Scotch Broom (Cytisus Scoparius) and Soil Nitrogen: Ecological Implications
SCOTCH BROOM (CYTISUS SCOPARIUS) AND SOIL NITROGEN: ECOLOGICAL IMPLICATIONS
by
Jacqueline Shaben
B.Sc, University of Victoria, 1999
A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE
in
THE FACULTY OF GRADUATE STUDIES
(ZOOLOGY)
THE UNIVERSITY OF BRITISH COLUMBIA
October 2006
© Jacqueline Shaben, 2006 ABSTRACT Scotch broom (Cytisus scoparius), a leguminous shrub with nitrogen-fixing rhizobial root associations, is an invasive plant in the endangered Garry oak ecosystem (GOE) of
Southern British Columbia. Broom frequently spreads from disturbed areas such as power line right-of-ways, and thus is a threat to native ecosystems. To unravel the ecological relationship between Cytisus scoparius and soil nitrogen, I conducted two independent studies in 2004-2005.
The first study assessed broom's effect on nutrient availability and plant community composition in two Garry oak sites. The second evaluated the impact on seedling recruitment of broom following fertilization with biosolids obtained from sewage
treatment.
To determine if broom increased soil nitrogen availability, adjacent broom-invaded and
non-invaded plots at two GOE sites (Rocky Point and Bamberton) were investigated.
Here, soil-nutrient availability was compared using PRS™ ion-exchange membrane
probes and concurrent plant surveys. No differences were observed in nitrogen
availability between broom-present and -absent plots except for a weak trend of higher
NHV" at Rocky Point. By contrast, phosphorus availability at Rocky Point was
significantly lower in the broom-present plots. Plant richness was independent of broom
presence, however multivariate analysis showed that species identity and abundance
differed, with native species declining and introduced species increasing in the broom-
invaded plots. I conducted a complementary bioassay in which a native grass and an
exotic grass were grown individually either with Cytisus or the native Holodiscus
discolor, to test if grasses benefited from the presence of Cytisus. No difference in grass
growth between treatment combinations was observed.
The second study tested a novel method of broom control. With fertilization trials, I
compared the efficacy of sewage biosolids and ammonium nitrate on the suppression of
broom seedlings following broom removal. In addition, I conducted a greenhouse
germination experiment to explore the mechanism of broom suppression. Broom
seedlings were fewest and vegetation biomass tended to be highest in the biosolids plots.
The germination experiment indicated no difference in broom germination rate when
II grown without plant competition. This suggests that the negative effects of biosolids on broom are due to increased competition from other plants for light and water.
The management implications of these two studies are that Cytisus scoparius, though not correlated with dramatic increases in nitrogen availability, may alter phosphorus availability in GOE and thus influence plant composition. Addition of sewage biosolids to disturbed sites however, causes broom seedlings to lose their competitive advantage in dense vegetation that is facilitated by complete nutrient addition from biosolids.
Ill TABLE OF CONTENTS ABSTRACT II TABLE OF CONTENTS IV LIST OF TABLES VI LIST OF FIGURES VII ACKNOWLEDGEMENTS . VIII 1. GENERAL INTRODUCTION 1
1.1 INVASIVE PLANTS 1 1.2 STUDY SPECIES - SCOTCH BROOM 2 1.3 CONTROL OF SCOTCH BROOM 2 1.4 SCOTCH BROOM AND THE GARRY OAK ECOSYSTEM 3 1.5 OBJECTIVES 4 1.6 REFERENCES 6 2. IMPACT OF SCOTCH BROOM (CYTISUS SCOPARIUS) ON PLANT COMMUNITY STRUCTURE IN THE GARRY OAK ECOSYSTEM: UNRAVELING THE RELATIONSHIP BETWEEN BROOM, SOIL NUTRIENTS AND PLANT DIVERSITY 7
2.1 INTRODUCTION 7 2.2 METHODS 10 2.2.1 Sites 10 2.2.2 Soil nutrient measurements 13 2.2.3 Broom densities 14 2.2.4 Plant diversity 15 2.2.5 Bioassay 16 2.2.6 Statistical analyses 19 2.3 RESULTS 20 2.3.1 Broom densities/biomass 20 2.3.2 Soil nutrient measurements 23 2.3.3 Plant diversity 31 2.3.4 Bioassay 35 2.4 DISCUSSION 38 2.4.1 Soil nutrient measurements 38 2.4.2 Plant Diversity...... ; 39 2.4.3 Bioassay : 40 2.5 CONCLUSION 42 2.6 REFERENCES 45 3. POTENTIAL OF FERTILIZER APPLICATION TO SUPPRESS SCOTCH BROOM RE-ESTABLISHMENT 47
3.1 INTRODUCTION 47 3.2 METHODS 50 3.2.1 Sites 50 3.2.2 Site Preparation 51 3.2.3 Biosolids and Application Rate 54
IV 3.2.4 Monitoring broom re-establishment 55 3.2.5 Broom germination experiment 58 3.2.6 Statistical analyses 59 3.3 RESULTS 60 3.3.1 Soils 60 3.3.2 Monitoring broom re-establishment 66 3.3.3 Broom germination experiment 72 3.4 DISCUSSION 73 Caveat 73 3.4.1 Soils 73 3.4.1 Plant Monitoring 75 3.4.3 Germination experiment 77 3.5 CONCLUSION 78 3.6 REFERENCES 79
4. CONCLUSION 83
4.1 REFERENCES 84
APPENDICES 85
APPENDIX A - BIOSOLIDS APPLICATION RATE 85 APPENDIX B - NATIVE SEED MIXES USED 88 APPENDIX C - TRACE ELEMENT APPLICATION RATE 89 APPENDIX D - AMMONIUM NITRATE APPLICATION RATE 92 APPENDIX E - SPECIES LISTS 93
V LIST OF TABLES
Table 2-1 Treatment combinations for bioassay and corresponding description of species 17
Table 2.2 Initial characteristics of soil used in bioassay and of reference soil obtained from an intact Garry oak meadow 18
Table 2.3 Scotch broom densities and biomass/m2 at each site for both Rocky Point & Bamberton 21
Table 2.4 Selected initial soil characteristics and nutrient values from composite soil cores taken at Rocky Point and Bamberton in February, 2005 24
1 Table 2.5 Mean values of N03~, NFL," " and P in broom-invaded and un-invaded plots- four sample periods and entire growing season 30
Table 2.6 Average species richness and evenness values at Bamberton and Rocky Point 31
Table 2.7 Pre-bioassay TKN values, and pre- and post-bioassay soil NFL/ and NO3" values in ppm for all four bioassay combinations with native GOE soil values for reference 37
Table 3.1 Initial soil analysis results for the three sites prior to treatment 53
Table 3.2 Percent cover and corresponding cover class as adapted from the Braun- Blanquet percent cover scale 57
1 Table 3.3 Results of soil analyses for NFL;" " and N03" taken 3 to 5 months after initial establishment of site and one year after establishment 61
Table 3.4 Summary of comparisons of soil NFI/, NO3" and TKN changes between pretreatment sampling and the first and second years following fertilization 63
Table 3.5 Cation exchange capacity (meq/lOOg) by treatment at all three sites 1 year following site establishment 66
Table 3.6 Mean percent-cover class of broom for each treatment at three sites 68
Table 3.7 Average relative abundance of grasses, non-native grasses & native shrubs- all three sites, each treatment 69
Table 3.8 Average plant species richness by treatment at each of the three sites 70
VI LIST OF FIGURES
Figure 2.1 Map showing locations of two study sites on Southern Vancouver Island. ..11
Figure 2.2 Relationship between average Cytisus biomass (g/m2) and NFLt+ and NO3" availability rates at Rocky Point 22
Figure 2.3 Relationship between average Cytisus density (stems/m2) and nitrogen availability at Rocky Point 22
+ Figure 2.4 NH4 availability in broom-invaded and un-invaded plots for each of 4 sample periods at Bamberton and Rocky Point 26
Figure 2.5 NO3" availability in broom-invaded and un-invaded plots for each of 4 sample periods at Bamberton and Rocky Point 27
Figure 2.6 Phosphorus availability in broom-invaded and un-invaded plots for each sample period at Bamberton and Rocky Point 28
Figure 2.7 Average nutrient supply rates for the period of January to June 29
Figure 2.8 Multidimensional scaling ordinations of plant diversity by quadrat at Bamberton 32
Figure 2.9 Multidimensional scaling ordinations of plant diversity by quadrat at Rocky Point 33
Figure 2.10 Summed percent dissimilarity of plant species between un-invaded and broom-invaded plots at Rocky Point. 34
Figure 2.11 Summed percent dissimilarity of plant species between un-invaded and broom-invaded plots at Bamberton 34
Figure 2.12 Mean mass of grasses grown with Cytisus and Holodiscus 36
Figure 3.1 NH/ and N03" availability in ammonium nitrate, biosolid and un-treated plots at Iona Beach, Burnaby Mountain and Duncan 65
Figure 3.2 Mean number of broom seedlings by treatment at all three sites, sampled
in July 2005 67
Figure 3.3 Mean percent-cover class of broom for each treatment at all three sites 68
Figure 3.4 Dried vegetation biomass for all three treatments at each of the three sites...71
Figure 3.5 Fates of broom seeds and seedlings under different soil fertilizer treatments 72
VII ACKNOWLEDGEMENTS As solitary as these innumerable hours of thesis writing have been, they have consistently served to remind me of how immensely collaborative my project has actually been.
First and foremost, I would like to thank my supervisor, Judy Myers, for accepting me into her lab, encouraging me to stay despite my rough start to grad school and showing confidence in me the whole way through. I also truly appreciate the advice and editing effort that Drs. Art Bomke and Sue Grayston provided me these past three years.
Though I am still no entomologist, there is no denying that, under the tutelage of my lab mates, I have come a long way in understanding what goes on in the lab and greenhouses frequented by the Myers crew. Thanks guys, for making me feel welcome.
For her great sense of humour, field-work stamina, plant i.d. tenacity and problem- solving genius, I thank Kristen Stevenson for postponing her trip to New Zealand in order to work with me all summer, 2005. Kristen, you're the best!
This project was recipient to funding from the British Columbia Transmission
Corporation, the Endangered Species Recovery Fund (WWF) and CRD Parks. For this I am eternally grateful.
I shudder to think of what it would have been like to do my field work without the help of the many people who tromped around in the field with me: Dave Aldcroft, Carrina
Maslovat, Claire Penman, Katherine Knapstein, Tracy Fleming, Geoff Argue, Ian Myers-
Smith, Tanya Pearson and Alex, Bob Maxwell, Gillian Booth, Tom Bird, Amanda Bates,
Linnaea Bates-Bird, Susanna Solecki, Debbie Bryant, Juan Carlos Solano-Montaro, Fran
Iredale, Tony Yang, Tamara Richardson, Valerie Caron, Jerry Ericsson, Michelle
Franklin, Sam Yeamans, Karmen, Darryl Suen, Stephan, Sophie Boizard and my 5
Victoria volunteers.
I also owe a great deal of thanks to the folks at Western Agriculture and Pacific Soils
Analysis for their technical support the whole way through my study.
During the course of this project, I have been the guest of many generous hosts who housed me during my field work and writing. In particular I'd like to thank Val and Ross in Melbourne, Jodi and Kirk in the Gold Coast, Hiram and Susan in Chemainus, Susanna
VIII Solecki, Narrisa and family, the Pearsons and the Bates-Birds, all living in Victoria, for their open arms and generosity.
Of all the people who helped me through my degree, the most influential person was my new, dear friend, Jessica Beaubier. Were it not for Jess's humour, insight, emotional support and fashion sense to guide me, many of these days would have been very dreary indeed.
And finally, though most of my family thinks I'm wacko for spending three (more) years at university, my mom's endless support and Robert's incredible patience and understanding were what kept me chugging away at each task as it came up.
IX 1. GENERAL INTRODUCTION
1.1 INVASIVE PLANTS
The issue of invasive plants has become a globally widespread concern, not only in the field of land-management applications (Council 2004), but also for theoretical and conservation-minded academia (Lodge 1993, Myers and Bazely 2003). The effects that invasive plants can have on ecosystems range from competing with native species for limited resources such as space, light, water, pollinators etc. to large-scale changes in ecosystem processes such as disturbance regimes, hydrology and nutrient cycling which can result in a complete alteration of ecosystem function.
Given such a breadth of impact intensities, it is important to distinguish between those plant species which have relatively small impacts and those whose impacts are large, in order to "prioritize management efforts" (Parker et al. 1999). Standard terminology describing these plants based on their relative impacts has not been thoroughly accepted by researchers and scientific authors though attempts have been made to standardize definitions (Richardson et al. 2000). It is therefore important to establish definitions for groups of plants in the context of this thesis: introduced, exotic and non-native are used interchangeably to mean plants which have been brought by humans to areas where they previously were not established, and which are currently able to grow without cultivation; invasive and weedy are used to define non-native plants which grow unchecked and are unwanted due to undesirable qualities which they may have; ecological engineer is used to define an invasive species which is capable of altering its environment to suit its needs, thereby possibly negatively impacting one or more species and/or ecological processes in the new ecosystem.
Identification of those species which should be prioritized requires ecological understanding of these species in their new environment. Hence, research is needed to investigate: the mechanism by which a species affects its surroundings, the intensity of its effect and possible means by which to control it. .
1 1.2 STUDY SPECIES - SCOTCH BROOM Scotch broom (Cytisus scoparius) is one such plant that has the potential to have strong effects on its environment. As a leguminous shrub with symbiotic nitrogen-fixing root associations with Rhizobium sp. of bacteria, it can possibly change soil nutrient dynamics in its new environment, while its capacity for prolific seed production enables it to spread effectively into appropriate sites. Broom does well in disturbed areas including those with poor, sandy and gravelly soil, but it is intolerant of shade. Its ability to fix nitrogen and to grow in disturbed sites made broom a popular species for use in soil stabilization projects and in tree plantations as a companion crop in Pacific Northwestern states of the U.S. and in Southwestern British Columbia during the 1980's (Deutsch 1997). It was thought that as a by-product of broom, excess nitrogen would become available to commercially valuable trees, thereby enhancing their growth (Wheeler et al. 1987). Instead, broom hindered the growth of trees by shading them and has since spread along open-canopied corridors. These open, broom-invaded sites are not only considered a fire-hazard, but they also act as a seed source for broom to expand into more ecologically sensitive areas, such as, for example, the endangered Garry oak (Quercus garryand) ecosystem (GOE) of
Western Oregon and Washington and Southwestern British Columbia.
1.3 CONTROL OF SCOTCH BROOM
Controlling Scotch broom in these source areas is essential to containing its spread.
Fostering the growth of competitive species by fertilizing sites with an abundant broom seedbank, enables us to target the trait by which broom maintains its competitive edge, i.e. the ability to fix its own nitrogen, while profiting from its main weakness, that of shade intolerance.
These concepts are what guided the reasoning behind the restoration work at Discovery
Park in Seattle, Washington in 1995 (Leccese 1996), a very successful project in which
4.5 hectares of mature Scotch broom were removed and the property fertilized with treated sewage biosolids from the King County Department of Natural Resources' West
Point Sewage Treatment Plant, resulting in a natural park area where grasses and native shrubs replaced the Scotch broom. It is with the same restoration goal in mind that the
2 study for Chapter 3 of this thesis, "Potential of fertilizer application to suppress Scotch broom re-establishment", came about.
1.4 SCOTCH BROOM AND THE GARRY OAK ECOSYSTEM
Broom's ability to fix nitrogen enables it to establish in nitrogen-limited sites, while giving it the potential to increase the overall nitrogen budget of that site. Studies have already been done to demonstrate the nitrogen-fixation rate of individual Rhizobium nodules taken from broom plants and extrapolated to the overall biomass accumulation of broom (Helgerson et al. 1984, Wheeler et al. 1987). The assumption that the nitrogen which accumulates in the plant biomass increases the total soil nitrogen budget has also been tested and supported in a study by Dancer et al. (1977) on various leguminous shrubs, including Cytisus scoparius, invading clay mine tailings in England. An increased total nitrogen budget, however, does not necessarily amount to an increase in the amount of nitrogen that becomes available to plants, as plant roots are generally only able to take
up nitrogen in its mineralized forms (ammonium, NH/, and nitrate, N03") along with a few amino acids (Whitehead 1995, Schimel and Bennett 2004), whereas much of the nitrogen content of the soil is unavailable since the majority of it is immobilized in organic matter.
NHV and N03" are short-lived ions, with NO3" being dissolvable in water and therefore easily accessible to plant roots but at the same time, susceptible to loss via leaching.
NFL;+, on the other hand, is not easily dissolved in water because it binds with the negatively-charged colloidal material in the soil, yet when plant roots come into contact with it, it is also taken up. The more likely pathway by which plants obtain ammonium- derived nitrogen is following the nitrification of ammonium, meaning the process by which nitrogen is converted from NFL* to NO3" via NO2" (nitrite)(Whitehead 1995).
It therefore follows that in order to quantify the amount of nitrogen that is being made
available to plants, either NFI/ and N03" must be measured directly, or else a proxy for these values must be devised. The carbon to nitrogen ratio (C:N) can act as this proxy, and in a study by Haubensak and Parker (2004), C:N quantified across a broom-invasion gradient in Quercus garryana prairies in Washington state was shown to increase as invasion density increased. The way in which the C:N ratio predicts the amount of
3 nitrogen that will be mineralized is as follows: as organic matter is mineralized, the soil microbes responsible for the mineralization require both carbon and nitrogen for metabolism, hence as carbon increases, the greater the nitrogen will be immobilized as well (Whitehead 1995).
However, as of yet, no studies have measured in-situ nitrogen availability in broom- invaded soils in the GOE throughout the entire growing season. Now, with the development of ion-exchange membrane probes (Plant Root Simulator, PRS™, it is possible to detect nitrogen ions as they become available in the soil (Qian and Schoenau
1995). Using these probes makes it possible to determine if broom is truly altering the availability of soil nutrients in the GOE as a result of its nitrogen-fixing capacity.
The second chapter of this thesis, "Scotch broom's (Cytisus scoparius) impact on plant community structure: unraveling the relationship between broom, soil nutrients and plant diversity" consists of a field study in which the seasonal availability of nitrogen, phosphorus and other nutrients were measured in conjunction with thorough plant surveys. In addition, an experimental study explores the direct relationship between two types of grasses, native and introduced when grown with Scotch broom, in an effort to determine if an introduced grass can differentially benefit from being grown with broom.
1.5 OBJECTIVES For my thesis I explored two aspects of Scotch broom's relationship to soil nutrients: its effect on nutrient availability in a sensitive ecosystem and the effect of fertilization on seedling survivorship in disturbed sites. My objectives were:
1) To quantify the seasonal availability of soil nutrients associated with broom-
invaded soils of the GOE
2) To describe differences in plant community structure between broom-invaded
and non-invaded GOE. sites, with a special emphasis on relative abundance
between introduced species and native species.
3) To test a novel approach to controlling broom seedling emergence and success
using sewage biosolids.
4 4) To determine the mechanism by which seedlings are controlled under fertilized conditions.
5 1.6 REFERENCES
Council, F. B. 2004. Invasive Plant Strategy for British Columbia. Fraser Basin Council, Vancouver, B.C. Dancer, W. S., J. F. Handley, and A. D. Bradshaw. 1977. Nitrogen accumulation in kaolin mining wastes in Cornwall 1. Natural Communities. Plant and Soil 48:153- 167. Deutsch, B. L. 1997. Controlling Scotch broom in landscape design: case studies in the Pacific Northwest using soil amendments with biosolids. Master's Thesis. University of Washington, Seattle. Haubensak, K. A., and I. M. Parker. 2004. Soil changes accompanying invasion of the exotic shrub Cytisus scoparius in glacial outwash prairies of western Washington [USA]. Plant Ecology 175:71-79. Helgerson, O. T., J. C. Gordon, and D. A. Perry. 1984. N2 fixation by red alder (Alnus rubra) and Scotch broom (Cytisus scoparius) planted under precommercially thinned Douglas fir (Pseudotsuga menziesii). Plant and Soil 78:221-233. Leccese, M. 1996. Fresh Fields. Pages 44-47 in Landscape Architecture. Lodge, D. M. 1993. Biological Invasions. Trends in Ecology and Evolution 8:133-137. Myers, J. FL, andD. Bazely. 2003. Ecology and Control of Introduced Plants. Cambridge University Press, Cambridge. Parker, I. M., D. Simberloff, W. M. Lonsdale, and e. al. 1999. Impact: toward a framework for understanding the ecological effects of invaders. Biological Invasions 1:3-19. Qian, P., and J. J. Schoenau. 1995. Assessing nitrogen mineralization from soil organic matter using anion exchange membranes. Fert Research 40:143-148. Richardson, D. M., P. Pysek, M. Rejmanek, M. G. Barbour, F. D. Panetta, and C. J. West. 2000. Naturalization and invasion of alien plants: concepts and definitions. Diversity and Distributions 6:93-107. Schimel, J. P., and J. Bennett. 2004. Nitrogen mineralization: challenges of a changing paradigm. Ecology 85:591-602. Wheeler, C. T., O. T. Helgerson, D. A. Perry, and J. C. Gordon. 1987. Nitrogen fixation and biomass accumulation in plant communities dominated by Cytisus scoparius L. in Oregon and Scotland. Journal of Applied Ecology 24:231-237. Whitehead, D. C. 1995. Grassland Nitrogen. CAB International, Wallingford.
6 2. IMPACT OF SCOTCH BROOM (CYTISUS SCOPARIUS) ON PLANT COMMUNITY STRUCTURE IN THE GARRY OAK ECOSYSTEM: UNRAVELING THE RELATIONSHIP BETWEEN BROOM, SOIL NUTRIENTS AND PLANT DIVERSITY
2.1 INTRODUCTION Scotch broom (Cytisus scoparius), a leguminous shrub native to the Mediterranean, has become an invasive plant of ecological concern in several areas of the world, including parts of Australia (Downey and Smith 2000), New Zealand (Williams 1981) and North
America (Bossard and Rejmanek 1994, Prasad 1999). Broom is well adapted to a moderately dry, temperate climate, but is intolerant of shading by trees and other shrubs.
Its invasive nature is attributed to prolific seed production, persistent seeds, year-round photosynthesis due to photosynthetic stems, production of alkaloid compounds, formation of dense, sunlight-blocking stands, a lack of competition from natural enemies in new environments and the ability to fix nitrogen (Williams 1981, Wink et al. 1983,
Wheeler et al. 1987, Waterhouse 1988, Prasad 1999, Downey and Smith 2000). While broom frequently invades disturbed sites in British Columbia (B.C.), Washington and
Oregon, its successful invasion of intact fragments of the Garry oak ecosystem (GOE)
(Parker 2001), an endangered oak and grass habitat (COSEWIC 2000) makes it a species of particular conservation concern.
Like many legumes, broom has symbiotic root-associations with nitrogen-fixing bacteria
(Rhizobia sp.). The ability to fix atmospheric nitrogen allows broom to establish in sites with nutrient-poor soils and may afford it a competitive growth advantage over non-N- fixing plants. At the community scale, broom has the potential to drive changes in soil nutrient dynamics by altering the nutrients available to other plants. This is cause for concern in the management of GOE given that the negative effects of nitrogen-fixing shrubs on invaded plant communities has already been well established in other ecosystems, such as young volcanic sites in Hawaii (Vitousek and Walker 1989, Hughes and Denslow 2005), Californian coastal dune ecosystems (Maron and Connors 1996,
7 Pickart et al. 1998), grassland plains of southern US and northern Mexico (Hibbard et al.
2001) and in the South African fynbos (Stock et al. 1995).
Other researchers have speculated that broom has the same effect on invaded ecosystems as do other N-fixing shrubs. For example, while investigating broom's impact on native shrubs in forests of South Australia, Fogarty and Facelli (1999) determined that broom- invaded soils had higher levels of nitrate (NO3") and ammonium (NFL;+) than adjacent non-broom soils, suggesting that broom is probably responsible for changing soil nutrient availability.
One way in which increased soil fertility can alter plant community structure is by facilitating the invasion of exotic, competitive species such as nitrophyllic forage grasses
(Vitousek and Walker 1989, Weiss 1999). These grasses are particularly good at acquiring available soil nitrogen which allows them to grow quickly and compete successfully with other plants for water, light and other nutrients (D'Antonio and
Vitousek 1992). The dense, copious leaf litter produced by these grasses can alter the microclimate by attenuating sunlight, insulating the soil from temperature fluctuations, and moderating soil moisture, resulting in altered decomposition rates and possibly decreased seed germination of native plants (Facelli and Pickett 1991).
In the Garry oak ecosystem, Dactylis glomerata is one of the more common exotic forage grasses that invade meadows. Site managers speculate that Dactylis establishment is facilitated by increased nitrogen availability as a consequence of broom invasion, thereby resulting in a decline in native plant species and a change in community structure as was demonstrated in the old-field fertilization experiments with Dactylis by Gurevitch and
Unnasch (1989).
The objective of this study is to determine some of the impacts of broom on the ecology of native plants of the GOE in southern British Columbia and to contribute to the management of the small fragments of these ecosystems that remain in order to optimize native plant species conservation. In order to do this it is clear that Scotch broom's effects on plant community structure and diversity must be quantified and the mechanism(s) by which broom changes the plant community structure must be determined.
8 Here I evaluate five aspects of the impacts of broom in the GOE by:
1) Quantifying the seasonal soil nutrient availability correlated with Scotch broom- invaded sites in the GOE of southern Vancouver Island.
2) Comparing the level of soil nitrogen and phosphorus availability in sites with and without the presence of Scotch broom.
3) Determining if the level of soil nitrogen availability and plant community structure are related in these GOE sites.
4) Determining if there is a difference in plant community structure between broom- invaded and un-invaded Garry oak sites, and if so, whether the changes are represented by changes in native and exotic plant abundances.
5) Ascertaining whether or not an exotic nitrophyllic forage grass differentially benefits over a native grass growing in association with broom.
9 2.2 METHODS
2.2.1 Sites
To determine whether Scotch broom invasion is correlated with higher levels of plant- available nitrogen, contiguous invaded and un-invaded sites with similar slope, aspect and substratum, having no past history of Scotch broom invasion but currently being invaded, were sought in the GOE.
Appropriate sites were found at two locations on southern Vancouver Island: Bamberton and Rocky Point (Figure 2.1).
10 Figure 2.1 Map showing locations of two study sites on Southern Vancouver Island. Bamberton is south of Duncan, B.C., on the east side of the Malahat Highway. Rocky Point is on Department of National Defense property in Metchosin, B.C.
Five sites at Rocky Point were located at the Department of National Defense's Rocky
Point munitions depot in Metchosin B.C (48°19.500'N, 123°32.550'W). The lands at
Rocky Point were, for a large part, undeveloped and had a magnificent stand of Garry oak trees surrounded by Douglas-fir forest. The Garry oak meadowlands were used historically by the First Nations people for traditional purposes such as camas harvesting
11 and were periodically burned. More recently, European immigrants used the site for sheep grazing (Arthur Robinson, personal communication, 2006).
All five sites at Rocky Point had negligible slope and were south-facing. The sites were contained within a total area of 3 hectares and were a minimum of 50 metres apart and a maximum of 400 metres apart. Sites 1 and 2 were the western-most sites and were situated within the largest grove of Garry oak trees. Sites 3 and 4 were approximately
75m from the ocean shore of Juan de Fuca Strait with site 3 situated in a small grove of
Garry oak trees and site 4 surrounded by scrubby Garry oak. Site 5 was approximately
200m from the ocean in a clearing surrounded by mixed Garry oak and Douglas-fir forest.
The Bamberton property was situated 48°34.760' N, 123°31.300' W, approximately 40 km north of Victoria (14 km north of Goldstream Provincial Park). The Bamberton meadows were several hundred metres from the closest road (Malahat Flighway) and, other than a small old skid road through the Douglas-fir forest, had no sign of physical disturbance and received very little human traffic. The study sites were situated directly south of the abandoned historical Bamberton cement plant.
The three sites at Bamberton were east-facing and all were meadow openings in the otherwise encroaching Douglas fir (Pseudotsuga menzesii)/Arbutus (Arbutus menzesii) forest. The entire area of 4 hectares was on a steep slope. In order to eliminate any leaching from one plot to another, broom and non-broom plots were established side-by- side along the same contour in all three sites, which were 200 - 500 metres apart. Though the site did not support actual Garry oak trees, the meadows were composed almost entirely of plants that are associated with the GOE (eg, Zygadena venenosus, Camassia quamash, Allium cernuum, Elymus glaucus, Lomatium nudicaule and L. utriculatum) and are subject to the same types of disturbance and climate properties as the GOE. Hence, the habitat of the sites studied is considered to be part of the Garry oak associated ecosystem, and is included by the Garry Oak Ecosystem Recovery Team (Fuchs 2001) when considering the management, restoration and protection of GOE.
12 2.2.2 Soil nutrient measurements Soil nutrient measurements were conducted using two methods: standard extraction from soil core samples and ion-exchange membrane probes.
In February 2005, 15 soil cores (20mm diameter, 0-10cm depth, including organic and mineral soil) from each broom and non-broom plot, were pooled, air-dried for 72 hours and sieved through a 2mm sieve. No soil samples were taken at Site 3 at Bamberton because that site was not established until a later date. Soils were then analyzed for pH
(determined potentiometrically on a 1:1 soil/distilled water slurry), total C (determined directly on a LECO CR 12 Carbon Analyzer), total nitrogen (determined colourimetrically on a Technicon Autoanalyzer, using a semi-micro Kjeldahl digest), ammonium (NFL;"1") and nitrate (NO3") (determined colourimetrically on a Technicon autoanalyzer using a 1:2 soil to solution 1M potassium sulphate extract), available P
(determined colourimetrically using a 1:10 soil to solution Bray Pi extract), Ca, K, Mg, and Na (determined on an Atomic Absorption Spectrophotometer using a 1:5 ammonium acetate extract), and % organic matter was calculated from total C x 1.724, at Pacific
Soils Analysis labs, Richmond B.C.
Ion-exchange membrane (Plant Root Simulator, PRS™) probes were buried in the soils of both the broom-invaded and un-invaded plots at each site to a depth of between 5 and
10cm in the primary rooting zone and were left for burial periods of 5 weeks. The probes consisted of 20cm long plastic stakes, which support a 5cm piece of ion exchange membrane. Each pair of probes is made up of one anion-exchange membrane probe and one cation-exchange membrane probe. At each site, four samples were taken in both the broom-invaded and the un-invaded plots. Each sample consisted of four probe pairs (8 probes) that were pooled by probe type to give two composite soil analyses per sample, one for the anion-exchange membranes and one for the cation-exchange membranes.
Four, 5-week-long burials were consecutively repeated during the spring, starting on
January 22, 2005 and ending on June 12, 2005. Probes work by adsorbing ions as they become available in the soil, therefore the unit of measurement is a rate of ion availability and is described in mass of nutrient per membrane surface area per time
13 buried. This translates into Hgrams/lOcm /35days for individual burials and wgrams/10cm2/140days for the entire period.
Once probes were removed from the soil, they were immediately replaced in the same slot by new probes. The soil was then tamped down to ensure contact with the membrane.
Removed probes were immediately rinsed with de-ionized water and scrubbed lightly with a brush to remove all soil and plant material, packaged moist in sealed plastic bags, placed on ice in a cooler and sent to the Western Agriculture Innovations lab in
Saskatoon, Saskatchewan for analysis where they were eluted using a 0.5 M HC1 solution. The elution was analyzed for NO3", NH4"1", and PO43" using a colourimetric autoanalyzer, and Ca, Mg, K, Cu, Al, and Zn were detected using inductively-coupled- plasma spectrophotometry (Hangs et al. 2004).
Availability rates of ammonium, nitrate and phosphorus were compared between broom- invaded and uninvaded plots, for all four individual burial periods as well as for the sum of all four burials to obtain a total seasonal nutrient availability at both Rocky Point and
Bamberton.
2.2.3 Broom densities
The broom stands chosen for the study all varied in age and density. Since this would likely have some effect on the level of impact on nutrient availability, average broom biomass was determined for each site. For all five Rocky Point sites and for sites 2 and 3 at Bamberton an average of ten, lm2 quadrats were sampled along a transect that was laid through the centre of the broom plot. The first quadrat was placed at a randomly chosen spot within the first metre of the transect. Ensuing quadrats were systematically placed at
1-1.5m intervals, calculated based on the length of the transect. Shoot and root biomass per metre-squared (density) was determined by removing every plant in the quadrat by the roots (using a Weed Wrench ™) and weighing it in the field with a hanging scale accurate to 20 grams. The total weight of all plants in all quadrats in a site was summed and averaged to obtain the broom biomass/m per site.
Due to time and disposal space constraints, the large number of shrubs per square metre at sites 1 (~ 9 shrubs/m2) and 5 (~5 shrubs/m2) at Rocky Point, made it impractical to remove all of the plants in the quadrats for weighing. Therefore, for these two sites,
14 regressions were calculated as a way to predict the overall average biomass using non• destructive measurements taken on all of the broom plants in the quadrat, of which only a portion were removed (Parker 1996). The measurements consisted of: length of longest stem (cm), diameter of main stem at 5cm from the ground (mm) and number of stems at
5cm above the ground.
2.2.4 Plant diversity To quantify the effect of broom invasion on plant community composition, plant surveys were conducted in May and June 2005. Measures for species richness and abundance were obtained using 25 points in a 0.50m x 0.50m quadrat. Ten quadrats were surveyed for each broom and non-broom plot at each site. Leaf-area index was used as a non• destructive measure of abundance for each species. This was done by counting the number of times each plant species came into contact with a long slender dowel held vertically to the ground at each of the 25 evenly spaced points on the quadrat. Using this method, the relative amount of sunlight intercepted by the individual species was approximated and used to represent the relative abundance of a species. Plant abundances were then used to conduct species diversity analyses for each location.
For each location, separate t-tests were used to compare richness of plant species (species number) and evenness of plant abundances (Pielou's J') by treatment. Following this, multidimensional scaling (MDS) was used to graphically represent differences in species assemblages between sites and between broom presence and absence. To construct the
MDS plots, the data were first assembled into a log(x+l) transformed Bray-Curtis similarity matrix using the statistical software PRIMER 5.2.2 (Plymouth Routines in
Multivariate Ecological Research). In creating the Bray-Curtis matrix, a value is generated which represents the degree of similarity between each pair of samples. The resulting values were then ordinated onto a two-dimensional graph (MDS) which visually depicts the degree of similarity of individual samples by their graphical proximity to each other in two-dimensional space. To test for significant differences between species assemblages, an analysis of similarity (ANOSEVI) was used (PRIMER 5.2.2). This test is analogous to a one-way ANOVA. An ANOSIM generates a value of R, called a global
Rho whose value is between 0 and 1, with 0 being the null hypothesis of no difference
15 between treatments and 1 representing the case where all similarities within groups (sites) are greater than the similarities between groups.
To ascertain which species were responsible for driving the differences seen between the broom and non-broom plots, dissimilarity percentages were determined using similarity percentage analysis (SIMPER) in PRIMER. Plant species that contributed to the top 50% of the cumulated percent dissimilarity between broom and non-broom plots were highlighted. Of these species, some were introduced plants and others were native plants.
Of these, some species increased in abundance from broom absence to broom presence, and other species decreased in abundance. Hence, four categories were created (Introduced/Increase, Introduced/Decrease, Native/Increase, Native/Decrease) and the cumulative percent dissimilarity from each category was compared to quantify the overall difference in plant species richness and abundance between broom-invaded and un-invaded plots.
The species determined, with SIMPER, to be driving the majority of the change in community dynamics were then regressed with available nitrogen rates at each site to ascertain the level of correlation between nitrogen availability and plant differences.
2.2.5 Bioassay To determine whether broom increases the growth of native and invasive grass species, a bioassay experiment was conducted from May 2004 until June 2005. The secondary question was whether or not the nitrophyllic, invasive Orchardgrass {Dactylis glomerata) preferentially benefits from being grown with broom compared to Roemer's fescue
(Festuca idahoensis var roemeri), which is native to the GOE.
The experiment consisted of treatment combinations of one grass and one shrub per
300ml pot. The grasses were paired with one of two shrub species, Cytisus scoparius
(broom) and Holodiscus discolor (Oceanspray), a common sun-loving, non-nitrogen- fixing shrub found growing in similar conditions to those preferred by broom. Plant success was measured by two separate outcomes: plant growth, determined by plant biomass; and plant reproductive effort, determined by number of flower spikes.
16 Twenty replicates were planted per combination in June of 2004. Treatment combinations
were as follows:
Shrub species (S) Grass species (G) Species code Type code
I=introduced N=native
Cytisus scoparius Dactylis glomerata C:D IS:IG
Cytisus scoparius Festuca idahoensis C:F IS:NG
Holodiscus discolor Dactylis glomerata H:D NS:IG
Holodiscus discolor Festuca idahoensis H:F NS:NG
Table 2-1 Treatment combinations for bioassay and corresponding description of species type.
To most closely replicate the edaphic conditions experienced by these species in the field,
the plants were potted in soil collected from Garry oak habitat. To ensure that the plants
were not profitting directly from increased levels of nitrogen mineralized from organic
matter already in the soil, we collected mineral soil from a development site near Mill
Hill Regional Park (48° 27.290' N, 123° 28.930' W), where the topsoil had already been
scraped away. The resulting potting soil was a 50:50 combination of collected soil and
perlite. Soil was initially analyzed for pH, NH/, N03, TKN, P, K," Ca and Mg. The soil
was particularly deficient in phosphorus (personal communication, Bill Herman, Pacific
Soil Analysis) therefore potting soil was fertilized with triple superphosphate (see Table
2.2). At the end of the experiment, the soils from all four treatments were analyzed for
NFL;"1" and NO3" and compared with the original results in order to determine if any of the treatments had altered the soil levels of NFL;"1- and NO3".
17 + Soil Type PH TKN C:N EC C NH4 N03 P K Ca Mg
% mmhos/ % —ppm- -me/100- cm
Pre-bioassay 5.0 0.51 46.7 0.24 23.8 6 5.4 5 230 450 33 GOE mineral soil only
Reference GOE 4.5 1.37 5.8 0.46 7.9 10 19.7 21 150 1400 160 soil inch organic matter
Table 2.2 Initial characteristics of soil used in bioassay and of reference soil obtained from an intact Garry oak meadow.
Grasses were grown from seed and were transplanted to the pots when they had two leaves approximately 10cm long. Year-old Holodiscus seedlings were purchased from a greenhouse. Cytisus seedlings were collected from the wild at Mill Hill Regional Park so as to have natural soil microbial community associations (ie. Rhizobia inoculation).
Cytisus seedlings were approximately two years old and were of similar size to the
Holodiscus seedlings. Broom seedlings of the most similar weight, stem width, length and number of stems were selected, using rank sum method.
To prevent stress and competition for water, pots were maintained at relatively constant moisture levels. To prevent growth imbalance due to shading, pots were arranged randomly and repositioned periodically in an unheated greenhouse with natural light only. In Fall 2004, the Cytisus plants were infected with a powdery mildew and twice were treated with a fungicide before being transplanted into one-gallon pots and subsequently transferred outdoors for the remainder of the experiment.
The experiment was terminated on June 5, 2005. Numbers of flower spikes on grass plants were counted, then the aboveground portions of the plants were removed, sorted
18 into individual paper bags that were coded by treatment, dried to constant weight and then weighed.
2.2.6 Statistical analyses All statistical analyses were measured using a 5% level of significance.
Regressions for broom biomass at sites 1 and 5 at Rocky Point were calculated using
SAS version 9. The number of stems did not significantly enhance biomass predictions in sites 1 and 5, so this variable was removed from the model and log of stem diameter was used to predict the natural log of broom field weight for site 1 and log of broom field weight for site 5.
For each burial period, the average nutrient supply rate within each plot was calculated
(eg. combined the four sample values, then divided by four) and for each location separately, the nutrient values were used in a one-way ANOVA, blocked by site (JUMP
IN 5.1), to determine statistical significance between the broom and non-broom plots.
To compare average nutrient availability for the entire burial period, the average nutrient supply rate within each plot from all four burial periods was summed and then analyzed separately for each location, using a one-way ANOVA blocked by site (JUMP FN 5.1).
The relative treatment effect of broom on the growth and flower production of the two grass species was determined using a general linear model (JUMP IN 5.1) with grass species, shrub species and the interaction between grass species and shrub species as predictor variables for the dependent variables, grass biomass and number of flowers.
Paired t-tests (JMP IN 5.1) were used to test the difference between pre- and post- treatment concentrations for NFL}"1" and NO3" in bioassay soils.
19 2.3 RESULTS
2.3.1 Broom densities/biomass A forward stepwise regression was used to determine the significant parameters for predicting the biomass of broom for sites 1 and 5 at Rocky Point, resulting in the following regressions:
Site 1 Ln field weight = -0.21472 + 1.74684(log stem diameter) + 0.0061 (stem length)
Site 5 Log(lO) field weight = 4.09314 - 26.2184(l/stem diameter)
Regressions to determine correlation between broom biomass and nitrogen availability and between density and nitrogen availability were conducted for the Rocky Point site.
Similar data are not available for the Bamberton site due to premature broom removal.
Values for the density, biomass/m2 and average biomass/m2 are given in Table 2.3 below.
20 Location Site Density Biomass Average plant (plants/m2) (g/m2) biomass (g) Rocky Point 1 9.3 3016 324.3 2 4.2 1632 388.6 3 3.8 270 71.1 4 2.9 2348 809.7 5 4.9 2716 554.3 Bamberton 2 3.6 468 130.0 3 1.7. 456 268.2
Table 2.3 Scotch broom densities and biomass/m2 at each site for both Rocky Point and Bamberton. The average plant biomass at each site gives an indication of the relative abundance of older, larger shrubs versus smaller, younger shrubs or seedlings per site.
2 2 At Rocky Point, neither average broom biomass per m nor density per m was related to nitrogen availability (Figures 2.2 and 2.3). Since broom biomass and broom stem density were not correlated with levels of nitrogen availability, neither the density nor the biomass measure was included as a covariate in the analysis comparing nitrogen availability in broom-invaded and un-invaded plots.
21 300
• NH4+: r2 = 0.003, p = 0.93, df =4
O N03-: r2 = 0.21, p = 0.44, df = 4 250
«3 200 T3 O
150 E o o 100 H 3
50 5
500 1000 1500 2000 2500 3000 3500
Average Cytisus biomass (g/m )
Figure 2.2 Relationship between average Cytisus biomass (g/m2) and NHU* and NO3" availability rates at Rocky Point. Bars are SE.
300 -j • NH4+: r2 = 0.19, p=0.46,df = 4 O N03-: r2 = 0.63, p=0.10,df = 4 250 -
>l re 200 - •0 0
150 - E 0u 100 -
5 50 -
0 - 10
Average Cytisus density (stems/m )
Figure 2.3 Relationship between average Cytisus density (stems/m ) and nitrogen availability at Rocky Point. Bars are SE.
22 2.3.2 Soil nutrient measurements
Results from the initial soil analyses are given in Table 2.4. There are considerable differences in levels of pH, percent carbon, TKN, phosphorus, potassium and calcium between Rocky Point and Bamberton, rendering it appropriate to compare nutrient availabilities between broom-invaded and un-invaded plots separately for Rocky Point and Bamberton. Average pH and Ca were both considerably lower at Rocky Poinfthan at Bamberton, while %C, TKN and P were considerably higher at Rocky Point than at
Bamberton. Of the two nutrients of primary interest, (phosphorus and nitrogen), only phosphorus displayed a noteworthy difference as it was twice as high in the non-broom plots as in the broom-invaded plots at both locations.
23 Site Nutrient Cytisus presence/absence PH % C TKN mg/g P ppm K ppm Ca ppm Mg ppm Broom No- Broom No- Broom No- Broom No- Broom No- Broom No- Broo No- broom broom broom broom broom broom m broom Rocky 5.12 5.2 16.8 19.5 10.9 11.6 20.8 45.4 178 243 1380 2010 252 406 Point ±0 ±0 ±0.2 ±0.3 ±0.1 ±0.2 ±0.7 ±1.5 ±2.0 ±3.4 ± 15.1 ±21.5 ±4.9 ± 13.0
Bamberton 7.9 7.8 10.4 10.4 5.4 4.3 1.5 2.8 325 260 7000 6750 172.5 135 ±0 ±0 ±0.6 ±0.3 ±0 ±0.1 ±0.2 ±0.4 ±10.6 ± 15.9 ±353.6 ±88.4 ± 15.0 ±8.8
Table 2.4 Selected initial soil characteristics and nutrient values from composite soil cores taken at Rocky Point (n=5) and Bamberton (n=2) in February, 2005. Errors are SE. Using the PRS probes, NH/ availability tended to be higher in broom-invaded plots at
Rocky Point, though the differences were not significant. Meanwhile, at Bamberton,
NFL;"*" availability was an order of magnitude lower than at Rocky Point and there were no detectable differences between broom-invaded and un-invaded soils (Figure 2.4). Nitrate availability was not significantly different between broom-invaded and un-invaded plots at either Rocky Point or Bamberton over the four months of measurement (Figure 2.7;
Table 2.5). Meanwhile, at Rocky Point, phosphorus availability was significantly higher in non-broom plots for three of the four sample periods as well as for the cumulated growing season (Figures 2.6, 2.7; Table 2.5).
25 + February NH4 March NH4
Broom-invaded E=S No Broom
+ n April NH4 MayNHl4 l
Bamberton Rocky Point Bamberton Rocky Point
Figure 2.4 NH/ availability in broom-invaded and un-invaded plots for each of 4 sample periods at Bamberton and Rocky Point. Bars are SE. 100 February NO3" March NO3
80 Biooin-inv.Kled O -o No Broom 60
« £ 40 g o
20
80 • April NO3 May NO3
Bamberton Rocky Point Bamberton Rocky Point
Figure 2.5 NO3" availability in broom-invaded and un-invaded plots for each of 4 sample periods at Bamberton and Rocky Point. Bars are SE. 14 n February P March P 12 I Bioonviiw.ided 10 IIIo Broom
8^
6
I 4 2
10 April P May P 12
10
8
6
4 aB Bamberton Rocky Point Bamberton Rocki y Point
Figure 2.6 Phosphorus availability in broom-invaded and un-invaded plots for each of 4 sample periods at Bamberton and Rocky Point. Bars are SE.
28 Rocky Point
N03 NH4 P
Figure 2.7 Nutrient supply rates averaged for the total sampling period of January 23 to June 12, 2006. Bars are SE
29 + N03" NH4 P Location Month Broom No F P Broom No F P Broom No F P Broom (df) Broom (df) Broom (df) Bamberton Feb 6.43 4.33 0.15 0.76 1.00 1.00 0.00 1.00 2.27 1.15 0.96 0.51 2.56 0.95 (14) 0.38 0.38 (1,1) 0.54 0.16 (1,1) Mar 5.70 5.45 0.08 0.80 0.62 0.62 0.00 1.00 2.33 2.28 0.00 0.96 0.86 0.10 (1,2) 0.35 0.43 (1,2) 0.66 1.14 (1,2) Apr 2.95 0.81 16.66 0.06 0.23 0.13 1.00 0.42 1.80 2.81 1.64 0.33 0.91 0.29 (1,2) 0.23 0.13 (1,2) 0.26 0.60 (1,2) May 4.38 3.92 5.79 0.14 1.37 1.38 0.03 0.87 2.33 2.34 0.00 0.99 1.01 0.84 (1,2) 0.49 0.52 (1,2) 0.46 0.79 (1,2) Total 17.32 13.05 1.03 0.42 2.88 2.80 0.26 0.66 7.98 8.18 0.00 0.95 5.24 2.43 (1,2) 1.00 0.88 (1,2) 1.77 2.67 (1,2) Rocky Feb 72.70 73.49 0.00 0.96 25.73 19.56 0.14 0.73 2.45 7.92 10.33 0.03 | Point 15.35 15.12 (1,4) 5.23 8.98 (1,4) 0.51 1.03 (1,4) i Mar 22.14 14.67 0.73 0.44 9.21 7.88 0.10 0.77 2.22 7.46 7.50 0.05 12.37 5.45 (1,4) 2.26 3.77 (1,4) 0.50 1.92 (1,4) Apr 6.86 8.56 0.89 0.40 5.07 6.49 0.93 0.39 2.40 7.63 7.60 0.05
0.73 1.16 (1,4) 0.98 2.07 (1,4) 0.28 1.10 (1,4) < May 6.96 8.94 0.60 0.48 12.58 6.84 2.11 0.22 4.43 9.90 6.22 0.07"" 1.48 3.03 (1,4) 2.32 1.84 (1,4) 0.54 2.15 (1,4) Total 88.33 88.78 0.00 0.97 53.89 36.71 0.74 0.44 11.64 33.01 8.29 0.05 J 35.27 33.51 (1,4) 9.39 13.92 (1,4) 1.62 7.61 (1,4)
Table 2.5 Mean values in (wg/10cm2/35days) of NO3", NH44" and P in broom-invaded and un-invaded plots for all four sample periods and for the entire growing season («g/10cm2/140days). SE is below mean; F statistic, df and p values given for each comparison, with significant p values highlighted.
o 2.3.3 Plant diversity
At neither location was broom presence/absence significantly correlated with the total plant species richness or with species evenness using Pielou's J'(Table 2.6).
Species Richness Species Evenness (J')
Broom present Broom absent Broom present Broom absent
Bamberton 12.53 ± 1.1 11.9 + 1.8 0.84 ±0.02 0.83 ±0.02
Rocky Point 11.08 ±1.1 11.42 ±1.2 2.21 ±0.17 2.35 ± 0.23
Table 2.6 Average species richness and evenness values at Bamberton (N = 3) and Rocky Point (N = 5). Errors are SE.
The MDS ordinations for both locations represented in Figures 2.8 and 2.9 illustrate different results at each location. At Bamberton (Figure 2.8), quadrats showed a significant pattern of clustering by site (ANOSIM, R = 0.921) and also, but less strongly, by presence or absence of broom (ANOSIM, R = 0.657).
31 Stress: 0.20
Figure 2.8 Multidimensional scaling ordinations of plant diversity by quadrat at Bamberton. Quadrats were clustered using species richness and abundance data (log(x+l) transformed; Bray-Curtis similarity) and were compared by broom-invaded and un- invaded at Bamberton.
At Rocky Point (Figure 2.9), the quadrats show a pattern of clustering by site and by presence or absence of broom which are both significant (ANOSFM, R = 0.81, R = 0.77, respectively). The results indicate that species diversity is more similar in plots invaded by broom than in plots lacking broom.
32 Figure 2.9 Multidimensional scaling ordinations of plant diversity by quadrat at Rocky Point. Quadrats were clustered using species richness and abundance data (log(x+l) transformed; Bray-Curtis similarity) and were compared by broom-invaded and un- invaded at Rocky Point.
From the SIMPER analysis, 60% of the plants that were responsible for driving the top
50% of differences seen between the broom-invaded and un-invaded plots at Rocky Point were introduced species whose abundances increased in the broom-invaded plots, while
21% of the top 50% Rocky Point plants driving the dissimilarity between broom-invaded and un-invaded plots were native species whose abundances decreased in the broom- invaded plots (Figure 2.10). At Bamberton, 60% of the plants that were responsible for driving the majority of the differences seen between broom-invaded and un-invaded plots were native species whose abundances decreased in the broom-invaded plots, while 24% of the top 50% Bamberton plants driving the dissimilarity were native species whose abundances increased in the broom plots (Figure 2.11).
33 80 Rocky Point
Introduced Native Figure 2.10 Summed percent dissimilarity of plant species between un-invaded and broom-invaded plots at Rocky Point.
Bamberton 100
(5 80
• Increase in broom 1 Decrease in broom 60 A
1
£ 20 A u
Introduced Native
Figure 2.11 Summed percent dissimilarity of plant species between un-invaded and broom-invaded plots at Bamberton.
At Rocky Point the three species that mainly accounted for the differences between broom-invaded and un-invaded plots were; Anthoxanthum odoratum (15.2% of the total percent dissimilarity), Holcus lanatus (10.2%) and Bromus hordaceous (5.6%), all of which are introduced grasses whose abundances increased in the broom sites. Dactylis
34 glomerata, the exotic grass species, which is of particular interest to managers of Garry oak ecosystems, actually decreased in broom-invaded plots and represented 4.7% of the dissimilarity between broom and non-broom. At Bamberton the four species that mainly accounted for the differences between broom-invaded and un-invaded plots were Alium cernuum (5.8%), Festuca idahoensis (5.3%) and Camassia sp. (5.2%), all native species whose abundances declined in the broom, and Lonicera hispidula (5.2%), a native trailing shrub which increased in the broom-invaded plots. Dactylis glomerata was not present in any of the Bamberton plots.
Regressions were used to determine if the abundance of Anthoxanthum odoratum, Holcus lanatus and Bromus hordaceous, the invasive grasses that increased in broom-invaded plots, were correlated with nitrogen levels at the five Rocky Point sites. The results indicated that neither NO3" nor NH4+ values were significantly correlated with the abundance of any of the three plant species.
At Bamberton, abundance of Festuca idahoensis was positively correlated with NO3" (r2
= 0.99, F(i,4) = 306.6, p = 0.04).
2.3.4 Bioassay
The bioassay was terminated in June 2005. For both grass species, dry weight did not differ between individuals grown with Holodiscus or with Cytisus, though Dactylis was consistently larger than Festuca. Mean overall Dactylis biomass was 17.0g ± l.lg and mean Festuca biomass was 5.2g ± 0.8g (Fi, 59 = 80.60, p<0.001).
Mean biomass of Dactylis grown with broom was 16.5g ± 1.6g and with Holodiscus was
17.4g ± 1.6g, while mean biomass of Festuca grown with broom was 7.0g ± 1.5g and with Holodiscus, 3.9g ± 0.6g (Figure 2.12).
35 30 -,
Cytisus r»i Holodiscus
0 Dactylis (I) Festuca (N)
Grass species
Figure 2.12 Mean mass of grasses grown with Cytisus and Holodiscus. Error bars are SE.
Both grass species tended to produce more flower spikes when grown with Cytisus than when grown with Holodiscus, although these were not significant differences. Mean number of flower spikes on Dactylis with broom was 3.4 ± 1.0 flowers compared to
1.9 ± 0.8 flowers on Dactylis grown with Holodiscus (F(i>26) = 1.41, p= 0.24) and mean number of flower spikes on Festuca with broom was 3.9 ± 0.8 spikes compared to
1.1 ± 0.9 spikes on Festuca with Holodiscus (F^i) = 3.29, p = 0.08).
Soils were analyzed for available nitrogen at the end of the bioassay (Table 2.7). For all four treatments, Nrl/ levels increased from the pre-bioassay level of 6ppm, to an average increase of 11.1 ± 1.4 ppm, (t-ratio = 7.68, df = 11, p < 0.0001), but there was no difference across treatments (F(i,n) = 0.28, p = 0.84). The average NO3" levels did not change significantly for any of the treatments (t-ratio = -2.14, df = 11, p = 0.97).
36 + Soil Type TKN NH4 N03
Reference GOE 1.37 10 19.7
Pre-bioassay soil 0.51 6 5.4
C:D 16.3 2.7
C:F 15.3 3.5
H:D 19.3 6.3
H:F 17.3 3.5
Table 2.7 Pre-bioassay TKN values, and pre- and post-bioassay soil NHA+ and NO3" values in ppm for all four bioassay combinations with native GOE soil values for reference.
37 2.4 DISCUSSION 2.4.1 Soil nutrient measurements This study was designed to determine if Scotch broom facilitates changes in plant community structure and composition in Garry oak meadows through mediation of changes in the nutrient supply, in particular by increasing available nitrogen. Though plant available ammonium (NH/) tended to be higher in the non-broom plots at one site,
Rocky Point, the high variability in the nitrogen supply at both sites, Bamberton and
Rocky Point, resulted in there being no significant correlation between broom invasion and soil nitrogen. It is possible that nitrogen availability had differed between plot types, and that the extra nitrogen was taken up by plants rather than being adsorbed by the probes. However, by using the ion-exchange membrane probes, continuous nitrogen availability was measured throughout the growing season and therefore, these values should amount to the same nitrogen utilization as by plants, thereby accounting for the same nutrient levels as were actually available.
It appears that Scotch broom in the Garry oak ecosystem has less of an effect on the availability of soil nitrogen than do other in-situ, micro-site factors such as climatic influences as well as physical and biological soil characters that facilitate the soil processes which are responsible for nitrogen mineralization. These findings support those of Waterhouse (1988) who found no difference in total nitrogen between broom-invaded and un-invaded soils in Australia, but contrasts to the study of Haubensak and Parker
(2004) who found an increase in total nitrogen and a decrease in carbon to nitrogen ratio in soils sampled along an increasing gradient of broom invasion in a Garry oak prairie grassland. As the ratio of carbon to nitrogen increases, the rate of nitrogen mineralization also typically increases, and thus this measure is used as a proxy for nitrogen availability.
Haubensak and Parker's results therefore indicate that broom-invaded Garry oak sites have the potential for greater nitrogen availability.
Though broom-invasion was not correlated with nitrogen availability, the supply rate of phosphate in the broom plots at Rocky Point was significantly lower than in the un- invaded plots. From the initial nutrient analyses of soil cores it became apparent that phosphorus levels were very low in both the broom-invaded and non-broom plots at
38 Bamberton, while at Rocky Point they were moderate in the broom-invaded plots and high in the non-broom plots (Neufeld 1980). These results support those found in a similar study by Caldwell (2006), where broom-associated soils in a California coastal prairie grassland had lower levels of inorganic phosphorus than did adjacent non-broom associated soils. Caldwell also detected a 123% increase in the soil enzyme phosphatase in soils under Scotch broom, which he speculates could be responsible for driving the lower levels of available soil phosphorus detected in the same soils. Caldwell suggests that though phosphatase catalyzes the release of inorganic phosphate from organic phosphate, it is then likely taken up quickly by broom to be used in the nitrogen-fixation process. Therefore in the present study the lower availability of phosphorus in broom plots at both Bamberton and Rocky Point could be due to increased phosphorus uptake by
Scotch broom.
The levels of nutrient availability detected at Rocky Point were much higher than those detected at Bamberton. This is not surprising given that the soils of Bamberton had a history of CaCC»3 deposition from dust emitted by the Bamberton cement plant during the first half of the 20th century (Robert Maxwell, personal communication). Concentrations of Ca+ were much higher at Bamberton than at Rocky Point and, at this pH (7.8), these are likely to bind anions such as phosphate and nitrate (Binkley and Vitousek 1989), rendering them less available for plant use.
2.4.2 Plant Diversity Broom presence was not correlated with a difference in either total plant species richness or evenness. However, the MDS ordinations indicated that there were differences in actual species identity and abundance between the sites and between the treatments, at both Bamberton and Rocky Point. The SIMPER analysis was useful for teasing out the identities of the species that were driving these observed differences between the broom and non-broom plots. The greatest differences between the two treatments were primarily attributed to an increase in exotic (mainly grasses) species with broom at Rocky Point and to a decrease in native species with broom at both Rocky Point and Bamberton. This conclusion supports the hypothesis of a negative correlation between broom invasion and native plant diversity. The question remains however, whether the introduced plants at
39 Rocky Point have in fact been facilitated by the broom or if it is simply a site effect where broom and other exotic species have had the opportunity to concurrently gain a foothold. Overall, these diversity results support the results of Parker et al. (1997) who also noted a negative relationship between broom invasion and native species richness in
Garry oak meadows in Washington state.
The fact that the broom plots at Bamberton did not have a large increase in introduced species is likely due to the relatively limited human access experienced at this location resulting in a minimal path for species introductions, humans being the main cause of invasive species introductions (Mack et al. 2000).
2.4.3 Bioassay
The bioassay showed the invasive grass, Dactylis, does not grow better than Festuca, a native grass, when grown with Cytisus, nor does Dactylis grow better with Cytisus than with Holodiscus, a native shrub. Therefore, it is possible that Dactylis is simply a larger growing plant than Festuca, regardless of the species with which it is grown or else that the grass growth was limited by low availability of magnesium or calcium (personal communication, Bill Herman, Pacific Soils Analysis). The management implications of this are that as long as it is able to become established, Dactylis will likely grow to a greater biomass than Festuca ultimately resulting in a greater capacity to compete with other plants. It is worth noting, however, that in the two sites studied here, Dactylis was not positively associated with broom-invaded plots, and at Rocky Point, it was actually more abundant in the non-broom plots, suggesting that Dactylis invasion is not necessarily a species of interest in terms of being linked to the presence of Scotch broom.
Though the results of the post-bioassay soils analyses did not indicate any increase in nitrogen availability, the possibility still remains that any additional soil nitrogen may not have been detected simply because it had been taken up by the plants. Due to cost constraints, however, the nitrogen concentration in plant tissues was not analyzed.
The greater number of flower spikes produced by Dactylis and Festuca when grown with broom could result in an enhanced reproductive output. This is interesting from a management perspective since greater seed set by a plant will increase its chance of spreading and should be considered in relation to the plant's phenology and the order of
40 establishment of the plant when attempting to predict which species will be most successful in invading a site (MacDougall and Turkington 2004).
Haubensak and Parker (2004) also conducted a similar bioassay on Achillea millefolium, a common native herbaceous perennial found in Garry oak ecosystems in Washington, to test if increased nitrogen mineralization in the broom-invaded soil would enhance growth. The plants were grown in soils collected from two Garry oak sites, one which had been invaded by broom for many years and one which had no broom. Despite higher nitrogen mineralization and nitrification in the broom-invaded soils, the Achillea actually grew ~ 30% less in the broom-invaded soil indicating possible counter-indicative effects of broom on plant growth which may be due to broom's production of quinolizidine alkaloid defensive compounds (Wink et al. 1983). The results of this present study do not corroborate those of Haubensak and Parker because broom had no effect on grass size. In this study, however, the broom was present for only twelve months, and therefore the negative effects of broom on other plant species through, for example, production of alkaloids, may only be detected in long-term studies.
41 2.5 CONCLUSION The invasive shrub Scotch broom was shown to be negatively correlated with native plant community structure and positively correlated with invasive plants at two sites on
Vancouver Island: Bamberton and Rocky Point. Yet, a link between the plant diversity changes and an increase in nitrogen availability was only suggested by a weak trend of increased NTH/ in the broom-invaded plots at Rocky Point. However, these and all other measurements of soil nitrogen at the various sites in both Bamberton and Rocky Point were overshadowed by high variability seen within the same broom treatment.
It is possible that the introduced plants in the broom plots were responsible for rapidly taking up any increased nitrogen made available by the broom. However, since the PRS probes were designed to capture plant-available nitrogen at the same rate as plant roots, this is unlikely. Measurements of plant foliar nitrogen in both the broom-invaded and non-broom plots could have elucidated this. Regardless, the high variation seen within the treatments suggests that nitrogen availability in the Garry oak ecosystem is more likely driven primarily by other soil and climate factors such as soil temperature, microsite moisture, soil physical structure and microbial community structure.
While the soil study indicated only a weak trend in increased soil nitrogen corresponding with the invasion by Scotch broom, broom's effects on soil nutrient cycling cannot be completely ignored as a mechanism for plant community change, as there was a significant decrease in phosphorus associated with broom invasion at Rocky Point. The surprising result of decreased phosphorus availability in the broom-invaded plots is interesting and may be worthy of further investigation to ascertain the mechanism by which this nutrient is made less available, i.e. through uptake by broom, and also to determine if, in fact, phosphorus availability is higher under broom due to the production of acid phosphatase from Scotch broom. It is, of course, always possible that the differences seen in soil nutrient availability between broom-invaded and un-invaded plots were not actually caused by the broom, but rather that these differences existed prior to broom invasion, and may have even had some influence on the invasibility of the plots. It would, therefore, be of use to science (though ethically unsound) to conduct a more
42 rigorous experimental study in which broom was planted into a pristine GOE meadow and the pre- and post-experiment soil nutrients analyzed and compared.
Meanwhile, several possible mechanisms other than increased fertility may explain the plant community shift observed in the diversity survey. For example, it is possible that shading, water competition and/or the production of alkaloid compounds by broom are responsible for the quantifiable differences observed between the broom-invaded and un- invaded plots. In fact it has been suggested before by Waterhouse(1988), that
"Allelopathic inhibition of plants beneath broom cannot be discounted without further investigation" given that there are known alkaloid compounds in broom plant tissues
(Wink et al. 1983). This short-term observational study, however, was not designed to control for all of these possible variables and therefore future experimental studies would benefit from quantifying these variables and taking them into account.
The bioassay experiment was designed to establish the direct effect of broom on the growth of two grasses, Dactylis glomerata and Festuca idahoensis. The results obtained rule out a short-term direct effect of broom on grass growth while at the same time suggesting that grasses produce more flowers when grown with broom. Whether the increased number of flowers is a response to higher nitrogen availability or to greater stress due to allelopathic compounds is unknown and would require further experimentation to determine the mechanism. In the future, it would also be useful from a management perspective to ascertain how much of an effect an increase in flower production has on the long-term success of Dactylis while taking into account the effects of establishment order, amount of disturbance and nutrient levels, similar to the work of
MacDougall and Turkington (2004) who showed that the competitive abilities of introduced and native grasses in a Garry oak meadow shifted with planting order.
Determining the most effective way to manage Garry oak ecosystems in order to maximize native plant communities depends on knowing the mechanism by which negative effects are brought about. It therefore continues to be a challenge to tease out the factor(s) responsible for the changes in plant community structure and will continue to require more experimental approaches to the question "By what mechanism does broom impact plant community structure and diversity in the Garry oak ecosystem?"
43 Hence this project is a small contribution to the expanding body of knowledge of invasion ecology of the Garry oak ecosystem of Southern Vancouver Island. While this work corroborates the plant diversity work done in the Garry oak glacial outwash prairies by Parker et. al (1997) and soil work done in Australia by Waterhouse (1988), it at the same time refutes the results of the soil nitrogen work done in the Garry oak glacial outwash prairies of Washington state by Haubensak and Parker (2004). It does, however, corroborate the phosphorus availability findings of Caldwell (2006) in California coastal prairies. The research also once again raises the question of the role of broom's alkaloid production in plant community structure and diversity in the Garry oak ecosystem.
44 2.6 REFERENCES
Binkley, D., and P. Vitousek. 1989. Soil nutrient availability. Pages 75-96 in R. W. Pearcy, J. R. Ehleringer, H. A. Mooney, and P. W. Rundel, editors. Plant Physiological Ecology - Field Methods and Instrumentation. Chapman and Hall, London. Bossard, C. C, and M. Rejmanek. 1994. Herbivory, growth, seed production and resprouting of an exotic invasive shrub Cytisus scoparius. Biological Conservation 67:193-200. Caldwell, B. A. 2006. Effects of invasive Scotch broom on soil properties in a Pacific coastal prairie soil. Applied Soil Ecology 32:149-152. COSEWIC. 2000. Canadian Species at Risk List. Pages available online at http://www.cosewic.gc.ca/COSEWIC/Default.cfm in. Committee on the Status of Endangered Wildlife in Canada, Ottawa, Ontario. Dancer, W. S., J. F. Handley, and A. D. Bradshaw. 1977. Nitrogen accumulation in kaolin mining wastes in Cornwall 1. Natural Communities. Plant and Soil 48:153- 167. DAntonio, C, and P. Vitousek. 1992. Biological invasions by exotic grasses, the grass/fire cycle and global change. Annual Review of Ecology and Systematics 23:63-87. Downey, P. O., and J. M. B. Smith. 2000. Demography of the invasive shrub Scotch broom (Cytisus scoparius) at Barrington Tops, New South Wales: insights for management. Austral Ecology 25:477-485. Facelli, J. M., and S. T. A. Pickett. 1991. Plant litter: its dynamics and effects on plant community structure. Botanical Review 57:1-32. Fogarty, G., and J. M. Facelli. 1999. Growth and competition of Cytisus scoparius, an invasive shrub, and Australian native shrubs. Plant Ecology 144:27-35. Fuchs, M. 2001. Towards a Recovery Strategy for Garry Oak and Associated Ecosystems in Canada: Ecological Assessment and Literature Review. Technical Report BGEI/EC-00-030 Environment Canada, Canadian Wildlife Service, Victoria. Gurevitch, J., and R. S. Unnasch. 1989. Experimental removal of a dominant species at two levels of soil fertility. Canadian Journal of Botany 67:3470-3477. Hangs, R. D., K.D.Greer, and C. A. Sulewski. 2004. The effect of interspecific competition on conifer seedling growth and nitrogen availability measured using ion-exchange membranes. Canadian Journal of Forest Research 34:754-761. Haubensak, K. A., and I. M. Parker. 2004. Soil changes accompanying invasion of the exotic shrub Cytisus scoparius in glacial outwash prairies of western Washington [USA]. Plant Ecology 175:71-79. Hibbard, K. A., S. Archer, D. S. Schimel, and D. W. Valentine. 2001. Biogeochemical changes accompanying woody plant encroachment in a subtropical savanna. Ecology 82:1999-2011. Hughes, R. F., and J. S. Denslow. 2005. Invasion by a N2-fixing tree alters function and structure in wet lowland forests of Hawaii. Ecological Applications 15:1615- 1628.
45 MacDougall, A. S., and R. Turkington. 2004. Relative importance of suppression-based and tolerance-based competition in an invaded oak savanna. Journal of Ecology 92:422 - 434. Maron, J. L., and P. G. Connors. 1996. A native nitrogen-fixing shrub facilitates weed invasion. Oecologia 105:302-312. Neufeld, J. H. 1980. Soil Testing Methods and Interpretations. British Columbia Ministry of Agriculture, Victoria. Parker, I. M. 1996. Ecological factors affecting rates of spread in Cytisus scoparius, an invasive exotic shrub. PhD Dissertation University of Washington. Parker, I. M. 2001. Safe site and seed limitation in Cytisus scoparius (Scotch broom): invasibility, disturbance and the role of cryptogams in a glacial outwash prairie. Biological Invasions 3:323-332. Parker, I. M., W. Harpole, and D. Dionne. 1997. Plant community diversity and invasion of the exotic shrub Cytisus scoparius: testing hypotheses on invasibility and impact. Pages 149-161 in P. V. Dunn and K. Ewing, editors. Ecology and Conservation of the Southern Puget Sound Prairie Landscape. The Land Conservancy, Seattle Washington. Pickart, A. J., L. M. Miller, and T. E. Duebendorfer. 1998. Yellow bush lupine invasion in Northern California coastal dunes: Ecological impacts and manual restoration techniques. Restoration Ecology 6:59. Prasad, R. 1999. Scotch broom, Cytisus scoparius L. in British Columbia, in. Pest Management Methods Network. R.N. Mack, D. Simberloff, W. M. Lonsdale, H. Evans, M.Clout, and F. Bazzaz. 2000. Biotic Invasions: Causes, Epidemiology, Global Consequences and Control. Issues in Ecology 5. Stock, W. D., K. T. Wienland, and A. C. Baker. 1995. Impacts of invading N2-fixing Acacia species on patterns of nutrient cycling in two Cape ecosystems: evidence from soil incubation studies and 15N natural abundance values. Oecologia 101:375-382. Vitousek, P. M., and L. R. Walker. 1989. Biological invasion by Myricafaya in Hawai'i: plant demography, nitrogen fixation, ecosystem effects. Ecological Monographs 59:247-265. Waterhouse, B. M. 1988. Broom (Cytisus scoparius) at Barrington Tops, New South Wales. Australian Geographical Studies 26:239-248. Weiss, S. B. 1999. Cows, cars and checkerspot butterflies. Conservation Biology 13:1476 - 1486. Wheeler, C. T., O. T. Helgerson, D. A. Perry, and J. C. Gordon. 1987. Nitrogen fixation and biomass accumulation in plant communities dominated by Cytisus scoparius L. in Oregon and Scotland. Journal of Applied Ecology 24:231-237. Williams, P. A. 1981. Aspects of the ecology of broom (Cytisus scoparius) in Canterbury, New Zealand. New Zealand Journal of Botany 19:31-43. Wink, M., L. Witte, T. Hartmann, C. Theuring, and V. Voltz. 1983. Accumulation of quinolizidine alkaloids in plants and cell suspension cultures: Genera Lupinus, Cytisus, Baptisia, Genista, Laburnum and Sophora. Planta Medica 48:253-257.
46 3. POTENTIAL OF FERTILIZER APPLICATION TO SUPPRESS SCOTCH BROOM RE-ESTABLISHMENT
3.1 INTRODUCTION The leguminous shrub Scotch broom (Cytisus scoparius) has been introduced to many regions around the world from its native Mediterranean habitat of Spain and southern
France. Through mostly intentional introductions, broom has become established in areas of Australia (Downey and Smith 2000), New Zealand (Partridge 1989) (Williams 1981) and the Pacific Coast region of the US (Bossard and Rejmanek 1994) and British
Columbia, Canada (Prasad 1999), where it has become a problematic weed.
Broom's weedy nature can be attributed to many of the characteristics that are typical of invasive plants. These include: quick, efficient growth that results in dense, sunlight- blocking stands; photosynthetic stems that enable it to grow year round; a release from natural enemies; alkaloid toxins that may be responsible for the inhibition of herbivory and/or other plants; symbiotic root associations with nitrogen-fixing bacteria which allow it to grow on nutrient-poor, disturbed sites; and the prolific production of long-lived seeds resulting in a copious seedbank. It is this last attribute that makes Scotch broom such a difficult plant to control and is, therefore, the aspect by which ecosystem managers are often defeated.
Standard methods of controlling broom consist of mowing, pulling, burning, herbicide application, introduction of grazing animals such as sheep and goats and attempts to establish effective biological control agents (Clark 2000, Hore 2000, Alexander and
D'Antonio 2003). The appropriateness and effectiveness of any of these methods is situation dependent, however, there is currently no magic bullet for controlling the surge of seedlings germinating from the seedbank following broom removal other than by repeated removals (Waterhouse 1988).
Three approaches are used for dealing with broom seedlings following mechanical removal.
1) Encourage a flush of seedlings, such as with fire, and then remove the new
seedlings before they mature and set seed. Repeat this method until the seedbank
is exhausted.
47 2) Create optimal conditions for competing vegetation such that germinated
broom seeds are unable to compete sucessfully.
3) Directly inhibit the germination of broom seeds.
The drawback with the first method is that repeated fire makes it difficult for native plants to become established unless they are adapted to fire. In addition, the added disturbance and the loss of nutrients associated with frequent fires may encourage other non-native plants to establish in broom's stead. And lastly, burning is often not an option for a variety of reasons, such as proximity to human habitation and/or utilities.
The second and third approaches show more promise and are the subject of the experiments conducted in this research. As a sun-loving nitrogen-fixer, broom typically outcompetes other plants in nutrient-poor, open-canopied sites, particularly those sites that have recently undergone disturbance. So, in order to encourage growth of alternative vegetation, the conditions responsible for limiting the other plants' growth must be mitigated. For instance, by improving the soil's fertility, other species such as grasses can grow more vigorously, thereby competing with broom for water and sunlight. In a study conducted by Bossard (1991), it was suggested that broom seedling establishment could be decreased by increasing vegetation cover.
A successful example of broom suppression by improving the soil's fertility is at
Discovery Park in Seattle, Washington where nitrogen-rich biosolids from the King
County Department of Natural Resources' West Point Sewage Treatment Plant were applied as soil conditioner to a landscape-sized restoration project in late summer of
1995. The overall goal of this project was to transform a 4.5 hectare broom-infested property to a self-perpetuating, biologically diverse, natural park area with open vistas for recreation and wildlife habitat (Deutsch 1997).
The restoration goals for Discovery Park were accomplished with the following steps:
The broom was bulldozed and removed, soils were tilled repeatedly, biosolids were then incorporated into the predominantly sandy soil at an application rate of 112 dry tonnes/hectare, followed by seeding with a mixture of grasses and native shrubs. More than ten years after the project's establishment, it is still very uncommon to see broom on the site. Instead, the landscape is a vast field of grasses (native and introduced), trailing
48 shrubs such as native Rubus ursinus, legumes and other herbaceous plants with scattered hedges of larger shrubs such as Elderberry (Deutsch 1997).
The success of this restoration project is undeniable, but the mechanism behind its success is unclear since no experimental trial or post-establishment monitoring was conducted at the site. One possible mechanism is that by increasing the soil-nitrogen levels of the site, the advantage that broom would typically gain over other plants by being able to fix nitrogen is negated. It is also possible that broom is hindered directly simply by increasing nitrogen availability.
Biosolids, however, are not only rich in nitrogen. They are composed of a suite of macro- and micro-nutrients as well as organic matter that all serve to condition the soil by improving its fertility, its water-holding capacity and its ion-exchange capacity - conditions that are conducive to high plant yield. Therefore, with this nutrient input it is possible that other plants are able to grow vigorously enough to out-compete broom seedlings for water and light.
The objectives of this study are to:
• Test the effectiveness of both biosolids and ammonium nitrate fertilizers on the
inhibition of broom re-invasion in a site with a large Scotch broom seedbank.
• Compare the effectiveness of ammonium nitrate to that of biosolids on the
inhibition of broom seedling growth.
• Ascertain the relative success of non-native grass growth and native shrub growth
in the different treatments and determine how the success of these species guilds
correlates to broom suppression.
• Determine whether broom re-establishment is suppressed directly by the
treatment(s) via inhibition of seed germination or indirectly via competition with
other plants that have been facilitated by increased nutrient supply.
49 3.2 METHODS
3.2.1 Sites
Three sites in British Columbia having the following attributes were chosen for the experiments: 1) A history of at least 5 years of dense broom that had flowered and produced seed, 2) little to no slope and 3) at least 30 metres away from any lakes, streams, wells and dwellings and at least 10 metres away from any roadways.
Site 1 - Iona Beach
49°13.01' N, 123°11.45' W
This property is situated in the Canfor block of Iona Beach Regional Park,
Richmond, British Columbia, which is an area that has no public access. The soil
is composed of sand tailings dredged from the Fraser River. The site was covered
in a dense stand of broom, with many shrubs c. 15 years and older. Other ground
cover consisted of Bromus tectorum, mosses and lichens. A slight slope of <2%
runs East-West. Twenty-one 3.5m x 3.5m plots were established in October 2003
and were blocked along the slope gradient with seven blocks of three plots each.
Broom removal occurred in July 2004 by hand-pulling the smaller shrubs and
brush-sawing the larger plants - all plants were then carried off site.
Site 2 - Burnaby Mountain Transmission Line Right-of-Wav
49°16.13' N, 122°53.80' W
This property runs along a Northeast - Southwest corridor between dense
Douglas-fir and maple forest. Public access along the trail adjacent to the site is
common, therefore snow-fencing was used to block access to people and animals
and project signs were posted. The loamy soil is > lm deep and was disturbed in
1999 to install a gas pipeline. The site was covered in a dense stand of even-aged
broom c. 5 years of age and producing flowers and seeds. Other ground cover
consisted of Rubus discolor (Himalayan blackberry) and a few grasses. There was
a very minor slope of <2%. In February 2005, entire broom plants, including roots, were removed by hand. In April 2005, twelve 6m x 2m plots were
established and blocked along the slope gradient in 4 blocks of 3 plots.
50 Site 3 - Duncan Transmission Line Right-of-Way
48°44.02' N, 123°42.98'W
This privately-owned property runs along an East-West corridor flanked on both
sides by Douglas-fir forest. The soil is loamy with broken boulders and has been
compacted by heavy machinery. The second generation broom was dense after
having been removed once before by heavy machinery. Other ground cover
consisted of mowed grasses and some Oregon grape (Mahonia nervosa). The site
is level. In May 2005, broom was removed by mowing, soil was loosened with an
excavator, roots and large rocks were removed by hand and twelve, 6m x 2m plots
were established in four blocks of three plots.
3.2.2 Site Preparation The three sites were established on different dates. The first site, Iona Beach was cleared in July 2004 then rototilled, fertilized and seeded on September 30, 2004. The Burnaby
Mountain site was cleared in mid-February, 2005 then rototilled, fertilized and seeded on two consecutive days in early April, 2005. The Duncan site was cleared, rototilled, fertilized and seeded on two consecutive days in mid-May, 2005.
Prior to plot establishment soil analyses were conducted for major nutrients (N, P, K, Ca,
Mg, Na) and trace elements (As, Cd, Cr, Co, Pb, Hg, Mo, Ni, Se, Zn) at all three sites
(Table 3.1). Samples were air dried and screened through a 2mm sieve. They were then analyzed for pH (determined potentiometrically on a 1:1 soil/distilled water slurry), total
C (determined directly on a LECO CR 12 Carbon Analyzer), Total Kjeldahl nitrogen or
TKN, (determined colourimetrically on a Technicon Autoanalyzer, using a semi-micro
Kjeldahl digest), ammonium (NFL;"1") and nitrate (NO3") (determined colourimetrically on a Technicon autoanalyzer using a 1:2 soil to solution 1M potassium sulphate extract), available P (determined colourimetrically using a 1:10 soil to solution Bray Pi extract) and Ca, K, Mg, and Na (determined on an Atomic Absorption Spectrophotometer using a
1:5 ammonium acetate extract) at Pacific Soils Analysis labs, Richmond B.C. Percent organic matter was calculated by multiplying total C x 1.724. Trace elements were
51 analyzed using Inductively Coupled Plasma Optical Emission Spectroscopy (ICP). This information was used for determining the maximum application rate of the biosolids based on nutrient and element analyses done on the biosolids (Appendix A).
Soil bulk densities were also determined for each site by removing soil and gravel from a small hole, measuring the volume of the hole by lining it with a plastic bag and filling it with water then measuring the volume of water held. The soil which was removed to create the hole was then oven-dried, large stones were removed and their volume subtracted from the overall water volume and the remaining soil (including some coarse fragment) weighed to determine the mass per unit volume (kg/m3) of the soil.
All plots were rototilled using a hand rototiller, seeded with native shrubs and grass seed
(Appendix B) and treated with either ammonium nitrate fertilizer at an application rate of
350kgN/ha applied in 3 installments or biosolids fertilizer at an application rate of 350 kgN/ha applied in 1 installment, plus an un-treated control. At Burnaby Mountain and
Duncan, an average of 74 ± 0.9 broom seeds per plot were also added to ensure a minimum number of seeds per plot because these sites had shorter histories of broom establishment and, therefore, had less substantial seedbanks.
The Burnaby Mountain site was fraught with the difficulties associated with being situated in a public area. On more than one occasion the plots were driven through by motor bikes, while on another occasion the stakes delineating the plots were removed and scattered about the site. Lastly, during a big windstorm, a large big-leaved maple (Acer macrophyllum) branch fell directly on one of the ammonium nitrate plots. The branches were soon removed, and though there seemed to be some imprint into the soil by the limbs, these were no-longer noticeable when the plant surveys were conducted.
52 BP TKN NH4* NO, C P K Ca Mg Na As Cd Cr Co Cu Pb Hg Mo Ni Date pH (g/cm3) (%) mg/kg % mg/kg Iona Beach 9/17/04 5.9 1.0 0.02 1.6 2.28 0.74 3.8 45 162 57 28 <10 <0.2 20.7 8.3 11.7 5 0.02 <4 29.7 Bby
Mtn. 4/22/05 4.9 1.5 0.3 11.3 11.4 4.9 27.7 62.3 111.7 34.3 85.7 <10 <0.5 11.7 5.3 13.7 11.3 0.1 <4 7.3
Duncan 5/17/05 5.2 1.5 0.17 14.0 7.80 3.8 32 84 512.5 58 43.8 <10 <0.5 26 12.7 37.0 8.3 0.04 <4 21.3
Table 3.1 Initial soil analysis results for the three sites prior to treatment. Average trace metal values were obtained from Cantest Laboratories (Burnaby B.C.) and bulk density was obtained following the procedure described in Methods. All other values were obtained from Pacific Soils Analysis (Richmond B.C.). N=4. 3.2.3 Biosolids and Application Rate
Biosolids used in this study are composed of de-watered, pasteurized sewage sludge from the Greater Vancouver Regional District's (GVRD) Annacis Island Wastewater
Treatment Plant. The dewatered organic material is recovered from residential (80%), industrial (10%) and commercial (10%) wastewater sources and undergoes anaerobic digestion at 55° C for 20 - 30 days resulting in a 99.99% pathogen-free, nutrient-rich fertilizer. Biosolids composition is a constant ratio of water, organic matter, soil, nutrients and trace metals.
Yearly average biosolids nutrient composition obtained from the GVRD was used to determine the appropriate biosolids application rate prior to fertilizing each site. To determine the actual application rates, post application analyses were conducted on a biosolids sample obtained from the batch of biosolids used at each of the three sites.
Application rates were determined as per the B.C. Biosolids Best Management
Guidelines (McDougall et al. 2001). Because nitrogen (N) is typically the limiting nutrient for most plant growth and is also the most likely nutrient in biosolids to cause ecosystem damage when over-applied, the biosolids application rate is based on the following four parameters:
1) Background soil N levels - TKN, N03", NH/
2) Biosolids N availability = sum of predicted mineralized nitrogen + available
NH/ + available NO3", as determined from lab analysis.
3) Expected loss of N due to volatilization during tilling - predicted to be 40% of
the original NH44" available in the biosolids.
4) The plants' N requirements - dependent upon the species present and whether
or not they are to be removed from the site as a crop.
For this study, what seemed to be a high N application rate (350 kg N/ha) was in fact appropriate for these sites owing to low soil nitrogen levels at all three sites and the fact that as a restoration project, the sites would not be fertilized more than once, nor would plants be removed.
54 Calculations for biosolids application rates for the three sites are found in Appendix A. In addition, a calculation was computed to ensure that biosolids application did not result in an over application of trace elements of concern by the Organic Materials Recycling
Regulations (OMRR) (McDougall et al. 2001) (Appendix C).
The ammonium nitrate application rate was also based on a total of 350 kg N/ha/year and calculations for this are found in Appendix D. To mimic the slow release of nitrogen from the biosolids, the total amount of ammonium nitrate to be applied was divided into three installments, two of which were applied during the first year (during site set-up and half-way though the growing season) while the third installment was applied at the beginning of the second growing season (May 2006). Three (Burnaby and Duncan) to five (Iona Beach) months after site establishment and prior to the second ammonium nitrate application, soils were sampled for a second time and analyzed for NFL;"1" and NO3" availability. A third soil sample was taken in May of 2006, and analyzed for TKN, NFL/ and NO3".
I predicted that NFL,"1" and NO3" would increase significantly in the ammonium nitrate and biosolids plots from the initial analyses to the first post-treatment analysis, and that the
biosolids plot would continue to show higher levels of NFL/ and N03" in the second analysis. It was also predicted that there would be consistently higher levels of NFL/,
NO3" and TKN in the ammonium nitrate and biosolids plots than in the control plots.
3.2.4 Monitoring broom re-establishment
Plant monitoring was conducted in the summer of 2005 at the time of peak grass flowering at each site. This was to coincide with maximum plant growth for the biomass sampling and for easier grass identification.
Broom seedling density
To determine if broom seedling survivorship was affected by treatment, plots were sampled using randomly-systematically arranged, lm x lm quadrats. Four quadrats were used at Burnaby Mountain and Duncan, while at Iona beach three quadrats per plot were used due to the smaller plot size. Broom seedlings were censused in each quadrat and averaged per square metre per plot. Broom seedling surveys were conducted at Iona
55 Beach in June, Burnaby Mountain in July and two times at Duncan, once in July and once in October, as a response to seemingly low seedling numbers in the first survey.
Percent broom cover, relative abundance of competing vegetation and species richness
Native grass seed that was obtained from a private seed producer (Many Vaartnou,
Richmond, B.C.) was sown in the experiment in an effort to shift the species composition from one that is dominated by exotic plants to one with a higher proportion of native plants. Grasses are considered to be good potential competitors with broom seedlings as they can be dense groundcover and can self-establish relatively quickly since most species go to seed within two years (Richards 1973). The concern that invasive exotic grasses may spread beyond the restoration area, can be mitigated by introducing native grass seed to the area. Given that these grasses have to compete with exotic grasses that may have already established, it was, therefore, practical and interesting to ascertain the relative success of introduced and native plants, especially that of grasses, in the different treatments.
To do this, percent cover of every plant species was determined for each plot using four random quadrats (0.5m x 0.5m) in each of the Duncan and Burnaby Mountain plots and three quadrats in the smaller Iona Beach plots. I adapted the Braun-Blanquet scale of percent cover (Mueller-Dombois and Ellenburg 1974) by splitting Class 3 into two classes in order to refine the scale of my data (Figure 3.2).
56 Class Braun-Blanquet Adapted Braun- % cover Blanquet % cover
1 <1 % 1 sighting
2 1-5% 1-5%
.3 6-25% 6 - 10%
4 26 - 50% 11-25%
5 51-75% 26 - 50%
6 76 - 100% 51 -75%
7 76 - 100%
Table 3.2 Percent cover and corresponding cover class as adapted from the Braun- Blanquet percent cover scale (Kent and Coker 1992).
These values were averaged across the number of quadrats surveyed in each plot to obtain one value per species present per plot. (For species lists, see Appendix E). Percent cover of broom was compared across treatments at each site. As well, species were grouped into three guilds (grasses, introduced grasses and native shrubs), and their percent cover relative to the total plant cover was compared across treatments. In addition, the average species richness was determined for each treatment and compared across treatments.
Competitive vegetation
To determine the effect of treatment on overall vegetation yield, a subsample of above- ground biomass was collected from each plot within a week following the plant surveys.
Each subsample consisted of a 0.5m x 0.5m quadrat systematically arranged in each plot
57 from which all above-ground vegetation was clipped at the soil surface. Total vegetation was oven-dried to a constant weight then weighed to the nearest 0.0lg on an electronic digital scale in the laboratory. This estimate of productivity by treatment was compared with the broom seedling density data in an effort to account for seedling density results.
3.2.5 Broom germination experiment
To test the direct effect of fertilization treatment on broom seed germination under controlled conditions, a greenhouse-based pot experiment was conducted from August
28, 2005 until September 30, 2005.
Broom seeds collected from mature broom pods were checked for insect damage and sorted then scarified using 120 grit sandpaper. Soil used for potting was collected from the Duncan site, air dried, sieved to 2mm, then autoclaved to inhibit any weed growth or disease. Eight seeds were placed in each of twenty-seven, 4-inch pots, which were then randomly filled with soil that had been treated with biosolids, ammonium nitrate or unfertilized control.
Fertilized soils were treated with biosolids or ammonium nitrate to an amount equivalent to 350 kg/ha of nitrogen available per year. For the biosolids, these rates were based on laboratory analyses of the biosolids which were used to calculate a field application rate in kg/ ha (as per Appendix A), then a theoretical volumetric field application rate was calculated assuming an application depth of 10cm in soil with a bulk density of 1.5 g/cm
(determined in the field). This rate was then extrapolated for the volume of soil used in 9,
4-inch pots, resulting in a total of 130 grams biosolids being applied. The ammonium nitrate application rate was based on applying an equivalent of the first 1/3 of the yearly nitrogen availability rate of 350kg N/ha/year (as per Appendix D) to soil 10cm deep with a bulk density of 1.5 g/cm3, then extrapolating it to the volume of soil used in the experiment. The resulting amount was 1.88 grams of ammonium nitrate, which was dissolved in tap water then added to the dried soil. All soils were watered to field capacity with tap water.
Pots were filled to 1cm of the rim, eight seeds per pot were sown to 2mm depth and marked with toothpicks. Pots were placed onto individual saucers and arranged randomly in the greenhouse under a water mister for the first week of the experiment: After the first
58 week, germinated seeds were counted and left to continue growing. Consecutive seedling surveys were conducted over the next 4 weeks to tally the live seedlings and those that died. Cumulative totals of germinants and of seedlings that died were summed at the end of the first two weeks and again at the end of the last two weeks of the experiment to determine if there was a temporal effect on the survivorship of broom seedlings.
3.2.6 Statistical analyses Comparisons of soil NFL/ and NO3" availability between pre-treatment sampling and the first and second years following fertilization were conducted using t-tests (JMP 5.1).
Comparisons of the soil analysis results between treatment types for a given date were done using one-way ANOVAs (JMP 5.1).
General linear models (SAS version 9) using site, block, treatment and site-by-treatment were used to determine the effect of fertilization on: broom seedling number, percent cover of broom, competitive vegetation biomass and species richness as well as on the relative abundance of the following species guilds: invasive grasses vs other plant species; total grasses vs other plant species; and native shrubs vs other plant species. Iona
Beach was analyzed separately from Burnaby Mountain and Duncan for broom seedling density, since there were so many more broom seedlings counted at Iona Beach, possibly a result of having been established the previous fall.
One-way ANOVAs (JMP 5.1) were used to determine the effect of treatment on broom seed germination after two and four weeks of the germination experiment.
The level of significance for all statistics was set at 5%, and Bonferroni adjustments were made to the significance level for multiple comparisons.
59 3.3 RESULTS
3.3.1 Soils
The post-treatment soil samples that were taken three to five months after each site's initial establishment, give an indication of the relative persistence of available soil nitrogen following fertilization (Table 3.3).
60 + Pre-treatment Analysis NH N03 TKN 1 4 1 AMM AMM AMM Site NH4+ N03- TKN Date NI BIOS CTRL NI BIOS CTRL NI BIOS CTRL 10.0 ± 3.0 ± 2.1 ± 3.1 ± 0.5 ± 0.3 ± Iona Beach 1.6 2.28 0.02 Jun-05 2.5 0.7 ; 0.5 1.2 0.4 0.3 n/a n/a n/a 1.0 + 1.6 ± 1.6 + 1.2 ± 1.2 ± i.o ± 0.07 ± 0.08 ± 0.06 ± May-06 0 0.8 0.9 0.4 0.4 0.6 0.01 0.01 0.01 14.25 + 19.5 ± 15.5 ± 17.5 + is.o± 14.0 ± v Burnaby Mtn 11.3 11.4 0.3 I Jul-05 4.0 4.0 3.6 3.1 4.5 5.2 n/a n/a n/a 3.3 ± 4.5 ± 4.0 ± 1.8 ± 1.6 ± 1.1 ± 0.3 ± 0.37 ± 0.32 ± May-06 0.5 1.0 1.2 0.6 0.9 0.5 0.01 0.01 0 57.25 ± 72.25 ± 9.5 ± 151.5 ± 82.75 ± 49.25 ± Duncan 14.00 7.80 0.17 1 Jul-05 11.4 10.6 2.5 18.4 22.3 8.2 n/a n/a n/a 4.5 ± 8.0 ± 4.5 ± 3.9 ± 6.1 ± 4.4 ± 0.24 ± 0.34 ± 0.14± May-06 1.7 2.9 2.4 1.1 1.1 1.9 0.07 0.08 0.01
Table 3.3 Results of soil analyses for NFL;"1" and NO3" taken 3 to 5 months after initial site set-up and one year after site set-up (see Table 1 for initial dates and other nutrient values). NHV" and NO3" values are in ppm; TKN is in %. Errors are SE. N= 7 for Iona Beach and N= 4 for Burnaby Mountain and Duncan. Comparisons ofNH4+, NO3 and TKN levels pre- and post-fertilization
In the first growing season (2005), NFL/ increased significantly in the ammonium nitrate treated plots at Iona Beach while the levels of NO3" actually decreased in the biosolids and control plots. At Burnaby Mountain, though the NFL/ and NO3" levels tended to increase in all three treatments, they were not significantly different from the pre- treatment values. At the Duncan site, NFL/ increased in both the ammonium nitrate plots and the biosolids plots while NO3" increased in all three treatments (Table 3.4).
By the second growing season (2006), NFL/ levels had decreased significantly from the original levels in the ammonium nitrate treatment plots at Iona Beach and remained
unchanged in the other two treatments, while the N03" levels had increased in all three
treatments. At Burnaby Mountain, both NFL/ and N03" had decreased in all three treatments, while at Duncan, NFL/ decreased in the control and ammonium nitrate plots, and NO3" remained the same as original levels in all three treatments.
Total Kjeldahl nitrogen (TKN) increased in all three treatments at Iona Beach, and remained the same in all treatments at both Burnaby Mountain and Duncan (Table 3.4.)
62 Ammo nium Nitrate Biosolids Control
+ + NH4 N03- TKN NH4 NO, TKN NH, NO3- TKN
Iona 7 = 1.5 Same N/A Same F == 13.7 N/A Same I F = 17.6 N/A Beach p = 0.02 T p =-0.00= 4 lpp = 0.002 Bby. Same Same N/A Same Same N/A Same Same N/A 03 Mtn. Duncan ^ F = 13.3 + F =- 59.2 N/A +F = 27.48 • F = 11.1 N/A Same + F = 22.4 N/A I p = 0-/01 lpp == 0.001 I p = 0.002 I p = 0.-02 p = 0.003 Iona J F = 7.8 t F = 14.1 + F = 20.251 Same j F = 14.1 |F = 13.78 Same t F = 15.2 t F = 26.7 Beach *p = 0.02 I p = 0.004 I- p = 0.001 I p = 0.004 lnp. = 0.004 I p = 0.003 I p = 0.007 03 Bby. F = 24.4 F == 142.2 Same I F = 16.2 I F = 125.1 Same I F = 18.2 I F = 170.2 Same Mtn. •p = 0.00.3 To.p.<< 0.001. • P=:p:01 •p < 0.001 • p = 0.01 •p < 0.001 Duncan I F = 7 Same Same Same Same Same f F = 7.1 Same Same o = 00:0:l 3 I p = 0.04 Table 3.4 Summary of comparisons of soil NFL/1", NO3" and TKN changes between pre-treatment sampling and the first and second years (2005, 2006) following fertilization. Iona Beach, N =7; Burnaby Mountain and Duncan, N = 4. as LO Comparisons ofNH/ andNO^ levels between treatments During the first growing season (2005), at Iona, the NH4+ availability was greater in the ammonium nitrate treated plots than in both the biosolids and control plots, while there were no significant differences in NO3" between any of the treatments. At Burnaby Mountain, there was no difference in NFL* or NO3" levels between any of the treatments. At Duncan, both the ammonium nitrate and biosolids treatment plots had higher NEV" than did the control plots, while the NO3" levels were higher in the ammonium nitrate treatment plots than in the biosolids and the control plots. By the second growing season (2006), there were no differences in either NtVor NO3" between any of the treatments at any of the sites (Figure 3.1). 64 Iona Beach p = 0.004 Ammonium Nitrate Biosolid Control 05 NH4+ 05 N03- 06 NH4+ 06 N03- Figure 3.1 NFL/ and N03" availability in ammonium nitrate, biosolid and un-treated plots at Iona Beach, Burnaby Mountain and Duncan, 1 and 2 years after fertilizer application (05, 06). Iona Beach, N = 7; Burnaby Mountain and Duncan, N = 4. Bars followed by different letters are significantly different, p < 0.05. 65 Cation exchange capacity differences between treatments lacked a consistent pattern at any of the three sites (Table 3.5). At the Duncan site, the CEC was improved in the biosolids treatment, whereas at the two other sites, the CEC was actually lower in the treatment plots than in the control plots. Site Cation Exchange Capacity Amm. Nitrate Biosolids Control Iona 4.02 3.93 4.29 Burnaby 30.8 31.7 36.6 Duncan 15.2 29.5 16.1 Table 3.5 Cation exchange capacity (meq/lOOg) by treatment at all three sites one year following site establishment (May, 2006). 3.3.2 Monitoring broom re-establishment Broom seedling re-establishment At all three sites, biosolids treated plots had the lowest broom seedling density of the three treatments, differences which were significant at both Burnaby Mountain and Duncan (F(2,i2) = - 4.20, p = 0.0012) (Figure 3.2). 66 140 Ammonium Nitrate Biosolids 120 Control c 100 80 GO 40 20 H a b b cd T Iona Beach Burnaby Mtn Duncan Sites Figure 3.2 Mean number of broom seedlings by treatment at all three sites, sampled in July 2005. Iona Beach, df = 2,18 ; Burnaby Mountain and Duncan, df = 2,12; Different letters denote significantly different values, p < 0.05. Error bars are SE. 67 Percent cover of broom, relative abundance of competing vegetation and species richness At both Iona Beach and Burnaby Mountain percent cover of broom was lowest in the biosolids plots, while at Duncan, both the ammonium nitrate and biosolids plots tended to have lower percent cover of broom than the control plots (Figure 3.3, Table 3.6). Iona Beach Bumalty fvltn. £ Duncan o o CD 3 CT (0 C O «5 CQ • 5 C o 3 1 1 CQ C (0 CD Amm. Ni. Bi osolid Control Treatment Figure 3.3 Mean percent-cover class of broom for each treatment at all three sites. Asterisks denote differences between treatments significant to p < 0.0028. Treatment Site Amm. Ni. Biosolids Control F P Iona Beach 1.38 (0.38) 0.28 (0.14) 1.71 (0.10) 15.39 0.0005 Burnaby Mtn. 1.50 (0.38 0.13 (0.16) 2.63 (0.48) 8.8 0.0123 Duncan 0.13 (0.07) 0.25 (0.16) 0.87 (0.20) 6.88 0.0321 Table 3.6 Mean percent-cover class of broom for each treatment at three sites. (SE in brackets). 68 The relative proportion of grasses to other species in the plots was higher in the biosolids treatment than in the control (F(2,24) = 3.69, p = 0.001), non-native grasses from the seedbank being the predominant group of grasses, these also being higher in the biosolids treatment than in the control (F(224) = 2.57, p = 0.017). There was no difference in relative abundance of native shrubs between treatments (Table 3.7). Table 3.7 Average relative abundance of grasses, non-native grasses and native shrubs for all three sites for each treatment. Iona Beach, N = 7; Burnaby Mountain and Duncan, N = 4. AH Grasses Non-native Grasses Native Shrubs Ammonium Nitrate 57 % 46 % 8 % Biosolids 62 % 47 % 8 % Control 50% 37 % 8 % 69 At all three sites, the control plots had a greater number of species than the biosolids treated plots (F(2,24) = 2.70, p = 0.01) (Table 3.8). Average number of species ± sd Site N Am. Nitrate Biosolids Control Iona 7 11 ±0.9 10 + 0.6 11 ±0.6 Burnaby 4 7 ±0.4 5± 1.1 8 ±0.6 Duncan 4 17 ± 1.0 16 ± 1.1 18 ±0.8 Table 3.8 Average plant species richness by treatment at each of the three sites. Competitive vegetation biomass There was a significant interaction between site and treatment for biomass production, therefore treatment effects were compared separately for each site. At all three sites, the unfertilized control plots produced consistently lower biomass than either of the other two treatments. At Iona Beach, the biosolids plots produced greater vegetation biomass than either the ammonium nitrate or the control plots, while at Burnaby Mountain, though the biosolids plots demonstrated higher productivity than both of the other treatments, they were only significantly higher than the control plots (Figure 3.4). Vegetation productivity was very similar between the two treatments and the control at the Duncan site. 70 140 Ammonium Nitrate Biosolid Control Iona Beach Burnaby Mtn. Duncan Sites Figure 3.4 Dry vegetation biomass for all three treatments at each of the three sites. Error bars = SE. df = 2,24. Iona Beach, N = 7; Burnaby Mountain and Duncan, N = 4. Bars followed by different letters are significantly different at p < 0.05. 3.3.3 Broom germination experiment Biosolids and ammonium nitrate addition had no effect on the germination of broom seeds at 2 or 4 weeks after sowing. (Mean seedlings germinated: Ammonium nitrate = 7.00, Biosolids = 6.33, Control = 6.89, F(2>24) = 0.89, p= 0.42). There was, however, a treatment difference in the number of seedlings that died. No seedlings died in controls even after 4 weeks, whereas in the biosolids treated pots, 10% of the seedlings died (n=9, SE = 0.43) and in the ammonium nitrate plots, 21% of the seedlings died (n=9, SE = 0.44) (F(2>24) = 4.76, p = 0.02) (Figure 3.5). 8 1 Germ.2wks Died2wks Germ. 4wks Died4wks Seedling fate at 2 and 4 weeks Figure 3.5 Fates of broom seeds and seedlings under different soil fertilizer treatments. Different letters denote significantly different values, p < 0.05. N = 9; Error bars are SE. 72 3.4 DISCUSSION Caveat The findings presented here represent the results of the first year of what will be a three- year monitoring project assessing the potential of fertilization to inhibit Scotch broom (Cytisus scoparius) growth. It is, therefore, important to bear this in mind when discussing the results as it is quite possible that the overall outcome of the study will be different from these preliminary results. 3.4.1 Soils Neither the NFL/ nor the NO3" levels responded as predicted when compared between sites and between treatments in the first year following treatment. This could be due to three possible factors: the short-lived nature of these molecules, the rate of nutrient uptake by the plants and the influence of site-specific soil processes influenced by moisture and temperature. At Iona Beach, it is interesting to note that despite the low values of NFL/ and NO3" in comparison to the two other sites, the levels of available nitrogen in the ammonium nitrate plots at Iona Beach are relatively high compared to those observed in the other treatments at the same site. The higher nitrogen availability corresponds with lower competitive vegetation biomass than in the biosolids plots, suggesting that plants in the ammonium nitrate plots at Iona Beach were not able to take up all the nitrogen available to them because their growth was limited by other macro nutrients, such as P and K, which were initially very low (Neufeld 1980) and were only ameliorated in the biosolids treated plots. Not only did the ammonium nitrate plots at Iona Beach produce less competitive vegetation, but they also corresponded with a higher number of broom seedlings, whereas the biosolids plots produced the lowest number of seedlings with the highest amount of competitive vegetation. This suggests that in sandy sites such as Iona Beach, increasing the nitrogen alone does not suppress broom, but increasing the biomass of competitive plants, by improving the soil's overall fertility, does suppress broom. When TKN was measured in 2006, it had increased in all three treatments at Iona Beach, but not at either of the other two sites. Relative to the extremely low initial values of soil 73 nitrogen and the high C:N ratio (37) seen at Iona Beach, the additions of ammonium nitrate and biosolids seem to have made a comparatively large contribution to the overall nitrogen level. It is, therefore, not surprising that there was a notable increase in TKN in the ammonium nitrate and biosolids plots. The increase in TKN levels in the control plots may be explained by increased soil aeration from rototilling the plots, which could have led to greater decomposition by soil microbes. At Burnaby Mountain, the slight increases in NFL;4- and NOV availability observed in all the treatment plots in the first analysis following treatment, coupled with the initial C:N ratio of 16.3, indicates that there was likely a sufficient amount of nitrogen for plant growth (Bill Herman, Pacific Soil Analysis, Personal Communication) at this site prior to treatment and that the nutrient additions were comparatively small relative to the initial total soil nitrogen. The moderate levels of initial soil nitrogen make sense given the several years of invasion by Rhizobium-nodulated Scotch broom in a soil with good cation-exchange capacity (30.8 - 36.6 meq/lOOg). It is also possible that the soil analyses offer only one small snapshot into the seasonal nitrogen dynamics and that the plants simply had not yet had a chance to incorporate the nitrogen that was made available to them. However, given the difference observed in vegetative biomass production (higher in the biosolids plots), it seems more probable that plants grew to their potential and were limited by nutrients other than nitrogen. Meanwhile, it is interesting to note the nitrogen availability at Duncan which, in the treatment plots, was up to an order of magnitude higher than those values seen at Iona Beach and Burnaby Mountain, yet in the control plots, the NH4*" and N03" values were consistent with those seen at the other two sites. Despite these high values in the treatment plots, there was no significant difference in competitive vegetation biomass production between treatments, nor was there any real difference in numbers of broom seedlings, unlike at Iona Beach and Burnaby Mountain. These results suggest that by establishing the Duncan site in mid-May, which is well into the growing season on Vancouver Island, the plants that grew in these plots did not have enough time to grow large enough to take advantage of the available nitrogen. 74 Given the transient nature of NIL,"1" and NO3", it remains difficult to ascertain the overall nitrogen dynamics based on one soil sample per season, especially since predicting peak nitrogen availability (which should also correspond with peak plant growth) is so challenging. The timing of the 2005 soil samples was likely too late in the season to catch peak N-availability therefore the 2006 data were collected earlier in the growing season (May 2006). However, despite the earlier sampling date, the 2006 nitrogen availability data still does not display any consistent trends among the sites, nor are there any differences between the treatments. A better indication of the plants' overall seasonal nitrogen uptake in each treatment may have been obtained by measuring foliar nitrogen, while the use of ion-exchange membranes inserted in the soil over the course of the entire growing season, would have given a better approximation of the seasonal availability of soil nitrogen. The NFL/ and NO3" data from 2006 cannot be compared with plant productivity data since the plant data are not yet available, however some preliminary conclusions can be made. The decrease in available nitrogen from 2005 levels seen in the control plots as well as the treatment plots indicates that the earlier sampling date coincided with greater plant growth activity than the previous year, hence most of the available nitrogen was taken up by plants rather than lingering in the soil. Therefore, it is doubtful that these soil data give an accurate indication of overall nitrogen availability at the three sites, lending greater credence to the usefulness of ion-exchange membrane probes for future monitoring of seasonal nutrient availability. 3.4.1 Plant Monitoring At all three sites, broom seedlings tended to be fewer in the biosolids plots than in either of the other treatments. At the Duncan site, however, few seedlings occurred in any of the treatments. It is unclear why there were so few seedlings at the Duncan site, although there are several possibilities. Firstly, Duncan was the last site to be established and this may not have given the broom seeds enough time to germinate by the time of monitoring. In Ingrid Parker's study on the demographics of Scotch broom in the Pacific Northwest (Parker 2000), she reports that broom seeds begin to germinate in early March, though some germination occurs throughout the summer. This may also explain why there were 75 fewer seedlings at Burnaby Mountain, in comparison to Iona Beach, since the Burnaby Mountain site was established shortly before the Duncan site. It is also probable that, despite the addition of broom seeds to both the Burnaby Mountain and Duncan sites, there may have still been more seeds in the Iona Beach seedbank. It. was impossible to quantify the number of seeds present at any of the three sites prior to establishing them. However, based on the histories of the individual sites it is quite likely that there would have been more seeds in the Iona Beach seedbank since this site had the longest history of broom and had never been disturbed, which allowed more time for seeds to accumulate. Nevertheless, the fact that there were fewer broom seedlings in the biosolids plots than in either of the other treatments supports the idea that biosolids fertilization is a good way to suppress Scotch broom. While it was initially postulated that Scotch broom may in fact be suppressed directly by increased nitrogen, the less negative effect of the ammonium nitrate treatment on broom seedling success, coupled with the inverse relationship seen between broom seedling numbers and vegetation biomass, suggests the mechanism by which broom is suppressed by biosolids is an indirect effect of increased competition with other vegetation. In 2005, the vegetation was composed primarily of grasses, the majority of which were non-native, nitrophyllic forage grasses such as Bromus tectorum (cheatgrass), Anthoxanthum odoratwn (sweet vernalgrass) and Holcus lanatus (common velvetgrass) which is a common result of biosolids-fertilized sites (Richards 1973, Deutsch 1997). These results support the well-known outcome of competition between legumes, such as clovers, and forage grasses in high nitrogen situations, where the grasses, if left un- mowed, shade out the smaller nitrogen-fixing legume (Baylor 1974). The conclusion of the mechanism by which broom is suppressed is also supported by the results of the germination experiment, where the effect of competition with other plants was controlled for and there was no difference in germination between treatments. The pattern of percent cover of broom observed in the three sites mirrors that seen for the number of broom seedlings, which is not surprising since the data are correlated. However, in future years of monitoring, broom seedlings will likely grow at different 76 rates, making the percent cover of broom seedlings a better measure of broom success as time progresses. Broom counts, on the other hand, will remain useful as a way of tracking plant demographics by using life tables (Myers and Bazely 2003). Likewise, the low percent cover and the similarity of relative abundance of native shrubs between treatments, is not of great concern in this, the first year of monitoring. As plant succession progresses, the shrub species will increase in size, along with the broom plants, making the relative abundance of native shrubs more relevant to the successful suppression of broom by competition with native plants. 3.4.3 Germination experiment Broom seeds did not exhibit any difference in their germination response when sown in soil fertilized with either ammonium nitrate or biosolids under controlled conditions in the greenhouse, which suggests that any differences in suppression seen in the field would not be due to direct effects of increased nitrogen. This result supports the notion that broom control by biosolids fertilization is more likely caused by increased competition by other plants whose growth has been encouraged by biosolids addition (Deutsch 1997). The significant difference in the number of seedlings that died during the experiment is interesting however, and though monitoring the deaths of seedlings was not the goal of the experiment, and hence was not monitored for a very long period, it might be worthwhile to conduct a follow-up experiment to ascertain the potential for seedling survival in different treatments. 77 3.5 CONCLUSION By demonstrating fewer broom seedlings and lower percent cover of broom in the biosolids-treated plots, these preliminary results support the use of biosolids soil amendments as a means of suppressing Scotch broom re-invasion. Though the results may not be quite as dramatic as those seen at Discovery Park in Seattle, Washington, (Deutsch 1997), where biosolids were incorporated at a rate that is six times greater than the rate used in these experiments, the intermediate effects seen in these field experiments remain promising as a method to reduce Scotch broom re-invasion in previously disturbed sites, in particular those adjacent to sites of high conservation concern (such as the endangered Garry oak ecosystem of Southern Vancouver Island and the Gulf Islands, British Columbia). The results of the competitive vegetation biomass measures coupled with those of the germination experiment support the conclusion that any broom suppression related to biosolids fertilization is caused by competition with other vegetation whose growth was improved by inputs of nutrients. It is unfortunate that the most successful competitive vegetation guild is that of the introduced grasses, but once again, these data represent the first season of a three-year monitoring project, and many of the native grasses that were planted are perennial species, which require two seasons to reach maturity and thus they may show increasing trends in the future. Therefore, further monitoring in 2006 and 2007 will shed more light on the effect of biosolids on broom recruitment. 78 3.6 REFERENCES Alexander, J. M., and C. M. D'Antonio. 2003. Control methods for the removal of French and Scotch broom tested in coastal California. Ecological Restoration 21:191-198. Baylor, J. E. 1974. Satisfying the nutritional requirements of grass-legume mixtures, in D. A. Mays, editor. Forage Fertilization. American Society of Agronomy, Crop Science Society of America, Soil Science Society of America, Madison Wis. Binkley, D., and P. Vitousek. 1989. Soil nutrient availability. Pages 75-96 in R. W. Pearcy, J. R. Ehleringer, H. A. Mooney, and P. W. Rundel, editors. Plant Physiological Ecology - Field Methods and Instrumentation. Chapman and Hall, London. Bossard, C. C. 1991. The role of habitat disturbance, seed predation and ant dispersal on establishment of the exotic shrub Cytisus scoparius in California. The American Midland Naturalist 126:1-16. Bossard, C. C, and M. Rejmanek. 1994. Herbivory, growth, seed production and resprouting of an exotic invasive shrub Cytisus scoparius. Biological Conservation 67:193-200. Caldwell, B. A. 2006. Effects of invasive Scotch broom on soil properties in a Pacific coastal prairie soil. Applied Soil Ecology 32:149-152. Clark, A. 2000. Controlling broom (Cytisus scoparius) in pasture on the Barrington Tops - a graziers perspective. Plant Protection Quarterly 15:161-162. COSEWIC. 2000. Canadian Species at Risk List. Pages Available online at http://www.cosewic.gc.ca/COSEWIC/Default.cfm in. Committee on the Status of Endangered Wildlife in Canada, Ottawa, Ontario. Council, F. B. 2004. Invasive Plant Strategy for British Columbia. Fraser Basin Council, Vancouver, B.C. Dancer, W. S., J. F. Handley, and A. D. Bradshaw. 1977. Nitrogen accumulation in kaolin mining wastes in Cornwall 1. Natural Communities. Plant and Soil 48:153- 167. D'Antonio, C, and P. Vitousek. 1992. Biological invasions by exotic grasses, the grass/fire cycle and global change. Annual Review of Ecology and Systematics 23:63-87. Deutsch, B. L. 1997. Controlling Scotch broom in landscape design: case studies in the Pacific Northwest using soil amendments with biosolids. Master's Thesis. University of Washington, Seattle. - Downey, P. O., and J. M. B. Smith. 2000. Demography of the invasive shrub Scotch broom (Cytisus scoparius) at Barrington Tops, New South Wales: insights for management. Austral Ecology 25:477-485. Facelli, J. M., and S. T. A. Pickett. 1991. Plant litter: its dynamics and effects on plant community structure. Botanical Review 57:1-32. 79 I Fogarty, G., and J. M. FacelH. 1999. Growth and competition of Cytisus scoparius, an invasive shrub, and Australian native shrubs. Plant Ecology 144:27-35. Fuchs, M. 2001. Towards a Recovery Strategy for Garry Oak and Associated Ecosystems in Canada: Ecological Assessment and Literature Review. Technical Report BGEI/EC-00-030 Environment Canada, Canadian Wildlife Service, Victoria. Gurevitch, J., and R. S. Unnasch. 1989. Experimental removal of a dominant species at two levels of soil fertility. Canadian Journal of Botany 67:3470-3477. Hangs, R. D., K.D.Greer, and C. A. Sulewski. 2004. The effect of interspecific competition on conifer seedling growth and nitrogen availability measured using ion-exchange membranes. Canadian Journal of Forest Research 34:754-761. Haubensak, K. A., and I. M. Parker. 2004. Soil changes accompanying invasion of the exotic shrub Cytisus scoparius in glacial outwash prairies of western Washington [USA]. Plant Ecology 175:71-79. Helgerson, O. T., J. C. Gordon, and D. A. Perry. 1984. N2 fixation by red alder {Alnus rubra) and Scotch broom (Cytisus scoparius) planted under precommercially thinned Douglas fir (Pseudotsuga menziesii). Plant and Soil 78:221-233. Hibbard, K. A., S. Archer, D. S. Schimel, andD. W. Valentine. 2001. Biogeochemical changes accompanying woody plant encroachment in a subtropical savanna. Ecology 82:1999-2011. Hore, C. 2000. Herbicides for broom (Cytisus scoparius (L.) Link): testing alternatives to Grazon. Plant Protection Quarterly 15:167-168. Hughes, R. F., and J. S. Denslow. 2005. Invasion by a N2-fixing tree alters function and structure in wet lowland forests of Hawaii. Ecological Applications 15:1615- 1628. Kent, M., and P. Coker. 1992. Vegetation description and analysis. CRC Press, Boca Raton, Ann Arbor. Leccese, M. 1996. Fresh Fields. Pages 44-47 in Landscape Architecture. Lodge, D. M. 1993. Biological Invasions. Trends in Ecology and Evolution 8:133-137. MacDougall, A. S., and R. Turkington. 2004. Relative importance of suppression-based and tolerance-based competition in an invaded oak savanna. Journal of Ecology 92:422 - 434. Mack, R. N., D. Simberloff, W. M. Lonsdale, H. Evans, M. Clout, and F. Bazzaz. 2000. Biotic Invasions: Causes, Epidemiology, Global Consequences and Control. Issues in Ecology 5. Maron, J. L., and P. G. Connors. 1996. A native nitrogen-fixing shrub facilitates weed invasion. Oecologia 105:302-312. McDougall, N. R., M. D. VanHam, and M. J. Douglas. 2001. Best Management Practices Guidelines for Biosolids Land Application in British Columbia. Mueller-Dombois, D., and H. Ellenburg. 1974. Aims and Methods of Vegetation Ecology. James Wiley and Sons, New York. 80 Myers, J. H., and D. Bazely. 2003. Ecology and Control of Introduced Plants. Cambridge University Press, Cambridge. Neufeld, J. H. 1980. Soil Testing Methods and Interpretations. British Columbia Ministry of Agriculture, Victoria. Parker, I. M. 1996. Ecological factors affecting rates of spread in Cytisus scoparius, an invasive exotic shrub. PhD Dissertation University of Washington. Parker, I. M. 2000. Invasion dynamics of Cytisus scoparius: a matrix model approach. Ecological Applications 10:726 - 743. Parker, I. M. 2001. Safe site and seed limitation in Cytisus scoparius (Scotch broom): invasibility, disturbance and the role of cryptogams in a glacial outwash prairie. Biological Invasions 3:323-332. Parker, I. M., W. Harpole, and D. Dionne. 1997. Plant community diversity and invasion of the exotic shrub Cytisus scoparius: testing hypotheses on invasibility and impact. Pages 149-161 in P. V. Dunn and K. Ewing, editors. Ecology and Conservation of the Southern Puget Sound Prairie Landscape. The Land Conservancy, Seattle Washington. Parker, I. M., D. Simberloff, W. M. Lonsdale, and e. al. 1999. Impact: toward a framework for understanding the ecological effects of invaders. Biological Invasions 1:3-19. Partridge, T. R. 1989. Soil seed banks of vegetation on the Port Hills and Banks Peninsula, Canterbury, New Zealand, and their role in succession. New Zealand Journal of Botany 27:421-436. Pickart, A. J., L. M. Miller, and T. E. Duebendorfer. 1998. Yellow bush lupine invasion in Northern California coastal dunes: Ecological impacts and manual restoration techniques. Restoration Ecology 6:59. Prasad, R. 1999. Scotch broom, Cytisus scoparius L. in British Columbia, in. Pest Management Methods Network. Qian, P., and J. J. Schoenau. 1995. Assessing nitrogen mineralization from soil organic matter using anion exchange membranes. Fertilization Research 40:143-148. Richards, N. A. 1973. Oldfield vegetation as an inhibitor of tree vegetation, in R. Goodland, editor. Powerlines and the Environment. Cary Arboretum of New York, Mill Brook, New York. Richardson, D. M., P. Pysek, M. Rejmanek, M. G. Barbour, F. D. Panetta, and C. J. West. 2000. Naturalization and invasion of alien plants: concepts and definitions. Diversity and Distributions 6:93-107. Schimel, J. P., and J. Bennett. 2004. Nitrogen mineralization: challenges of a changing paradigm. Ecology 85:591-602. Stock, W. D., K. T. Wienland, and A. C. Baker. 1995. Impacts of invading N2-fixing Acacia species on patterns of nutrient cycling in two Cape ecosystems: evidence 81 from soil incubation studies and 15N natural abundance values. Oecologia 101:375-382. Ussery, J. G. 1997. Managing invasive plant species in Garry oak meadow vegetation communities: a case study of Scotch broom. Simon Fraser University, Burnaby, B.C. Vitousek, P. M., and L. R. Walker. 1989. Biological invasion by Myrica faya in Hawai'i: plant demography, nitrogen fixation, ecosystem effects. Ecological Monographs 59:247-265. Waterhouse, B. M. 1988. Broom (Cytisus scoparius) at Barrington Tops, New South Wales. Australian Geographical Studies 26:239-248. Weiss, S. B. 1999. Cows, cars and checkerspot butterflies. Conservation Biology 13:1476-1486. Wheeler, C. T., O. T. Helgerson, D. A. Perry, and J. C. Gordon. 1987. Nitrogen fixation and biomass accumulation in plant communities dominated by Cytisus scoparius L. in Oregon and Scotland. Journal of Applied Ecology 24:231-237. Whitehead, D. C. 1995. Grassland Nitrogen. CAB International, Wallingford. Williams, P. A. 1981. Aspects of the ecology of broom (Cytisus scoparius) in Canterbury, New Zealand. New Zealand Journal of Botany 19:31-43. Wink, M., L. Witte, T. Hartmann, C. Theuring, and V. Voltz. 1983. Accumulation of quinolizidine alkaloids in plants and cell suspension cultures: Genera Lupinus, Cytisus, Baptisia, Genista, Laburnum and Sophora. Planta Medica 48:253-257. 82 4. CONCLUSION In this thesis I have explored two aspects of Scotch broom (Cytisus scoparius) ecology. The first, Scotch broom's ecological role in the Garry oak ecosystem, investigates broom's effect on soil nutrient cycling and its correlation with native and exotic plant diversity. Within this study, the question of whether or not the success of an exotic forage grass, Dactylis glomerata, was directly facilitated by growing in association with broom was also examined. The second aspect of Scotch broom ecology which was explored was the susceptibility of broom seedlings to control, by way of fertilization using treated sewage biosolids in non- ecologically sensitive sites. Within this study, I explored possible mechanisms by which the treatments could possibly suppress the seedlings. Broom invasion in the GOE negligibly increased ammonium and significantly decreased phosphorus availability, while also being negatively correlated with native plant diversity. Dactylis glomerata, however, was not associated with broom presence in the field study, nor did it gain any size benefit from being grown with broom, yet both grass species used in the bioassay showed increasing trends in their flower production, indicating a possible increased reproductive output. Meanwhile, biosolids were shown to have positive controlling effects on broom seedling success at all three study sites, which is encouraging news for land managers with Scotch broom infestations. It also seems as though the mechanism by which the broom seedlings are controlled is through competition with other plants whose growth has been facilitated by the improvement in soil fertility. In conclusion, negative correlations were seen between broom invasion and native plant diversity in the federally-listed endangered Garry oak ecosystem. Although the evidence is not conclusive that broom has a causal effect on nutrient cycling, there seemed to be a link between slightly higher NFL/ levels at one of the two research sites (Rocky Point) as well as a decrease in phosphorus availability asociated with broom invasion (at the same site). The results of these studies underscore the responsibility of land owners with broom on their properties, to manage their lands appropriately. For those with ecologically sensitive 83 lands, such as the GOE, broom should be removed in an ecologically sensitive manner, such as described by Ussery (1997) while it behooves owners of disturbed, less ecologically sensitive sites to also take pro-active, responsible actions towards removing, and hence controlling the spread of, broom. The preliminary results here suggest that fertilization using treated sewage biosolids as a broom control method, has the potential to be a valuable tool in the ecologist's toolkit when used in the appropriate sites. 4.1 REFERENCES Ussery, J. G. 1997. Managing invasive plant species in Garry oak meadow vegetation communities: a case study of Scotch broom. Simon Fraser University, Burnaby, B.C. 84 APPENDICES APPENDIX A - Biosolids application rate Calculations: The calculations in steps 1 - 4 are to determine the amount of nitrogen that will be made available to plants (in kg/dry tonne biosolids) during the first year after biosolids application. The values are based on initial nitrogen levels determined through nutrient analysis of the biosolids (see Table 1-1 below) a predicted mineralization rate of 30% and a predicted loss to volatilization of 40% (as per the B.C. Biosolids Best Management Guidelines (McDougall et al. 2001)). The calculations in steps 5 and 6 are to determine the amount of wet biosolids needed to apply in order to achieve a total crop nitrogen rate of 400 kg N/ha. The calculations in step 7 are to ascertain the actual amount of nitrogen applied at the site based on the actual amount of biosolids that were available per plot that day. Stepl - Determine the amount of organic nitrogen which will mineralize in the first year, from total Kjeldahl nitrogen amount (kg/dry tonne biosolids). Eg. Mineralization rate from 2003 averages (used at Iona Beach): 44.0 kg/DT x 30% =13.2 kgN/D7/ biosolids Step2 - Amount of nitrogen that will be made available to plants after considering the amount lost to volatilization during the application process, obtained using the value for available ammonium (kg/DT biosolids), Eg. Initially available ammonium from 2003 averages (used at Iona Beach): 9.212 kg/DT x (1 - 40%) = 5.527 kg/DT biosolids Step3 - Initially available nitrate values obtained directly from biosolids analyses. Eg. Initially available nitrate from 2003 averages (used at Iona Beach): = 0.0084 kg N037DT biosolids 85 Step4 - Total nitrogen available in first year after biosolids application determined by summing the above three values. Total available N per dry tonne of biosolids = 13.2 kg + 5.527 kg + 0.008 kg = 18.74 kg N/DT biosolids Step5 - To determine the application rate of biosolids (DT/ha) necessary to provide a crop N requirement of 400kg N/ha, in the first year following application, based on the nitrogen availability (kg N/DT biosolids) obtained in steps 1-4. Application rate (DT/ha) = 400 kg N/ha - 18.74 kg N/DT biosolids = 21.34 DT/ha Step6 - To determine the wet biosolids application rate based on the % moisture and the dry biosolids application rate. Wet biosolids application rate (WT/ha) = 21.34 DT/ha 31.3% = 68.18 WT/ha = 6.82 Wkg/m2 Step7 - Back calculation to determine how much N was made available to plants based on actual total amount of biosolids applied to plots (64 Wkg/10.24 m2) Total wet weight per ha: 64 Wkg/10.24m2 x 10 000m2/ha = 62500 Wkg biosolids/ha Total dry weight per ha: 62500 Wkg/ha x 31.3% solid = 19562.5 Dkg biosolids/ha Total N applied per ha: 19.562 DT biosolids/ha x 18.74 kg N/DT biosolids = 366.59 kg N/ha 86 Results for biosolids nutrient analyses used to calculate biosolids application rate. Average Average Iona Burnaby Values 2003 Values 2004 Beach Mtn. Total Solids (%) 31.3 30.7 30.5 31.0 TKN (ppm dry wt.) 44.0 43.6 28.0' 33.9 NH/ (ppm dry wt.) 9000 9000 2000 936 NO3" (ppm dry wt.) 8.4 4.3 10 1.9 87 APPENDIX B - Native seed mixes used Iona Beach Burnaby Mountain Duncan Grasses Bromus sitchensis Bromus richardsonii Bromus richardsonii Calamagrostis nutkaensis Bromus sitchensis Bromus sitchensis Calamagrostis stricta Bromus vulgaris Bromus vulgaris Deschampsia cespitosa Calamagrostis nutkaensis Calamagrostis nutkaensis Deschampsia elongata Deschampsia cespitosa Deschampsia cespitosa Hordeum brachyantherum Deschampsia elongata Deschampsia elongata Leymus mollis Elymus glaucus Elymus glaucus Elymus trachycaulus Elymus trachycaulus Festuca rubra Festuca rubra Hordeum brachyantherum Hordeum brachyantherum Shrubs Cornus sericea Herbaceous Carex macrocephala plants 88 APPENDIX C - Trace element application rate Calculations: The following calculations are for predicting if biosolids application exceeds the acceptable agricultural limits as set out by the Organic Matter Recycling Regulation (OMRR). The trace element values are based on those that were determined from soil and biosolids analysis in 2003 and 2004 (see Table 1-2 below). Stepl. Total amount, in kg/ha, of each trace element added to site based on pre• determined application rate, in dry tonnes/ha, (see Appendix 1) and biosolids trace element concentration, in ppm. Eg. For Chromium at Iona Beach: \9.6DTIha 68 ppm x = 1.3 kg/ha 1000% IDT Step2. Total amount, in kg/ha, of trace element background soil levels based on soil bulk density and total volume of soil in one hectare of soil to a depth of 15 cm. 20.7 ppm x I500kg/m3 x 1500m3/ha = 46 575 000 mg/ha = 46.575 kg/ha Step3. Expected final amount of trace metal in soil obtained by summing values from steps 1 and 2. 1.3 kg/ha+ 46.575 kg/ha = 47.875 kg/ha Step4. Convert final amount back to a concentration (ppm) in order to compare to OMRR agricultural limits by dividing by a conversion factor set by: _ 1500m3/haxl500kg/m3 ~ 1000000 = 2.25 kg/ha per million 89 47.875% I ha 2.25kg I ha I million 21.28 ppm Values in parts per million (ppm) for trace elements in biosolids and in site soils which are used to calculate the total predicted soil concentrations following biosolids application at calculated rates. Elements Biosolids Background Site Predicted Soil Agricultural Averages Levels Concentration Limits (ppm) (ppm) (ppm) (ppm) *2003 2004 *Iona BBY Duncan *Iona BBY Duncan Arsenic 5.3 5.4 0 0 0 0.1 0.0 0.0 ~ 25 Cadmium 3 2.9 0 0 0 0.0 . 0.0 0.0 9 Chromium 82 68 20.7 11.7 26.0 21.3 12.3 26.6 50 Cobalt 6.6 8.1 8.3 5.3 12.7 8.4 5.4 12.8 40 Copper 1258 1115 11.7 13.7 37.0 27.7 23.4 46.7 150 Lead 71 89 5 11.3 8.3 5.9 12.1 9.1 350 Mercury 3.7 2.9 0.02 0.1 0.0 0.1 0.1 0.1 0.6 Molybdenum 14 14 0 0.0 0.0 0.2 0.1 0.1 5 Nickel 26 23 29.7 7.3 21.0 30.0 7.5 21.2 150 Selenium 5.9 5 0 0.5 0.0 0.1 0,5 0.0 2 Zinc 862 961 38 45.3 48.7 49.0 53.7 57.1 200 * Application rates at Iona Beach were based on 2003 averages from the Greater Vancouver Regional District's Annacis Island monthly biosolids monitoring program. 91 APPENDIX D - Ammonium nitrate application rate Calculations: To obtain 350 kg N/ha/yr in three installments for a plot of 12 m2: Step 1. Since ammonium nitrate releases 34% available nitrogen by weight, I must first determine how many kilograms of ammonium nitrate is required to obtain 350 kg available N/ha. 350 kg available N/ha/yr -f 0.34 available N/kg ammonium nitrate = 1029 kg ammonium nitrate/ha/yr Step 2. Amount of ammonium nitrate required for a 12 m2 plot. 1029 kg ammonium nitrate/ha/yr 10 000 m2/ha x 12 m2 = 1.24 kg ammonium nitrate/plot/yr Step 3. To obtain the rate for the first of three applications. 1.24 kg ammonium nitrate/plot/year -f- 3 installment/year = 0.41 kg ammonium nitrate/plot/installment APPENDIX E - Species lists Complete list of species found at three field sites. I = Introduced; N = Native; G = Grass; H = Herbaceous; S = Shrub. Iona Beach Burnaby Mountain Duncan I/N G/H/S Species I/N G/H/S Species I/N G/H/S Agrostis Agrostis Agrostis exarata N G capillaris 1 G capillaris 1 G Agrostis Agrostis Aira caryophyla 1 G stolonifera 1 G stolonifera 1 G Anthoxanthum Anthoxanthum Aira praecox 1 G ode rat um 1 G odoratum 1 G Bromus Deschampsia carinatus N G elongata N G Bellis perrenis 1 H Bromus Leucanthemum sitchensis 1 G vulgare 1 H Cerastium sp. 1 H Bromus Plantago tectorum N H lanceolata 1 H Cirsium arvense 1 H Cardamine Polygonum Convolvulus oligosperma N H persicaria 1 H arvensis 1 H Carex Taraxacum Cytisus macrocephala 1 H officianale 1 H scoparius 1 S Pteridium Deschampsia Cirsium arvense 1 S aquilinum N H elongata N G Cytisus Rumex scoparius N G occidentalis N H Elymus sp. unk. G Deschampsia Cytisus Gnaphalium elongata 1 H scoparius 1 S oligonosum 1 H Draba verna N H Rubus discolor 1 S Holcus lanatus 1 G Epilobium Rubus Hordeum brachycarpum N G parviflorus N S brachyantherum N G Hordeum Rubus brachyantherum N G spectabilis N S Hypericum sp. 1 H Lupinus Leymus mollis N S Rubus ursinus N S polyphyllus N H Mahonia Madia aquifolium N G glomerata N H Matricaria discoidea 1 H Myosotis discolor 1 H Navarretia squarrosa N H Pascopyrum smithii N G Poa pratensis 1 G Polygonum persicaria 1 H Pteridium aquilinum N H Ranunculus N H 93 Iona Beach Burnaby Mountain Duncan Species IVN G/H/S Species I/N G/H/S Species I/N G/H/S occidentalis Rubus ursinus N S Rumex acetosella 1 H Rumex occidentalis N H Spergularia rubra 1 H Taraxacum officinale 1 H Trifolium sp 1 H 94