, P.Z., Pearson, K.M., [et al.]. 2008. Fire history of piñon-juniper woodlands at upper ecotones with ponderosa pine forests in Arizona and New Mexico. Canadian Journal of Forest Research 38: 2097–2108.

91. Huxman, T.E., Wilcox, B.P., Breshears, D.D., Scott, R.L., Snyder, K.A., Small, E.E., Hultine, K., Pockman, W.T. and Jackson, R.B., 2005. Ecohydrological implications of woody plant encroachment. Ecology, 86(2), pp.308-319.

92. Intergovernmental Panel on Climate Change (IPCC), 2007. The Physical Science Basis, contribution of Working Group I to The Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Solomon, S., D. Qin, M. Manning, Z. Chen, M. Marquis, K.B. Averyt, M. Tignor, and H.L. Miller (Eds.). Cambridge University Press, Cambridge, United Kingdom and New York, New York. 996 pp.

93. IPCC, 2007: Climate Change 2007: Synthesis Report. Contribution of Working Groups I, II, and III to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Pachauri, R.K. and A. Reisinger, editors. IPCC, Geneva, Switzerland. 104 pp. URL: http:// www.ipcc.ch/ipccreports/ar4-syr.htm. Accessed: November 2008.

94. Jacobi, W.R., H.S.J. Kearns, D.W. Johnson. 2005. Persistence of piñon pine snags and logs in southwestern Colorado. Western Journal of Applied Forestry 20: 247-252.

95. Jacobs, B.F., Romme, W.H. and C.D. Allen. 2008. Mapping “old” vs. “young” piñon–juniper stands with a predictive topo‐climatic model. Ecological Applications 18(7): 1627-1641.

96. Jacobs, B.F. 2015. Restoration of degraded transitional (piñon–juniper) woodland sites improves ecohydrologic condition and primes understory resilience to subsequent disturbance. Ecohydrology 8:1417-1428.

97. Janetski, J.C. 1999. Role of piñon-juniper woodlands in aboriginal societies of the desert west. USDA Forest Service Proceedings RMRS-p-9.

98. Johnsen, T. N. 1962. One-seed juniper invasion of Northern Arizona Grasslands. Ecological Monographs. The Ecological Society of America 32(3):187-207.

99. Keane, R.E., Agee, J.K., Fulé, P., Keeley, J.E., Key, C., Kitchen, S.G., Miller, R., and L.A Schulte. 2008. Ecological effects of large fires on US landscapes: benefit or catastrophe? International Journal of Wildland Fire 7: 696–712.

100. Keinath, D.A. 2004. Fringed Myotis (Myotis thysanodes): a technical conservation assessment. [Online]. USDA Forest Service, Rocky Mountain Region. Available: http://www.fs.fed.us/r2/projects/scp/assessments/fringedmyotis.pdf

101. Kleinman, S.J., DeGomez, T.E., Snider, G.B. and K.E. Williams. 2012. Large-scale piñon ips (Ips confusus) outbreak in southwestern United States tied with elevation and land cover. Journal of Forestry 110(4): 194-200.

102. Kleinman, S.J., DeGomez, T.E., Snider, Gary B., and K.E. Williams. 2012. Large-Scale pinyon Ips (Ips confusus) outbreak in southwestern United States tied with elevation and land cover. Journal of Forestry, 110: 194–200. https://doi.org/10.5849/jof.11-060

103. Koepke, D.F., Kolb, T.E. and H.D. Adams. 2010. Variation in woody plant mortality and dieback from severe drought among soils, plant groups, and species within a northern Arizona ecotone. Oecologia 163(4): 1079-1090.

104. Kyllo, R. 2016. Piñon-Juniper Woodlands of the Western United States: Are we on the Brink of Piñon Oblivion? Thesis. Northern Arizona University.

105. Laing, S.P. 1988. Cougar habitat selection and spatial use patterns in southern Utah. - M.Sc. thesis, University of Wyoming, 68 pp.

106. Lanner, Ronald M. 1981. The pinon pine: A natural and cultural history. Reno, NV: University of Nevada Press. 208 p.

107. Laundre J.W., and L. Hernandez. 2003. Winter hunting habitat of pumas Puma concolor in northwestern Utah and southern Idaho, USA. Wildlife Biology 9: 123-129.

108. Linton, M.J., J.S. Sperry, and D.G. Williams. 1998. Limits to water transport in Juniperus osteosperma and Pinus edulis: implications for drought tolerance and regulation of transpiration. Functional Ecology 12: 906-911

109. Litzinger, W. J. 2003. A personal perspective on the ethnobotany of old-growth piñon-juniper woodlands. Pages 287-293 in L. M. Floyd, D. D. Hanna, W. H. Romme, M. Colyer, editors. Ancient Piñon-Juniper Woodlands: A Natural History of Mesa Verde Country. University Press of Colorado. 389 pp.

110. Madsen, M.D., Chandler, D.G., and J. Belnap. 2008. Spatial gradients in ecohydrological properties within a piñon-juniper ecosystem. Ecohydrology 1: 349-360.

111. Margolis, E.Q. 2014. Fire regime shift linked to increased forest density in a piñon-juniper savanna landscape. International Journal of Wildland Fire. 23(2): 234–245.

112. McCabe, M., Gohdes, D., Morgan, F., Eakin, J., Sanders, M., and C. Schmitt. 2005. Herbal Therapies and Diabetes among Navajo Indians. Diabetes Care 28 (6): 1534-1535. doi: 10.2337/diacare.28.6.1534-a.

113. McDowell, N.G., A. P. Williams, C. Xu, W.T. Pockman, L.T. Dickman, S. Sevanto, R. Pangle, J. Limousin, J. Plaut, D.S. Mackay, J. Ogee, J.C. Domec, C.D. Allen, R.A. Fisher, X. Jiang, J.D. Muss, D.D. Breshears, S.A. Rauscher and C. Koven. 2015. Multi-scale predictions of massive conifer mortality due to chronic temperature rise. Nature Climate Change Letters: NCLIMATE2873. Doi: 10.1038.

114. McDowell, N.G., Pockman, W.T., Allen C.D. [et al.], 2008. Mechanisms of plant survival and mortality during drought: why do some plants survive while other succumb to drought? New Phytology 178: 719-739.

115. Meddens, A.J., Hicke, J.A., Macalady, A.K., Buotte, P.C., Cowles, T.R. and C.D. Allen. 2015. Patterns and causes of observed piñon pine mortality in the southwestern United States. New Phytologist 206(1): 91-97.

116. Meeuwig, R.O., Budy, J.D., and R.L. Everett, 1990. Pinus monophylla Torr. & Frem. singleleaf pinyon. In: Burns, Russell M.; Honkala, Barbara H., technical coordinators. Silvics of North America.

Volume 1. Conifers. Agric. Handb. 654. Washington, DC: U.S. Department of Agriculture, Forest Service: 380-384.

117. Miller, M. E. 2005. The structure and functioning of dryland ecosystems- Conceptual models to inform long-term ecological monitoring: U. S. Geological Survey Scientific Investigations Report 2005-5197. 74 pp.

118. Miller, R. F., and P.E. Wigand. 1994. Holocene changes in semiarid pinyon–juniper woodlands. Bioscience 44: 465–474.

119. Miller, R.F., Tausch, R.J., McArthur, E.D., Johnson, D.D. and S.C. Sanderson. 2008. Age structure and expansion of piñon-juniper woodlands: a regional perspective in the Intermountain West. Res. Pap. RMRS-RP-69. Fort Collins, CO: U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station. 15 p., 69. Notes expansion and infill in several sites in OR UT NV and ID

120. Morrison, M.L. and L.S. Hall. 1999. Habitat relationships of amphibians and reptiles in the Inyo- White Mountains, California and Nevada. In: Monsen, Stephen B.; Stevens, Richard, compilers. Sustaining and restoring a diverse ecosystem: Proceedings: ecology and management of pinyon- juniper communities within the Interior West; 1997 September 15-18; Provo, UT. Proc. RMRS-P-9. Ogden, UT: U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station: 233- 237.

121. Neely, B., P. Comer, C. Moritz, M. Lammert, R. Rondeau, C. Pague, G. Bell, H. Copeland, J. Humke, S. Spackman, T. Schulz, D. Theobald, and L. Valutis. 2001. Southern Rocky Mountains: An Ecoregional Assessment and Conservation Blueprint. Prepared by The Nature Conservancy with support from the U.S. Forest Service, Rocky Mountain Region, Colorado Division of Wildlife, and Bureau of Land Management.

122. Negron, J.F., and J.W. Wilson 2003. Attributes associated with probability of infestation by the piñon Ips confuses (coleopteran: Scolytidae) in piñon pine, Pinus edulis. Western North American Naturalist 63: 440-451.

123. New Mexico Department of Game and Fish. 2016. State Wildlife Action Plan for New Mexico. New Mexico Department of Game and Fish, Santa Fe, New Mexico, USA. http://www.wildlife.state.nm.us/download/conservation/swap/New-Mexico-State-Wildlife-Action- Plan-SWAP-Final-2017.pdf

124. Nielson, J. 2009. Series: Piñon-Juniper Woodlands. Southern Colorado Plateau network Inventory and Monitoring Program. National Park Service Article Series. Available at: https://www.nps.gov/articles/series.htm?id=0216D798-933C-2108-EB4384D97499E89A

125. O’Meara, T.E., J.B. Haufler, L.H. Stelter, J.G. Nagy. 1981. Nongame wildlife responses to chaining of piñon–juniper woodlands. The Journal of Wildlife Management 45: 381-389.

126. Oliver G.V. 2000. The bats of Utah. Salt Lake City, UT: Utah Division of Wildlife Resources. Publication Number 00-14. 141 p.

127. Oliver G.V., and J. Tuhy. 2010. Ecological integrity tables of Utah animals of conservation concern. Salt Lake City, UT: Utah Division of Wildlife Resources. 630 p.

128. Ostoja S.M., Brooks M.L., Chambers J.C., and B.K.Pendleton. 2013. Species of Conservation Concern and Environmental Stressors: Local, Regional, and Global Effects. Chapter 6. RMRS-GTR-303. USDA Forest Service, Rocky Mountain Research Station. p 97-124.

129. Parisien, M.A., Snetsinger, S., Greenberg, J.A., Nelson, C.R., Schoennagel, T., Dobrowski, S.Z. and M.A. Moritz. 2012. Spatial variability in wildfire probability across the western United States. International Journal of Wildland Fire, 21(4): 313-327.

130. Parrish, J.R., F.P. Howe, and R.E. Norvell. 1999. Utah Partners in Flight Avian Conservation Strategy. Publ. No. 99-40. Utah Partners in Flight Program, Utah Division of Wildlife Resources, Salt Lake City.

131. Paulin, K.M., Cook, J.J. and S.R. Dewey. 1999. Pinyon-juniper woodlands as sources of avian diversity. Proceedings: Ecology and management of pinyon-juniper communities within the Interior West (S.B. Monsen and R. Stevens, Comps.). US Department of Agriculture, Forest Service, Rocky Mountain Research Station P-9. Pp.240-243.

132. Peterman W., Waring R.H., Seager T., and W.L. Pollock. 2012. Soil properties affect piñon pine – juniper response to drought. Ecohydrology 6: 455–463. http://onlinelibrary.wiley.com/doi/10.1111/gcb.12976/full

133. Pilliod D.S., and E. Wind. 2008. Habitat management guidelines for amphibians and reptiles of the Northwestern United States and Western Canada. Partners in Amphibian and Reptile Conservation, Technical Publication HMG-4, Birmingham, AL. 139 pp.

134. Poiani K.A., Goldman R.L., Hobson J., Hoekstra J.M., and K.S. Nelson. 2010. Redesigning biodiversity conservation projects for climate change: examples from the field. Biodiversity and Conservation 20: 185-201.

135. Ramsey R.D., and N.E West. 2009. Vegetation of Utah. In: McGinty EIL, editors. Rangeland Resources of Utah. Logan, UT: Utah State University Cooperative Extension Services. p 49-56.

136. Redmond, M.D. and N.N. Barger. 2013. Tree regeneration following drought‐and insect‐induced mortality in piñon–juniper woodlands. New Phytologist 200(2): 402-412.

137. Redmond, M.D., F. Forcella, and N.N. Barger. 2012. Declines in piñon pine cone production associated with regional warming. Ecosphere 3 (12): art120 DOI: 10.1890/ES12-00306.1

138. Redmond, M.D., Cobb, N.S., Clifford, M.J. and N.N. Barger. 2015. Woodland recovery following drought-induced tree mortality across an environmental stress gradient. Global Change Biology 21: 3685–3695. doi:10.1111/gcb.12976

139. Rehfeldt G.E., Crookston N.L., Warwell M.V., and J.S. Evan. 2006. Empirical analyses of plant- climate relationships for the Western United States. International Journal of Plant Sciences 167: 1123-1150.

140. Rhea, B., Bidwell, M. and C. Livensperger. Sensitive Species Assessment of Vulnerability to Climate Change on San Juan Public Lands, Colorado. Prepared by Mountain Studies Institute in cooperation with USDA San Juan National Forest Service and USDOI Bureau of Land Management Tres Rios Field Office. Durango, CO. Available for download from: www.mountainstudies.org.

141. Rocca M.E., Brown P.M., MacDonald L.H., and C.M. Carrico. 2014. Climate change impacts on fire regimes and key ecosystem services in Rocky Mountain forests. Forest Ecology and Management 327: 290-305.

142. Romero-Lankao, P., J.B. Smith, D.J. Davidson, N.S. Diffenbaugh, P.L. Kinney, P. Kirshen, P. Kovacs, and L. Villers Ruiz. 2014. Chapter 26: North America. In: Climate Change 2014: Impacts, Adaptation, and Vulnerability. Part B: Regional Aspects. Contribution of Working Group II to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change [Barros, V.R., C.B. Field, D.J. Dokken, M.D. Mastrandrea, K.J. Mach, T.E. Bilir, M. Chatterjee, K.L. Ebi, Y.O. Estrada, R.C. Genova, B. Girma, E.S. Kissel, A.N. Levy, S. MacCracken, P.R. Mastrandrea, and L.L. White (eds.)]. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, pp. 1439-1498.

143. Romme, W.H., Allen, C.D., Bailey, J.D., [et al.]. 2009. Historical and modern disturbance regimes, stand structures, and landscape dynamics in piñon-juniper vegetation of the western United States. Rangeland Ecology and Management 62: 203–222.

144. Romme, W.H., C.D. Allen, J.D. Bailey, W.L. Baker, B.T. Bestelmeyer, P.M. Brown, K.S. Eisenhart, L. Floyd-Hanna, D.W. Huffman, B. Jacobs, R.F. Miller, E. Muldavin, T.W. Swetnam, R.J. Tausch, and P.J. Weisberg. 2008. Historical and modern disturbance regimes, stand structures, and landscape dynamics in piñon-juniper vegetation of the western U.S. Colorado Forest Restoration Institute, Colorado State University, Fort Collins, CO. 35 pp.

145. Romme, W.H., Floyd-Hanna, M.L., and D.D.Hanna. 2003. Ancient piñon-juniper forests of Mesa Verde and the West: A cautionary note for forest restoration programs. In, Proceedings of the conference on Fire, Fuel Treatments, and Ecological Restoration: Proper Place, Appropriate Time. USDA Forest Service, Colorado State University.

146. Ronco, F.P., Jr. 1990. Pinus edulis Engelm. pinyon. In: Burns, Russell M.; Honkala, Barbara H., technical coordinators. Silvics of North America. Volume 1. Conifers. Agric. Handb. 654. Washington, DC: U.S. Department of Agriculture, Forest Service: 327-337.

147. Rondeau, R., M. Bidwell, B. Neely, I. Rangwala, L. Yung, and C. Wyborn. 2017. Piñon-Juniper Landscape: San Juan Basin, Colorado Social-Ecological Climate Resilience Project. North Central Climate Science Center, Ft. Collins, Colorado.

148. Rosenstock, S.S. and C. Van Riper III. 2001. Breeding bird responses to juniper woodland expansion. Journal of Range Management: 226-232.

149. Roundy, B.A., Young, K., Cline, N., [et al.] 2014. Piñon-juniper reduction increases soil water availability of the resource growth pool. Rangeland Ecology and Management. 67: 495–505.

150. Rowe R.J., Finarelli J.A., and E.A.Rickart. 2010. Range dynamics of small mammals along an elevational gradient over an 80-year interval. Global Change Biology 16: 2930-2943.

151. Ryan K.C., and E.D.Reinhardt.1988. Predicting postfire mortality of 7 western conifers. Canadian Journal of Forest Research 18: 1291–1297.

152. Sauer, J.R., Hines, J.E., and J. Fallon. 2008. The North American breeding bird survey, results and analysis 1966 - 2007. Version 5.15.2008. Laurel, MD: USGS Patuxent Wildlife Research Center. Online: http://www.mbr-pwrc.usgs.gov/bbs/bbs.html. [2012, Oct 30].

153. Schlanger S., and S. Larralde. 2008. Why is cultural resource site density high in in the piñon-juniper woodland? In Gottfried, Gerald J., Shaw, John D., Ford, Paulette L., compilers. 2008 .Ecology, management, and restoration of piñon-juniper and ponderosa pine ecosystems: combined proceedings of the 2005 St. George, Utah and 2006 Albuquerque, New Mexico workshops. Proceedings RMRS-P-51.Fort Collins, CO: U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station.

154. Schoennagel T., and C.R. Nelson. 2011. Restoration relevance of National Fire Plan treatments across the western U.S. Frontiers in Ecology and the Environment 9: 271-277.

155. Schussman, H. and E. Smith. 2006. Historical range of variation and state and transition modeling of historic and current landscape conditions for potential natural vegetation types of the Southwest. Southwest Forest Assessment Project. The Nature Conservancy. Tucson, AZ.

156. Seager, R., and G. Vecchi. 2010. Greenhouse warming and the 21st century hydroclimate of southwestern North America. Proceedings of the National Academy of Science. U.S.A. 107: 21277– 21282.

157. Seager, R., et al. 2007. Model projections of an imminent transition to a more arid climate in southwestern North America. Science 316: 1181–

158. Senft R.L., Coughenour M.B., Bailey D.W., Rittenhouse L.R., Sala O.E., and D.M. Swift. 1987. Large herbivore foraging and ecological hierarchies. Bioscience 37: 789.

159. Settele, J., R. Scholes, R. Betts, S. Bunn, P. Leadley, D. Nepstad, J.T. Overpeck, and M.A. Taboada. Chapter 4: Terrestrial and inland water systems. In: Climate Change 2014: Impacts, Adaptation, and Vulnerability. Part A: Global and Sectoral Aspects. Contribution of Working Group II to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change [Field, C.B., V.R. Barros, D.J. Dokken, K.J. Mach, M.D. Mastrandrea, T.E. Bilir, M. Chatterjee, K.L. Ebi, Y.O. Estrada, R.C. Genova, B. Girma, E.S. Kissel, A.N. Levy, S. MacCracken, P.R. Mastrandrea, and L.L. White (eds.)]. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, pp. 271-359.

160. Shaw, J.D., B.E. Steed, and L.T. DeBlander. 2005. Forest Inventory and Analysis (FIA) Annual Inventory answers the question: what is happening to piñon-juniper woodlands? Journal of Forestry 103(6): 280-285.

161. Sheffield, J., and E.F. Wood. 2008. Projected changes in drought occurrence under future global warming from multi-model, multi-scenario, IPCC AR4 simulations. Climate Dynamics 31: 79–105.

162. Shinneman, D.J., and W.L. Baker. 2009a. Historical fire and multidecadal drought as context for piñon-juniper woodland restoration in Western Colorado. Ecological Applications 19(5): 1231–1245.

163. Shinneman, D.J., and W.L. Baker. 2009b. Environmental and climatic variables as potential drivers of post-fire cover of cheatgrass (Bromus tectorum) in seeded and unseeded semiarid ecosystems. International Journal of Wildland Fire. 18: 191–202.

164. Short, H.L., and C.Y. McCulloch. 1977. Managing pinyon-juniper ranges for wildlife. GTR-RM-47. Fort Collins, CO: U.S. Department of Agriculture, Forest Service, Rocky Mountain Forest and Range Experiment Station. 10 p.

165. Short, H.L., Evans, W., and E.L. Boeker. 1977. The use of natural and modified piñon pine-juniper woodlands by deer and elk. The Journal of Wildlife Management: 543-559.

166. Sthultz, C.M., Gehring, C.A. and T.G. Whitham. 2007. Shifts from competition to facilitation between a foundation tree and a pioneer shrub across spatial and temporal scales in a semiarid woodland. New Phytologist 173(1): 135-145.

167. Swetnam T.W., Farella J., Roos C.I., Liebmann M.J., Falk D.A., and C.D. Allen. 2016. Multiscale perspectives of fire, climate and humans in western North America and the Jemez Mountains, USA. Philosophical Transactions of the Royal Society B 371: 20150168. http://dx.doi.org/10.1098/rstb.2015.0168

168. Tausch, R.J., West, N.E., Shaw, D.W., Aldon, E.F. and C. LoSapio.1994. Plant species composition patterns with differences in tree dominance on a southwestern Utah piñon-juniper site. D.W. Shaw, E.F. Aldon, and C. LoSapio [tech. coords.]. Desired Future Conditions for Pinion-Juniper Ecosystems, pp.16-23.

169. Thompson, R.S., Hostetler, S.W., Bartlein, P.J., and K.H. Anderson. 1998. A strategy for assessing potential future changes in climate, hydrology, and vegetation in the western United States (No. 1153). U.S. Geological Survey circular (1153): 1-20.

170. Tilford, G.L. 1997. Edible and Medicinal Plants of the West. Mountain Press Publishing Company. ISBN 0-87842-359-1.

171. Triepke, F.J. 2017. Assessing the climate change vulnerability of the southwestern U.S. Dissertation. University of New Mexico

172. Trujillo E., Molotch N.P., Goulden M.L., Kelly A.E., and R.C. Bales. 2012. Elevation-dependent influence of snow accumulation on forest greening. Natural Geoscience 5: 705–9

173. Tuhy, J.S., P.Comor, D. Dorfman, M. Lammert, J. Humke, B. Cholvin, G. Bell, B. Nely, S. ?Silbert, L. Whitham, and B. Baker. 2002. A conservation assessment of the Colorado Plateau Ecoregion. The Nature Conservancy, Moab Project Office, Moab UT. 110p +appendeces.http://fsweb.r3.fs.fed.us/eap/assessments/_c_colorado_plateau_ecoregion_assessme nt/Plan%2BApp.pdf

174. U.S. Bureau of Land Management (BLM). 2011. Notice of intent to prepare environmental impact statements and supplemental environmental impact states to incorporate greater sage grouse conservation measures in land use and land management plans. Notice of Intent, Federal Register, pp 76: 77008-77011.

175. U.S. Department of Agriculture (USDA) Forest Service. 2011. Navigating the Climate Change Performance Scorecard. A Guide for National Forests and Grasslands, Version 2. Available at: http://www.fs.fed.us/climatechange/advisor/scorecard/scorecard-guidance-082011.pdf.

176. U.S. Environmental Protection Agency (USEPA). 2013. Level III ecoregions of the continental United States: Corvalis, OR: U.S Environmental Protection Agency, National Health and Environmental Research Laboratory. Map scale 1:750,000. https://www.epa.gov/eco-research/level-iii-and-iv- ecoregions-continental-united-states. [Accessed January 16, 2017].

177. U.S. Geological Survey (USGS). 2011. National Gap Analysis Program (GAP)-Core Science Analytics and Synthesis. National Land Cover, Version 2. Washington, DC: U.S. Department of the Interior, U.S. Geological Survey. https://gapanalysis.usgs.gov/data/. [Accessed January 16, 2017].

178. Utah State, Department of Natural Resources. 2015. Utah sensitive species list. Salt Lake City, UT: 150 p.

179. Utah Wildlife Action Plan Joint Team. 2015. Utah Wildlife Action Plan: A plan for managing native wildlife species and their habitats to help prevent listing under the Endangered Species Act. Publication number 15-14. Utah Division of Wildlife Resources, Salt Lake City, Utah, USA.

180. Valdez, E.W., and P.M. Cryan. 2009. Food Habits of the Hoary Bat (Lasiurus cinereus) During Spring Migration Through New Mexico. The Southwestern Naturalist 54(2): 195–200.

181. Vander Wall, S.B., Hoffman, S.W., and W.K. Potts. 1981. Emigration behavior of Clark's nutcracker. The Condor 83: 162-170.

182. Waichler, W.S.; Miller, R.F.; Doescher, P.S. 2001. Community characteristics of old-growth western juniper woodlands. Journal of Range Management. 54:518–527.

183. Waring, K.M., D.M. Reboletti, L.A. Mork, C. Huang, R.W. Hofstetter, A.M. Garcia, P.Z. Fulé, and T.S. Davis. 2009. Modeling the impacts of two bark beetle species under a warming climate in the southwestern USA: ecological and economic consequences. Environmental Management 44(4): 824- 835.

184. Weisberg, P.J. Lingua, E. and R.B. Pillai. 2007. Spatial patters on Piñon-juniper woodland expansion in central Nevada. Rangeland Ecology and Management 60: 115-124.

185. Westerling, A. L., H. G. Hidalgo, D. R. Cayan, and T. W. Swetnam. 2006. Warming and earlier spring increase western U.S. forest wildfire activity. Science 313:940- 943.

186. Westerling, A.; Brown, T.; Schoennagel, T.; [et al.]. 2014. Briefing: Climate and wildfire in western U.S. forests. In: Sample, V.A.; Bixler, R.P., eds. Forest conservation and management in the Anthropocene: Conference proceedings. Proceedings. RMRS-P-71. Fort Collins, CO: U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station: 81–102.

187. Wilcox, B. P., D. D. Breshears, and C. D. Allen. 2003. Ecohydrology of a resource-conserving semiarid woodland: Effects of scale and disturbance. Ecological Monographs 73(2):223-239.

188. Wilcox, B.P. 1994. Runoff and erosion in intercanopy zones of piñon-juniper woodlands. Journal of Range Management. 47: 285-295.

189. Williams, A.P., Allen, C.D., Millar, C.I., [et. al]. 2010. Forest responses to increasing aridity and warmth in the southwestern United States. Proceedings of the National Academy of Sciences U.S.A. 107(50): 21289–21294.

190. Ziska, L.H., Reeves, J.B., and B. Blank. 2005. The impact of recent increases in atmospheric CO2 on biomass production and vegetative retention of cheatgrass (Bromus tectorum): implications for fire disturbance. Global Change Biology 11: 1325-1332.

191. Zlatnik, E. 1999. Juniperus osteosperma. In: Fire Effects Information System, [Online]. U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station, Fire Sciences Laboratory (Producer). Available: https://www.fs.fed.us /database/feis/plants/tree/junost/all.html [2018, August 30].

192. Zouhar, K.L. 2001. Pinus monophylla. In: Fire Effects Information System, [Online]. U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station, Fire Sciences Laboratory (Producer). Available: http://www.fs.fed.us/database/feis/

Appendix 1. Additional Resources To facilitate coordination among SRLCC stakeholders, we list additional resources for PJ assessment and study (Table 1). Activities represent either currently active working groups, collaborations, research or other projects and spatial datasets with information relevant to the assessment of PJ ecosystems.

Table 1. Online Resources and data sets of interest to PJ research Title Geographic Description Link scope HabiMap Arizona- Arizona only User-friendly, web-based http://www.habimap.org/hab "Mapping wildlife and data viewer for imap/ conservation potential" information contained within the State Wildlife

Action Plan available to anyone interested in Arizona's wildlife. Historical Range of Variation Arizona and Produced for USDA by The Gori and Bate, 2007. and State and Transition New Mexico Nature Conservancy http://azconservation.org/dl/ models Report describes HRV for TNCAZ_SWFAP_HRV_Piñon_J SW ecosystems based on uniper.pdf empirical information Forest Inventory and Analysis Nation wide Field sampled vegetation https://www.fia.fs.fed.us/tool (FIA) data National Program data (sampled every 5km) s-data/ and reporting tools Geographic distribution map North America Monthly climate surfaces Cole et al. 2008 (http://www for Piñon Juniper created (1 km scale) using .usgs.nau.edu / data from the Global global_change) Historical Climatology Network. High-resolution maps of Nevada and a Geospatial data for object- Coates et al. 2017. conifers portion of based high-resolution https://www.sciencebase.gov northeastern classification of conifers /catalog/item/59160b60e4b0 California within greater sage-grouse 44b359e32e67 habitat across Nevada and a portion of northeastern California The Desert Laboratory Varies Largest collection of Webb et al. 2007. Contact Repeat Photography repeat photography Meredith Hartwell at Collection, [email protected] if you need further information about the collection. http://pubs.usgs.gov/fs/2007/ 3046/fs2007-3046.pdf Ecoadapt assessment for Southern Climate change http://ecosadapt.org/library focal habitats (including California vulnerability assessment

Piñon Juniper woodlands) for focal habitat and climate change adaptation strategies Shaw et al. 2015 Climate Change Vulnerability Colorado Climate Change Decker et al. 2015. Colorado Assessment for Colorado Vulnerability Assessment Natural Heritage Program Bureau of Land Management for Colorado Bureau of [CNHP]. Land Management.

Piñon-Juniper landscapes: Colorado Framework and adaptation Rondeau et al. 2017 San Juan Basin, Colorado strategies for PJ under 3 Social-Ecological Climate climate scenarios Resilience Project

PJ symposium Range of PJ Presentations on topics of https://nhnm.unm.edu/P- current status, climate JSymposium change impacts, wildlife, information gaps The Piñon-juniper webzone Intermountain Reference materials for Oregon State university, USGS West including piñon and juniper species http://oregonstate.edu/dept/ Texas EOARC/piñon-juniper/w.html USDA Forest Service New Mexico/ Spatially explicit analysis of Treipke et al. 2017. Reports Southwest Region Arizona ecosystem vulnerability available on a per forest basis. Vulnerability Assessment based on NRV and climate Contact Jack Triepke for more projections information.

Appendix 2. Indicator Datasets The following describes the underlying datasets used to measure exposure, sensitivity and adaptive capacity for PJ woodlands in two focal areas within the Southern Rockies Landscape Conservation Cooperative for the final report “Spatially Explicit Vulnerability Assessment of Piñon-Juniper Ecosystems in the Four Corners and Upper Rio Grande Landscapes.” All data is available online (https://www.sciencebase.gov/catalog/item/5693e56ee4b09c7f9a21a41d) to facilitate additional analyses and assessments identified by users. Exposure 1. Temperature Source: Rehfeldt, Gerald E. 2006. A spline model of climate for the Western United States. Gen. Tech. Rep. RMRS-GTR-165. Fort Collins, CO: U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station. 21 p. (available at: http://charcoal.cnre.vt.edu/climate/future/). Description: Coded value - HUC12 with > 1.8°C increase (mean change for year 2030) in mean maximum of the warmest month for RCP4.5, year 2030. Data used comprised Ensembles of 17 climate models under the RCP 4.5 for year 2030, which is an average of conditions for the period 2020-2040. Justification: Williams et al. (2013) identified a forest drought-stress index (FDSI) for the southwestern U.S. that is based on warm-season vapor pressure deficit (VPD) and cold-season precipitation. They present strong evidence for the influence of growing season VPD (driven by temperature) and winter precipitation on drought mortality within Pinus edulis (and two other conifers). In particular, VPD was the main driver of the FDSI even if cold season precipitation remains at current levels. Higher temperatures which affect VPD can lead to hydraulic failure or starvation of tree species, thereby causing intense droughts, where hot droughts result in greater mortality. Furthermore, Redmond et al. (2012) noted that areas that experience the greatest increase in growing season temperatures experienced the greatest decline in seed cone production. Increased late summer temperature may have led to a 40% reduction in seed cone productions in piñon pine (Redmond et al. 2012). Similarly, Barger et al. (2009) noted strong correlations between piñon-juniper growth, winter precipitation and June temperature. Finally, temperature also effects the overwintering survival of Ips beetles (Meddens et al. 2015). Data Compilation: Calculated the predicted change in mean maximum temperature for the warmest month from current for the 20 year period 2020-2040 (2030) using Raster calculator. For each HUC12, assigned a value of ‘1’ where HUC12 > 1.8°C increase for RCP 4.5, year 2030. Uncertainty: A climate scenario is a plausible representation of a potential future climate. The bioclimatic models used to predict future climate are based on general circulation models (CGM) and predicted emission scenarios that may or may not occur. The accuracy of predictions from bioclimatic models ultimately is dependent on the aptness of the scenarios and the precision of the GCMs. Significant uncertainties in any future climate prediction remain, especially regarding the future course of human population and development (IPCC, 2013).

2. Decline in Winter Precipitation Source: Rehfeldt, Gerald E. 2006. A spline model of climate for the Western United States. Gen. Tech. Rep. RMRS-GTR-165. Fort Collins, CO: U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station. 21 p. (available at: http://charcoal.cnre.vt.edu/climate/future/). Data used comprised Ensembles of 17 climate models under the RCP 4.5 for year 2030, which is an average of conditions for the period 2020-2040. Description: Coded value - HUC12 with < 100 percent of normal precipitation of the winter months (Nov – Feb) for RCP4.5, year 2030. Data used comprised Ensembles of 17 climate models under the RCP 4.5 for year 2030, which is an average of conditions for the period 2020-2040. Justification: Williams et al. (2013) identified a forest drought-stress index (FDSI) for the southwestern U.S. that is based on warm-season vapor pressure deficit (VPD) and cold-season precipitation. They present strong evidence for the influence of growing season VPD (driven by temperature) and winter precipitation on drought mortality within Pinus edulis (and two other conifers). In particular, VPD was the main driver of the FDSI even if cold season precipitation remains at current levels. Similarly, Barger et al. (2009) noted strong correlations between piñon-juniper growth, winter precipitation and June temperature. Data Compilation: Calculated the change in precipitation values as a percent of the normal (current) precipitation for the 20 year period 2020-2040 (2030) using Raster calculator. For each HUC12, assigned a value of ‘1’ where HUC12 was < 100 percent of normal precipitation for the winter months (Nov – Feb) for RCP 4.5, year 2030. Uncertainty: A climate scenario is a plausible representation of a potential future climate. The bioclimatic models used to predict future climate are based on general circulation models (CGM) and predicted emission scenarios that may or may not occur. The accuracy of predictions from bioclimatic models ultimately is dependent on the aptness of the scenarios and the precision of the GCMs. Significant uncertainties in any future climate prediction remain, especially regarding the future course of human population and development (IPCC, 2013).

3. Change in Development Source: U.S. Geological Survey. 2005. Conterminous United States Land-Use/Land-Cover (CONUS LULC) for historical year and projected years through 2100 for IPCC-SRES scenarios A2 and A1B (available at : http://landcover-modeling.cr.usgs.gov/) Description: Coded value – HUC12 with expected increase in developed land cover classes for year 2040 under the A1B climate scenario. The scenario-construction process used by USGS incorporated input from an integrated modeling framework to provide top-down proportions of land use/land cover change, with overall socioeconomic assumptions from the IPCC SRES scenarios driving differences between projected scenarios. Within the southwest, agreement was high for trends of change. This assessment used values generated by comparing the difference in the proportion of land classification between year 2005 and 2040 under an A1B emission scenario. Areas are characterized by a high percentage (20% or greater) of constructed material (concrete, asphalt, buildings, etc.). Justification: Development of any kind that involves land clearing is a threat to PJ ecosystems. Invasive species are more likely in disturbed areas. Modification of riparian areas because of water withdrawals,

habitat fragmentation, and pollution from nearby development and industries threaten PJ ecosystems. As urban centers grow over time, they may fragment or eliminate nearby PJ ecosystems. Data Compilation: Calculated the area of those land cover classes that represent development, including: ‘Developed’, ‘Mechanically Disturbed National Forests’, ‘Mechanically Disturbed Other Public Lands’, ‘Mechanically Disturbed Private’, ‘Cropland’, and ,’Hay/Pasture Land’ per HUC 12 watershed. Then, calculated the percent change in developed land cover from current for the year 2040 under the A1B climate scenario. For each HUC12, assigned a value of ‘1’ where HUC12 > 0 percent increase in development for year 2040. Uncertainty: Many factors determine how human land use and natural disturbance modify the earth’s landscape. Land-cover change is inherently a local event. Projecting future land cover must account for both local ("bottom-up") to global ("top-down") drivers of potential land use change resulting in a high level of uncertainty associated with predicting future development on the landscape and may represent a wide range of plausible future conditions.

4. Road Density Source: U.S. Census Bureau. 2016. 2016 TIGER/Line Shapefiles: Roads. (Available at: http://www.census.gov/cgi-bin/geo/shapefiles/index.php?year=2016&layergroup=Roads). Description: Coded value – HUC12 with primary and secondary road density values in the upper 25 percentile (>0.10). Data contains the density of mapped linear road features for primary and secondary roads. Primary roads are generally divided limited-access highways within the Federal interstate highway system or under state management. Secondary roads are main arteries, usually in the U.S. highway, state highway, or county highway system. Justification: Roads can fragment PJ ecosystems and can cause these systems and wildlife that depend on them to be vulnerable. Roads can represent a barrier to movement and contribute to a focal resources’ sensitivity to disturbance. Furthermore, vehicle collisions can lead to direct mortality of wildlife and can block their normal behavior patterns. As intact patches of PJ shrink, wildlife become increasingly vulnerable to predation and isolation (Watson, 2005). Data Compilation: Calculated the line density of primary and secondary roads within each HUC12 watershed. HUC12 watersheds within the top 25th percentile for road density were coded ‘1’. Uncertainty: Effect of road crossings varies among road surface types, road construction methods, and stream crossing types (bridge, culvert, etc.). Influences of these types should be examined at the local scale.

5. High to Very High Fire Potential Source: Dillon, Gregory K. 2015. Wildfire Hazard Potential (WHP) for the conterminous United States (270-m GRID), version 2014 continuous. Fort Collins, CO: Forest Service Research Data Archive. https://doi.org/10.2737/RDS-2015-0047. (Available at: https://www.firelab.org/project/wildfire-hazard- potential). Description: Coded value – HUC12 with some areas at high or very high wildfire hazard potential. The WHP geospatial data depicts the relative potential for wildfire, as well as wildfire likelihood and intensity. Areas mapped with higher WHP values represent fuel conditions with a higher probability of

experiencing torching, crowning, and other forms of extreme fire behavior under conducive weather conditions. Justification: Wildland fire is a threat to Piñon-Juniper ecosystems; however the impacts may differ across the range of PJ functional types (Margolis, 2014). Factors that affect the fire regime of these ecosystems are driven by a site’s climate, stand density, and existing plant community. Still, climate change may increase fire frequency across functional types due to the overall increases in temperature, length of wildfire season, and severity of fire weather (Westerling et al. 2006). Though wildfires have naturally occurred in PJ ecosystems, PJ is slow to recover from high severity fire, and fire is a primary cause of loss of ecologically important mature PJ woodlands (Huffman et al. 2008). Areas that experience increased fire are likely to transition to grasslands or shrublands as post-fire recovery is very slow, particularly where there is a lack of precipitation (Rocca et al. 2014; Romme et al. 2003). Data Compilation: Calculated the percent of total area classified as having High or Very High wildfire hazard potential (WHP) within each HUC12 watershed. HUC12 > 0 percent High or Very High WHP were coded ‘1’. Uncertainty: The WHP geospatial data is not an explicit map of wildfire threat or risk, but only spatially approximates relative wildfire risk to ecological resources or values-at-risk. Furthermore, although increases in negative fire behavior is well documented in some areas, it may not be the case in all PJ ecosystems or functional types and may ignore beneficial fire.

6. Insect and Disease Threats Source: Weidner, E. & Todd, A. 2011. From the forest to the faucet: Drinking water and forests in the US. Washington, DC: USDA Forest Service. (Available at: https://www.fs.fed.us/ecosystemservices/FS_Efforts/forests2faucets.shtml). Description: Coded value – HUC12 containing forest identified as at high risk of insect and disease related mortality. Data was derived from the National Insect and Disease Risk Map (NIRDM) created by the Forest Health Technology Enterprise Team (FHTET). NIRDM quantifies threats due to insects and disease across the U.S. FHTET defines a threshold for mapping high risk as: “the expectation that, without remediation, 25 percent or more of the standing live basal area (BA) of trees greater than 1 inch in diameter will die over the next 15 years (starting in 2005) due to insects and diseases” (Krist et al. 2014). Justification: Collectively, insect and disease are leading contributors to the risk of mortality of tree species. Wide-spread losses from pests or pathogens in susceptible areas can reach epidemic levels, weakening species and causing further mortality from fire, invasive species, drought, browsing, extreme weather events, etc. (Krist et al. 2014). Data Compilation: The Forests to Faucets layer is spatially subdivided by HUC12 watershed already with attribute “Percent of HUC highly threatened by Insects and Disease (using FHTET National Insect and Disease Risk Map)”. Those HUC12 > 0 percent of insect and disease related mortality were coded ‘1’. Uncertainty: As the NIRDM dataset is a compilation of many data sets, data completeness, accuracy, and scale may vary. Furthermore, insect disturbance is often patchy and not uniformly distributed across forested stands, and so not all mapped areas may be affected by insects (Peterman et al. 2013).

7. Loss of Climate Suitability (Great Basin Conifer Woodlands) Source: Rehfeldt, Gerald E.; Crookston, Nicholas L.; Saenz-Romero, Cuauhtemoc; Campbell, Elizabeth M. 2012. North American vegetation model for land-use planning in a changing climate: A solution to large classification problems. Ecological Applications. 22(1): 119-141 - Supplemental Materials and raster data (Available at: https://forest.moscowfsl.wsu.edu/climate/publications.php).

Description: Coded value – HUC12 with greater than 30% loss in suitable climates for Great Basin Conifer Woodlands (GBCW) type biome in projection year 2030. GBCW is described as a “cold-adapted evergreen woodland [that] is characterized by the unequal dominance of two conifers – juniper (Juniperus) and piñon (Pinus). These trees rarely, if ever, exceed 12 m height and are typically openly spaced (woodland), except at higher elevations and other less xeric sites where interlocking crowns may present a closed (forest) aspect. The shorter, bushier junipers ("cedars") are generally more prevalent than piñons, but either may occur as an essentially pure stand. Structurally, these juniper-piñon woodlands are among the simplest communities in the Southwest. This woodland has its evolutionary center in the Great Basin and is one of the most extensive vegetative types in the Southwest.” Justification: More common and severe drought conditions over the past decade have weakened pinyon-juniper species and have made these ecosystems more susceptible to mortality from insect and disease as well as wildfire (Shaw, 2006). Due to the increasing number of invasive species colonizing these woodland sites, succession may be dramatically altered post-fire, preventing native grasses and forbs from occupying the site and reducing the ability of PJ to regenerate (Gori et al. 2007). Multiple climate models predict that temperatures in the Southwest will continue to increase while precipitation decreases, potentially leading to widespread range retractions and local extirpation (Allen, 2007; IPCC, 2013) Data Compilation: Calculated the change from current biome distribution of GBCW for the projection year 2030 using Raster calculator. For each HUC12, we assigned a value of ‘1’ where there was greater than a 30% predicted decrease in total area of suitable climate for Great Basin Conifer Woodlands. Uncertainty: A climate scenario is a plausible representation of a potential future climate. The bioclimatic models used to predict future climate suitability envelopes are based on general circulation models (CGM) and predicted emission scenarios that may or may not occur. The accuracy of predictions from bioclimatic models ultimately is dependent on the aptness of the scenarios and the precision of the GCMs. Significant uncertainties in any future climate prediction remain, especially regarding the future course of human population and development (IPCC, 2013). Other physical, biological or socio-economic events may possibly be more important drivers of the habitat suitability of GBCW habitat in the future. Sensitivity

1. Presence of Energy Development Source: 1) ESRI & the FracTrackerAlliance. 2015. Oil and gas wells in the US (2015). (Available at: https://www.arcgis.com/home/item.html?id=49102e45079445fabdb9b5c0679d96ee). 2) New Mexico State Land Office. 2017. Active oil and gas leases on New Mexico State trust lands. (Available at: http://landstatus.nmstatelands.org/GISDataDownLoadnew.aspx).

Description: Coded value – HUC12 with any oil or gas wells where “Status” is active within the boundary. This layer contains all active oil and gas wells in the United States by state. The well data were obtained from each state's oil and gas regulatory agency, with data collected between March and May of 2015. Justification: Land conversion from PJ ecosystems to other types of land cover or development pose a major threat to the ecosystem. PJ ecosystems have been significantly impacted by energy development, specifically oil and gas extraction, surface coal mining, and solar and wind development, and associated infrastructure (powerlines, pipelines, and roads, etc.)(NMDFG, 2006). Oil and gas impacts on wildlife and the environment include habitat loss, wildlife mortality and displacement, and introduction of invasive species (Ramirez et al. 2015). Data Compilation: Calculated the number of oil and gas wells where “Status” is Active within each HUC12 watershed. Oil and gas wells were considered Active, when “Status” included: 'Active' , 'Approved Permit' , 'Domestic Well' , 'Drilling' , 'Gas Storage (liquefied gas)' , 'Injecting' , 'New' , 'New Permit' , 'Permitted Location' , 'Producing Gas Well', 'Producing Oil Well', 'Shut in Oil Well' , 'Water Disposal Well (converted from oil well)' , 'Water Well', or similar; HUC12 > 0 oil or gas wells where “Status” is active were coded ‘1’. Uncertainty: The effects of oil and gas wells on PJ ecosystem depends on the intensity and extent of resource extraction and associated infrastructure as well as oversight and management of oil and gas activities. Also, as the oil and gas wells dataset is a compilation of many data sets, data completeness, accuracy, and scale may vary.

2. Current Development Source: Homer, C.G., Dewitz, J.A., Yang, L., Jin, S., Danielson, P., Xian, G., Coulston, J., Herold, N.D., Wickham, J.D., and Megown, K., 2015, Completion of the 2011. National Land Cover Database for the conterminous United States-Representing a decade of land cover change information. Photogrammetric Engineering and Remote Sensing, v. 81, no. 5, p. 345-354. (Available at: https://www.mrlc.gov/nlcd11_data.php). Description: Coded value – HUC12 with any percent area classified as medium or high intensity development. The National Land Cover Database (NLCD) uses Landsat imagery to quantitatively determine the percent impervious surface area or urban land cover extent at the 30-m grid cell. Developed, medium intensity is defined as “areas with a mixture of constructed materials and vegetation. Impervious surfaces account for 50% to 79% of the total cover. These areas most commonly include single-family housing units.” Developed, high intensity is defined as “highly developed areas where people reside or work in high numbers. Examples include apartment complexes, row houses and commercial/industrial. Impervious surfaces account for 80% to 100% of the total cover.” Justification: The distribution and amount of urban land cover has a significant impact on Piñon-Juniper ecosystems. Development of any kind that involves land clearing is a threat to PJ ecosystems, and fragmentation of PJ landscape is of concern (Gottfried et al. 1995). Areas that have already been developed, much of which has already been altered for agriculture, or road and infrastructure, present challenges for PJ habitats and associated wildlife.

Data Compilation: Calculated the current percent area of those land cover classes that represent high levels of development, including: “Developed, Medium Intensity” and, “Developed, High Intensity” per HUC 12 watershed. For each HUC12, assigned a value of ‘1’ where HUC12 > 0 percent of area currently classified as developed, high or medium intensity. Uncertainty: National Land Cover Database (NLCD) data products are created at a 30-meter grid spatial resolution raster data set, and are a consistent and scientifically reliable set of mapped fire and vegetation characteristics to be used for course-scale, national and regional planning; however, they should generally not be used at the pixel level. The appropriate use of these products for local land management planning varies by product, location, and the specific use, and data users should assess these products based on their specific needs. The use of this data for this project is scaled regionally, however, it is likely that some HUC12 watersheds are misrepresented as having or not having developed land cover.

3. Mechanically Disturbed Forests Source: LandCarbon Conterminous United States Land-Use/Land-Cover (CONUS LULC) for historical year and projected years through 2100 for IPCC-SRES scenarios A2 and A1B (available at: http://landcover- modeling.cr.usgs.gov/) Description: Coded value – HUC12 with mechanically disturbed areas. Area (km2) of HUC Classified in 2005 as Mechanically Disturbed National Forests, Mechanically Disturbed Other Public Lands, Mechanically Disturbed Private Land. Justification: Development of any kind that involves land clearing is a threat to PJ ecosystems, and fragmentation of PJ landscape is of concern (Gottfried et al. 1995). Areas that have experience mechanical disturbance (e.g. thinning), present challenges for PJ habitats and associated wildlife. Data Compilation: Calculated the area of those land cover classes that represent development, including: ‘Mechanically Disturbed National Forests’, ‘Mechanically Disturbed Other Public Lands’, ‘Mechanically Disturbed Private’ per HUC 12 watershed. Calculated the total area (km2) of each Value in the Percent Mechanically disturbed Layer for each HUC12 watershed. HUC12 with mechanically disturbed areas were coded ‘1’. Uncertainty: Many factors determine how human land use and natural disturbance modify the earth’s landscape. Land-cover change is inherently a local event. Projecting future land cover must account for both local ("bottom-up") to global ("top-down") drivers of potential land use change resulting in a high level of uncertainty associated with predicting future development on the landscape and may represent a wide range of plausible future conditions.

4. Piñon-Juniper Habitat Source: LANDFIRE, 2014, Existing Vegetation Type Layer, LANDFIRE 1.4.0, U.S. Department of the Interior, Geological Survey (Available at: https://www.landfire.gov/version_comparison.php.) Description: Coded value – HUC12 with <30% cover of Piñon-Juniper habitat types. The existing vegetation type (EVT) data layer represents the current distribution of the terrestrial ecological systems classification developed by NatureServe for the western Hemisphere. A terrestrial ecological system is

defined as a group of plant community types (associations) that tend to co-occur within landscapes with similar ecological processes, substrates, and/or environmental gradients. Justification: The presence of nurse trees of either piñon or juniper species is associated with a greater abundance of seedlings and saplings after mortality events (Floyd et al. 2015) and may be important for maintaining recruitment after drought events (Sthultz et al. 2007). Data Compilation: Calculated the area of those existing vegetation types that represent Piñon-Juniper plant community types, including: 3011 – ‘Rocky Mountain Aspen Forest and Woodland’, 3016 – ‘Colorado Plateau Piñon-Juniper Woodland’, 3019 – ‘Great Basin Piñon-Juniper Woodland’, 3025 – ‘Madrean Piñon-Juniper Woodland’, 3049 – ‘Rocky Mountain Foothill Limber Pine-Juniper Woodland’, 3059 – ‘Southern Rocky Mountain Piñon-Juniper Woodland’, 3115 – ‘Inter-Mountain Basins Juniper Savanna’, 3116 – ‘Madrean Juniper Savanna’, 3119 – ‘Southern Rocky Mountain Juniper Woodland and Savanna’. Second, calculated the total percent area of the combined Piñon-Juniper plant community types per HUC12 watershed. For each HUC12, assigned a value of ‘1’ where HUC12 < 30% cover of Piñon-Juniper Woodlands. Uncertainty: LANDFIRE data products are created at a 30-meter grid spatial resolution raster data set, and are a consistent and scientifically reliable set of mapped fire and vegetation characteristics to be used for course-scale, national and regional planning; however, they should generally not be used at the pixel level. The appropriate use of LANDFIRE products for local land management planning varies by product, location, and the specific use, and data users should assess these products based on their specific needs. The use of this data for this project is scaled regionally, however, it is likely that some HUC12 watersheds are misrepresented as having or not having an associated piñon-juniper plant community association.

5. Wildlife Diversity Source: U.S. Geological Survey. 2018. U.S. Geological Survey – Gap Analysis Project Species Range Maps CONUS_2001: U.S. Geological Survey data release, https://doi.org/10.5066/F7Q81B3R. (Available at: https://gapanalysis.usgs.gov/species/data/download/). Data downloaded for 17 species (black-chinned hummingbird, Bewick's wren, black throated grey warbler, black-throated sparrow, bushtit, Cassin's kingbird, fringed myotis, Gunnison's prairie dog, gray flycatcher, hoary bat, Juniper titmouse, Lewis's woodpecker, Mexican gartersnake, narrow-headed gartersnake, Pinon jay, Scott's oriole, spotted bat. Description: Coded value – HUC12 with at least 12 Piñon-Juniper associated species. GAP species range data constitute a course-scale total area of suitable habitat for a species or where a species is likely to be found. Justification: Persistent PJ forests include important habitat for wildlife species. PJ habitat that are able to support a large number of associated wildlife species holds more value than habitats with relatively few species. Data Compilation: Calculated the number of species distribution occurrence (i.e. having any area with suitable habitat for a species) per HUC12. HUC12 with > 12 species were coded ‘1’. Uncertainty: GAP acknowledges that these distribution maps are appropriate for National, regional or statewide biodiversity or conservation planning, county-level comprehensive planning or other coarse- filter applications such as evaluation of potential impacts of utility or transportation corridors, wilderness proposals, habitat connectivity proposals, climate change adaption proposals, regional open

space and recreation proposals, etc. on wildlife habitat. The GAP program recommends using data to map small areas (less than thousands of hectares or tens of kilometers), typically requiring mapping resolution at 1:24,000 scale and using aerial photographs or ground surveys. Many HUC12 watersheds are smaller than the indicated area. However, errors carried over from GAP estimates of species presence are minimized as we use distribution maps to designate each HUC12 watershed not to identify specific areas where species exists. Further, we overlay several datasets. Still, for underlying datasets, it is likely that some HUC12 are misrepresented as having or not having species.

6. Available Water Storage Capacity Source: Peterman, Wendy & Ferschweiler, Ken. 2015. A case study for evaluating potential soil sensitivity in arid land systems. Integrated Environmental Assessment and Management. 10.1002/ieam.1691. (Available at: https://databasin.org/maps/0215e43355f442e5b62679fc362a3055 From: Low Available Water Storage Capacity _AWC _0.10_ _ Southern Rockies LCC _ South). Description: Coded value – HUC12 with any soils with low available water capacity (AWC) <0.10. The dataset was compiled from data produced by the National Cooperative Soil Survey by the Conservation Biology Institute for the purpose of a soil vulnerability study. These data were created by field inventory of soils and mapping repeatable known patterns of soils that occur commonly across landscape features. Justification: Site characteristics such as soil texture and depth can influence soil water content and water availability for trees (Meddens et al. 2015). Peterman et al. (2013) was able to relate drought related mortality to soil water storage capacity. Their study found that 70% of mortality during 2003 and 2004 drought events were recorded for areas with an available soil water capacity of <100mm (84% where AWC <150mm). Soil texture tended to vary widely for areas with similar AWC and was indicated as probably less important. Specifically, water holding capacity appears to be important for piñon species’ recruitment (Redmond et al. 2015). Piñon regeneration was greatest where soil available water capacity was highest in sites in both Colorado (Redmond and Barger, 2013) and Arizona (Redmond et al. 2015). Additionally, Redmond et al. (2015) notes that mature piñon mortality was higher in areas with lower soil available water capacity. Data Compilation: Calculated the area within each HUC12 of those areas with low AWC <0.10. For each HUC12, we assigned a value of ‘1’ where there was any area with soils with low AWC. Uncertainty: The soil data comes from a digital soil survey, the most detailed level of soil geographic data developed by the National Cooperative Soil Survey. The soil map units are linked to attributes in the National Soil Information System relational database, which gives the proportionate extent of the component soils and their properties. However, large gaps in the Soil Survey Geographic Database (SSURGO) soil data layers are known to exist. In order to map soils across the U.S., data is interpolated at course scale, so large gaps between known soil survey data may increase uncertainty in the data. Finally, not all studies of piñon mortality events have found a relationship between soil properties and mortality severity (Meddens et al. 2015).

Adaptive Capacity 1. Increase in Suitable Climate Source: Rehfeldt, Gerald E.; Crookston, Nicholas L.; Saenz-Romero, Cuauhtemoc; Campbell, Elizabeth M. 2012. North American vegetation model for land-use planning in a changing climate: A solution to large classification problems. Ecological Applications. 22(1): 119-141 - Supplemental Materials and raster data (Available at: https://forest.moscowfsl.wsu.edu/climate/publications.php). Description: Coded value – HUC12 with greater than 30% area gain in suitable climate in Great Basin Conifer Woodland (GBCW) type biome in projection year 2030. GBCW is described as a “cold-adapted evergreen woodland [that] is characterized by the unequal dominance of two conifers – juniper (Juniperus) and piñon (Pinus). These trees rarely, if ever, exceed 12 m height and are typically openly spaced (woodland), except at higher elevations and other less xeric sites where interlocking crowns may present a closed (forest) aspect. The shorter, bushier junipers ("cedars") are generally more prevalent than piñons, but either may occur as an essentially pure stand. Structurally, these juniper-piñon woodlands are among the simplest communities in the Southwest. This woodland has its evolutionary center in the Great Basin and is one of the most extensive vegetative types in the Southwest.” Justification: More common and severe drought conditions over the past decade have weakened pinyon-juniper species and have made these ecosystems more susceptible to mortality from insect and disease as well as wildfire (Shaw, 2006). Due to the increasing number of invasive species colonizing these woodland sites, succession may be dramatically altered post-fire, preventing native grasses and forbs from occupying the site and reducing the ability of PJ to regenerate (Gori et al. 2007). Multiple climate models predict that temperatures in the Southwest will continue to increase while precipitation decreases, potentially leading to widespread range retractions and local extirpation (Allen, 2007; IPCC, 2013). Those area where suitable PJ climate remains will become increasingly important in sustaining populations. Data Compilation: Calculated the change from current biome distribution of GBCW for the projection year 2030 using Raster calculator. For each HUC12, we assigned a value of ‘1’ where there was greater than a 30% predicted increase in total area of suitable climate for Great Basin Conifer Woodlands. Uncertainty: A climate scenario is a plausible representation of a potential future climate. The bioclimatic models used to predict future climate suitability envelopes are based on general circulation models (CGM) and predicted emission scenarios that may or may not occur. The accuracy of predictions from bioclimatic models ultimately is dependent on the aptness of the scenarios and the precision of the GCMs. Significant uncertainties in any future climate prediction remain, especially regarding the future course of human population and development (IPCC, 2013). Other physical, biological or socio-economic events may possibly be more important drivers of the habitat suitability of GBCW habitat in the future.

2. Protected Land Designation Source: U.S. Geological Survey, Gap Analysis Program (GAP). May 2016. Protected Areas Database of the United States (PAD-US), version 1.4 Combined Feature Class (Available at: https://gapanalysis.usgs.gov/padus/data/download/). Description: Coded value – HUC12 with greater than 30% area with protected GAP designations. An area was designated as protected when it had a GAP status code or ‘1’ or ‘2’ meaning it is managed to

maintain a natural state. The USGS defines the GAP status codes as “a measure of management intent to conserve biodiversity defined as: Status 1: An area having permanent protection from conversion of natural land cover and a mandated management plan in operation to maintain a natural state within which disturbance events (of natural type, frequency, intensity, and legacy) are allowed to proceed without interference or are mimicked through management. Status 2: An area having permanent protection from conversion of natural land cover and a mandated management plan in operation to maintain a primarily natural state, but which may receive uses or management practices that degrade the quality of existing natural communities, including suppression of natural disturbance.” Justification: Fragmentation of PJ landscape is of concern (Gottfried et al. 1995). Large, protected landscapes allow native PJ communities and native species to persist for the long term while maintaining ecosystem function and process with minimal human management intervention and/or disturbance (IUCN, 2018). Data Compilation: Calculated the percent area within each HUC12 of those areas with protected GAP designation with a GAP status code of ‘1’ or ‘2’. For each HUC12, we assigned a value of ‘1’ where there was greater than 30% area with protected GAP designations. Uncertainty: As the GAP dataset is a compilation of many data sets, data completeness, accuracy, and scale may vary. Furthermore, some human activities have regional or far-reaching effects that are not restricted to protected area boundaries, i.e. air pollution or climate change, which may or may not affect PJ ecosystems within watersheds with protected GAP designations.

3. Piñon-Juniper Habitat Source: LANDFIRE, 2014, Existing Vegetation Type Layer, LANDFIRE 1.4.0, U.S. Department of the Interior, Geological Survey (Available at: https://www.landfire.gov/version_comparison.php.) Description: Coded value – HUC12 with greater than 60% cover of Piñon-Juniper habitat types. The existing vegetation type (EVT) data layer represents the current distribution of the terrestrial ecological systems classification developed by NatureServe for the western Hemisphere. A terrestrial ecological system is defined as a group of plant community types (associations) that tend to co-occur within landscapes with similar ecological processes, substrates, and/or environmental gradients. Justification: The presence of nurse trees of either piñon or juniper species is associated with a greater abundance of seedlings and saplings after mortality events (Floyd et al. 2015) and may be important for maintaining recruitment after drought events (Sthultz et al. 2007). Data Compilation: Calculated the area of those existing vegetation types that represent Piñon-Juniper plant community types, including: 3011 – ‘Rocky Mountain Aspen Forest and Woodland’, 3016 – ‘Colorado Plateau Piñon-Juniper Woodland’, 3019 – ‘Great Basin Piñon-Juniper Woodland’, 3025 – ‘Madrean Piñon-Juniper Woodland’, 3049 – ‘Rocky Mountain Foothill Limber Pine-Juniper Woodland’, 3059 – ‘Southern Rocky Mountain Piñon-Juniper Woodland’, 3115 – ‘Inter-Mountain Basins Juniper Savanna’, 3116 – ‘Madrean Juniper Savanna’, 3119 – ‘Southern Rocky Mountain Juniper Woodland and Savanna’. Second, calculated the total percent area of the combined Piñon-Juniper plant community types per HUC12 watershed. For each HUC12, assigned a value of ‘1’ where HUC12 < 30% cover of Piñon-Juniper Woodlands. Uncertainty: LANDFIRE data products are created at a 30-meter grid spatial resolution raster data set, and are a consistent and scientifically reliable set of mapped fire and vegetation characteristics to be

used for course-scale, national and regional planning; however, they should generally not be used at the pixel level. The appropriate use of LANDFIRE products for local land management planning varies by product, location, and the specific use, and data users should assess these products based on their specific needs. The use of this data for this project is scaled regionally, however, it is likely that some HUC12 watersheds are misrepresented as having or not having an associated piñon-juniper plant community association.

4. Low Fire Potential Source: Dillon, Gregory K. 2015. Wildfire Hazard Potential (WHP) for the conterminous United States (270-m GRID), version 2014 continuous. Fort Collins, CO: Forest Service Research Data Archive. https://doi.org/10.2737/RDS-2015-0047. (Available at: https://www.firelab.org/project/wildfire-hazard- potential). Description: Coded value – HUC12 with areas at low and very low wildfire hazard potential. The WHP geospatial data depicts the relative potential for wildfire, as well as wildfire likelihood and intensity. Areas mapped with higher WHP values represent fuel conditions with a higher probability of experiencing torching, crowning, and other forms of extreme fire behavior under conducive weather conditions. Justification: Wildland fire is a threat to Piñon-Juniper ecosystems; however the impacts may differ across the range of PJ functional types (Margolis, 2014). Factors that affect the fire regime of these ecosystems are driven by a site’s climate, stand density, and existing plant community. Still, climate change may increase fire frequency across functional types due to the overall increases in temperature, length of wildfire season, and severity of fire weather (Westerling et al. 2006). Though wildfires have naturally occurred in PJ ecosystems, PJ is slow to recover from high severity fire, and fire is a primary cause of loss of ecologically important mature PJ woodlands (Huffman et al. 2008). Areas that experience increased fire are likely to transition to grasslands or shrublands as post-fire recovery is very slow, particularly where there is a lack of precipitation (Rocca et al. 2014; Romme et al. 2003). Data Compilation: Calculated the percent of total area classified as having Low or Very Low wildfire hazard potential (WHP) within each HUC12 watershed. HUC12 > 0 percent Low or Very Low WHP were coded ‘1’. Uncertainty: The WHP geospatial data is not an explicit map of wildfire threat or risk, but only spatially approximates relative wildfire risk to ecological resources or values-at-risk. Furthermore, although increases in negative fire behavior is well documented in some areas, it may not be the case in all PJ ecosystems or functional types and may ignore beneficial fire.

References Allen, C. (2007). Interactions Across Spatial Scales among Forest Dieback, Fire, and Erosion in Northern New Mexico Landscapes. Ecosystems, 10, 797–808. https://doi.org/10.1007/s10021-007-9057-4 Barger, N., Henry, A., Woodhouse, C., Neff, J., & Gregory, A. (2009). Influence of livestock grazing and climate on pinyon pine (Pinus edulis) dynamic. Rangeland Ecology and Management, 62(6), 531– 539. Retrieved from https://www.colorado.edu/lab/barger/sites/default/files/attached- files/_influence_of_livestock_.pdf Gori, D., & Bate, J. (2007). Historical range of variation and state and transition modeling of historical and current landscape conditions for pinyon-juniper of the southwestern U.S. Tucson, AZ. Gottfried, G., Swetnam, T., Allen, C., & Betancourt, J. (1995). Pinyon-juniper woodlands [Chapter 6]. In Ecology, diversity, and sustainability of the Middle Rio Grande Basin. Retrieved from https://www.fort.usgs.gov/sites/default/files/products/publications/2997/2997.pdf Huffman, D., Fulé, P., Pearson, K., & Crouse, J. (2008). Fire history of pinyon–juniper woodlands at upper ecotones with ponderosa pine forests in Arizona and New Mexico. Canadian Journal of Forest Research, 38(8), 2097–2108. https://doi.org/10.1139/X08-053 IUCN. (2018). About - Protected Areas | IUCN. Retrieved October 1, 2018, from https://www.iucn.org/theme/protected-areas/about Krist, F., Ellenwood, J., Woods, M., McMahan, A., Cowardin, J., Ryerson, D., Sapio, F., Zweifler, M., & Romero, S. (2014). 2013-2027 National insect and disease forest risk assessment. Fort Collins, CO. Retrieved from http://www.fs.fed.us/foresthealth/technology Margolis, E. (2014). Fire regime shift linked to increased forest density in a piñ on-juniper savanna landscape. International Journal of Wildland Fire, 23, 234–245. https://doi.org/10.1071/WF13053 Meddens, A., Hicke, J., Macalady, A., Buotte, P., Cowles, T., & Allen, C. (2015). Patterns and causes of observed piñon pine mortality in the southwestern United States. New Phytologist, 206(1), 91–97. https://doi.org/10.1111/nph.13193 Peterman, W., Waring, R., Seager, T., & Pollock, W. (2013). Soil properties affect pinyon pine - juniper response to drought. Ecohydrology, 6(3), 455–463. https://doi.org/10.1002/eco.1284 Ramirez, P., & Mosley, S. (2015). Oil and gas wells and pipelines on U.S. wildlife refuges: challenges for managers. PloS One, 10(4), e0124085. https://doi.org/10.1371/journal.pone.0124085 Redmond, M., & Barger, N. (2013). Tree regeneration following drought- and insect-induced mortality in piñon-juniper woodlands. New Phytologist, 200(2), 402–412. https://doi.org/10.1111/nph.12366 Redmond, M., Cobb, N., Clifford, M., & Barger, N. (2015). Woodland recovery following drought-induced tree mortality across an environmental stress gradient. Global Change Biology, 21(10), 3685–3695. https://doi.org/10.1111/gcb.12976 Redmond, M., Forcella, F., & Barger, N. (2012). Declines in pinyon pine cone production associated with regional warming. Ecosphere, 3(12), 1–14. https://doi.org/10.1890/ES12-00306.1 Rocca, M., Brown, P., MacDonald, L., & Carrico, C. (2014). Climate change impacts on fire regimes and key ecosystem services in Rocky Mountain forests. Forest Ecology and Management, 327, 290– 305. https://doi.org/10.1016/J.FORECO.2014.04.005 Romme, W., Floyd-Hanna, M., & Hanna, D. (2003). Ancient Pinyon–Juniper forests of Mesa Verde and

the West: a cautionary note for forest restoration programs. In Proceedings of the Conference on Fire, Fuel Treatments, and Ecological Restoration: Proper Place, Appropriate Time. Colorado State University. Shaw, J. (2006). Population-wide changes in pinyon-juniper woodlands caused by drought in the American southwest: effects on structure, composition, and distribution. In IUFRO Landscape Ecology Conference (pp. 1–8). Locorotondo, Bari (Italy). Retrieved from https://www.fs.fed.us/rm/pubs_other/rmrs_2006_shaw_j005.pdf Sthultz, C., Gehring, C., & Whitham, T. (2007). Shifts from competition to facilitation between a foundation tree and a pioneer shrub across spatial and temporal scales in a semiarid woodland. New Phytologist, 173(1), 135–145. https://doi.org/10.1111/j.1469-8137.2006.01915.x Watson, M. (2005). Habitat fragmentation and the effects of roads on wildlife and habitats. Santa Fe, NM. Retrieved from http://www.wildlife.state.nm.us/download/conservation/habitat- handbook/project-guidelines/Effects-of-Roads-on-Wildlife-and-Habitats.pdf Westerling, A., Hidalgo, H., Cayan, D., & Swetnam, T. (2006). Warming and earlier spring increase western U.S. forest wildfire activity. Science, 313(5789), 940–943. https://doi.org/10.1126/science.1128834 Williams, A., Allen, C., Macalady, A., Griffin, D., Woodhouse, C., Meko, D., Swetnam, T., Rauscher, S., Seager, R., Grissino-Mayer, H., Dean, J., Cook, E., Gangodagamage, C., Cai, M., & McDowell, N. (2013). Temperature as a potent driver of regional forest drought stress and tree mortality. Nature Climate Change, 3(3), 292–297. https://doi.org/10.1038/nclimate1693