Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Species richness, abundance and distribution in experimentally fragmented landscapes Perdue, Mark Edward 2002

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2002-0202.pdf [ 4.9MB ]
Metadata
JSON: 831-1.0090270.json
JSON-LD: 831-1.0090270-ld.json
RDF/XML (Pretty): 831-1.0090270-rdf.xml
RDF/JSON: 831-1.0090270-rdf.json
Turtle: 831-1.0090270-turtle.txt
N-Triples: 831-1.0090270-rdf-ntriples.txt
Original Record: 831-1.0090270-source.json
Full Text
831-1.0090270-fulltext.txt
Citation
831-1.0090270.ris

Full Text

SPECIES RICHNESS, ABUNDANCE AND DISTRIBUTION IN EXPERIMENTALLY FRAGMENTED LANDSCAPES By  MARK EDWARD PERDUE B.Sc, The University of British Columbia, 1996  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF T H E REQUIREMENTS FOR T H E DEGREE OF MASTER OF SCIENCE In  THE FACULTY OF GRADUATE STUDIES Faculty of Forestry ( Department of Forest Sciences ) We accept this thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH COLUMBIA November, 2001 © Mark Edward Perdue, 2001  In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives.  It is understood that copying or  publication of this thesis for financial gain shall not be allowed without my written permission.  Department of The University of British Columbia Vancouver, Canada  DE-6 (2/88)  ABSTRACT Experimentally fragmented landscapes were created in prairiefieldsat the Konza Long Term Ecological Research Area in the Flint Hills of Kansas State. Landscapes  (2m X 2m) were  fragmented by isolating plots (0. Im X 0. Im) of open-habitat (bare ground) using Bromus inermis,  an aggressive grass species, as a barrier. Elements of fragmentation, habitat area,  dispersion, and isolation, were varied by altering the numbers of randomly located brome plots. From 1995 through 1998, species were recorded for each open-habitat plot and brome-occupied plot. The effects of fragmentation were considered in terms of species richness, abundance and spatial distribution, each correlated with the level of fragmentation. Fragmented landscapes had fewer species than contiguous landscapes. At intermediate levels of open-habitat, some species occupied alternate plot cover types more frequently, somewhat mitigating the effects of fragmentation. However, continued fragmentation resulted in the extirpation of most species. Fragmentation also reduced the abundance of satellite and core species. Core species abundance were dramatically reduced near 60% open-habitat, consistent with the changes in species richness. These changes coincided with the percolating threshold, suggesting that habitat dispersion did contribute to the effects of habitat loss. Changes in the spatial distribution of species, a fundamental expression of how species are affected by fragmentation, were characterized by spatial signature analysis. Changes in species' spatial signatures were compared to changes in available habitat. Spatial signature analysis was able to detect how each species responded to fragmentation spatially. Based on this analysis, successful species could be distinguished from those species imperiled by these experimentally fragmented landscapes.  ii  TABLE OF CONTENTS.  Page ABSTRACT  ii  TABLE OF CONTENTS  iii  LIST OF TABLES  v  LIST OF FIGURES  vi  CHAPTER I.  Introduction  1  CHAPTER II.  Experimentally Fragmented Prairie Landscapes  4  Introduction  4  Methods Study Site Replicating a Random Neutral Landscape Model Field Methods  6 6 8 11  Results Habitat Area Habitat Arrangement Sampling Seasons Experimental Design  11 11 15 21 24  Conclusion CHAPTER III.  28 29  Species Richness  Introduction  29  Methods Sample Period Selection Species Richness  30 30 31  Results Open-habitat Species Richness Sparse-brome Species Richness Sparse-brome Species Richness Dense-brome Species Richness Overall Species Richness Conclusion  31 31 33 34 35 37 39  iii  CHAPTER IV.  Species Abundance and Distribution  Introduction  41 41  Methods Species Abundance Spatial Distribution of Species Results Species Abundance Resilience to Fragmentation and Biological Guilds Brome Tolerance, Biology Guilds and Resilience to Fragmentation Spatial Distribution of Species Spatial Signature and Resilience to Fragmentation Spatial Signature and Brome Tolerance Spatial Signature and Biological Guilds Conclusion CHAPTER V. Conclusion  43 43 46 52 52 56 57 59 61 61 62 63 65  LITERATURE CITED APPENDIX I. Sample of Field Data Card APPENDIX II. Species List with Observations  69 74 75  APPENDIX III.  Seasonal Occurrence of 1996 Species  81  APPENDIX IV.  Summary Table of Regression line Models  82  APPENDIX V.  Sample Species Selection  103  APPENDIX VI.  Species Selection Criteria  106  APPENDIX VII.  K-S Test for Open-habitat Area  108  APPENDIX VIII. Classification of Spatial Signatures  109  APPENDIX IX.  Comparison of Biological Traits  114  APPENDIX X.  Spatial Signature of Species  118  APPENDIX XI.  Spatial Signature Classification of Sample Species  129  APPENDIX XII.  Spatial Signature Classification and Biological Guilds  135  APPENDIX XIII. Landscape Maps of Species Locations  136  APPENDIX XIV. Summary of Resilience  137  iv  LIST OF TABLES Page Table 1. List of sample species selected for the analysis of spatial occupancy.  44  Table 2. A summary of species biological guilds, number of attribute categories and basis for their categorization.  45  Table 3. Ranking of resilience and tolerance partitioned by biological guilds for 1996 sample species.  58  v  LIST OF FIGURES Page Figure 1. Map of Kansas and the Konza LTER.  8  Figure 2. Experimental layout (not to scale) of the three replicates (A, B, C) and 21 treatment levels (landscapes). Each landscape is 2m X 2m.  10  Figure 3. The design percent and actual percent open-habitat for each experimental landscape for spring and fall from 1995 through 1998.  13  Figure 4. Actual (transformed) coverage for each design percent for open-habitat including spring and fall samples distinguished by year.  14  Figure 5. Patch adjacency rules, nearest neighbour and diagonal neighbour.  15  Figure 6. Adjacent neighbour patch metrics of six landscapes observed in 1996 (spring).  16  Figure 7. Changes in the abundance and spatial distribution of brome from spring, 1995 through the fall, 1998 on landscape (100A). Figure 8. The patchiness (spatial signature) of open-habitat. Figure 9. Regression line of actual and design percent open-habitat for all spring sample periods (Y=predicted actual open-habitat, X=design open-habitat percent). Figure 10. Regression line of actual and design percent open-habitat available for all fall sample periods (Y=predicted actual open-habitat, X=design open-habitat percent).  18 20  22  23  Figure 11. The number of 5% interval open-habitat classes represented during the eight sample periods of the 4-year study.  25  Figure 12. Samples within each 10% open-habitat class for all years sampled.  27  vi  Figure 13. Open-habitat species richness plotted against the transformed percent available open-habitat and the percentage of plots classified as open-habitat for spring (A, C) and fall (B, D) samples.  33  Figure 14. Species richness in sparse-brome for spring (A, C) and fall (B, D) samples.  35  Figure 15. Species richness in dense-brome plots graphed against open-habitat area (%), dense-brome area (%) and the number of dense-brome plots (%).  37  Figure 16. Two regression lines showing the total number of species (overall species richness) in relation to the transformed percent available open-habitat for spring (A) and fall (B) data.  39  Figure 17. Distmguisliing the effects of successional status from the varying proportions of open-habitat area for Ambrosia psilostachya (Species 31).  47  Figure 18. Stylized comparison of spatial signature classifications. For clarity, symbols are simplified.  50  Figure 19. Stylized comparison of spatial signatures within the broader Established, Fragmented class.  51  Figure 20. Species abundance and richness for three classes of open-habitat area (%), 85% Open-habitat, 65% Open-habitat and 45% Open-habitat.  53  Figure 21. Maximum species abundance for each open-habitat area (%) class including all 1996 spring and fall data.  55  Figure 22. Spatial signature classification of sample species.  59  vii  CHAPTER I INTRODUCTION  Fragmentation describes changes in the spatial distribution of habitat. Implied are three related processes: habitat loss, dispersion, and isolation (Fahrig 1997, Bunnell 1999). Collectively, these phenomenon have been implicated in the mass extinction of species (Wilcox 1980), and are further described as the greatest threat to biological diversity worldwide (Harrison and Quinn 1989, Ehrlich 1992, Daily et al. 1993, Dale et al. 1994, Turner and R. C. Corlett 1996). Forests contain the majority of species on earth (Bunnell and Johnson 1998), and there is growing evidence that biological diversity is a positive system where forest inhabitants contribute to the ecological processes that are responsible for biological diversification, forest productivity and ecological integrity (Power et al. 1996, Tilman 1996, Tilman et al. 1996, Palmer and Maurer 1997, Symstad et al. 1998). Forest landscapes have been altered by timber harvesting activities which often occur across large areas (Wallin et al. 1996, Cissel et al. 1999, Hessburg et al. 1999a, Hessburg et al. 1999b, Hessburg et al. 1999c). Although research into the effects of fragmentation remain a central theme in conservation biology (Robinson et al. 1992, Maurer and Heywood 1993, Fahrig 1  and Merriam 1994, Rudis 1995, Wiens 1995, Fahrig 1997, With and King 1999b), predictions about the affects of habitat fragmentation differs; perhaps because of ecological complexities, experimental limitation, conceptual differences, or differences in scale. The tallgrass prairies are diverse and dynamic (Great Plains Flora Association 1986), characteristics that are ideal for exploring the affects of habitat fragmentation. The heterogeneity of prairie ecosystems can be captured within a relatively small spatial area, and their dynamic response to disturbance allows treatments to be monitored easily. This thesis explores the relationships between habitat area and species richness, abundance and spatial distributions in experimentally fragmented prairie landscapes. Specifically, five questions are posed.  1.  Does habitat loss affect grassland species richness?  2.  Do changes in species richness reflect changes in species abundance?  3.  Do changes in abundance reflect changes in the spatial arrangement  (distribution)  of species? 4.  Does the distribution of habitat affect the distribution of species?  5.  Is the spatial relationship between species and habitat associated with particular biological guilds?  This thesis contains five chapters. Chapter II (Experimentally Fragmented Prairie Landscapes) describes the study area, experimental design and a description of the experimental landscapes. These descriptions serve as benchmarks for subsequent landscape and species analyses. Chapter III (Species Richness) investigates speciesrichnessand composition in the experimental landscapes. Chapter IV (Species Abundance and Spatial Occupancy) relates these community metrics to measures of species' abundance and spatial distributions. The spatial  2  distribution of species are classified and compared according to habitat area and biological guilds. Chapter V (Conclusion) provides a summary, as well as opportunities for future work.  3  CHAPTER II EXPERIMENTALLY FRAGMENTED PRAIRIE LANDSCAPES  INTRODUCTION Percolation theory (Stauffer and Aharony 1994), as it relates to ecology, describes the relationship between landscape spatial structure and species movement through that structure. Based on geometry, the percolation model predicts the critical proportion of habitat that enables cross-landscape movement of species (O'Neill et al. 1991, Milne et al. 1992, Turner et al. 1993, Pearson et al. 1996). This threshold is scale independent, although specific to how an organism moves. Using percolation theory, neutral landscapes (NL) can be simulated. Neutral landscapes are grid-based, computer generated landscapes where habitat patterns are created independent of the biological and physical processes inherent in real landscapes (With 1997). While early NLs were relatively simple, increased computing capacity has permitted the development of more complex models that continue to contribute in studies of ecology. One such contribution has been the identification of the critical thresholds in habitat loss that define habitat fragmentation (With 1997).  4  Neutral landscape models can predict landscape characteristics from the proportion of habitat. In random NLs there is a geometric relationship between the area and arrangement of habitat. From this relationship, other landscape attributes such as habitat distribution can also be predicted (With and King 1997). The specific nature of this relationship depends on the distribution and shape of habitat (for example, randomly arranged versus aggregated habitat) (Turner 1989), and species perception (i.e. gap crossing capacity) (With et al. 1999, With and King 1999a). This relationship is not linear because the probability of isolating habitat varies, depending on the existing amount and distribution of habitat. Where the proportion of existing habitat is high, the probability that a given amount of habitat loss will create isolated patches is low; habitat loss creates perforations in an otherwise contiguous landscape. Continued habitat loss increases the chance of isolating habitat patches as critical habitat connections are severed. At low proportions of available habitat, continued loss removes patches of already isolated habitat. The percolating  threshold  is defined as the minimum proportion of habitat that permits  cross-landscape organism movement. In a percolating landscape, there are relatively few, large and connected patches. There are a growing number offieldtests supporting the observations generated within these models (Gustafson and Parker 1992, With and Crist 1996, Wiens et al. 1997, With 1997, Mclntyre and Wiens 1999, With et al. 1999). The results generated by NLs and small-scale field experiments represent a source of ecological insight not easily obtained in the study of large natural landscapes.  5  The objective of this chapter is to characterize the spatial structure of the experimental landscapes. The area of open-habitat will be plotted against the number of open-habitat patches for each landscape, and the result (the spatial signature) will be compared to the signatures reported for NLs. The relationship between the experimental and modeled signatures will determine the capacity to relate elements of habitat fragmentation to species richness, abundance and spatial distribution, which will be investigated in subsequent chapters.  METHODS  Study Site The Great Plains are the second largest biome in North America and encompass nearly 2.6 million square kilometers of the continental interior, roughly 14% of the continental land mass (Simms 1988). The Great Plains extent east from the Rocky Mountains to the deciduous forests of the Canadian Shield, and the Central Lowlands of the United States (Weaver and Albertson 1956). The Great Plains are plant species diverse, although it is species interactions more than uniqueness that is responsible for the ecological communities of the Great Plains (Great Plains Flora Association 1986). Within these communities, evolving species interactions and distributions combine to make grasslands dynamic. These dynamics are further embedded in extreme environments: a continental climate, intervals offireand drought, soil fertility, precipitation and grazing. The results are landscapes that are spatially and temporally diverse and seasonally dynamic.  6  The value of studying grassland ecosystems extends beyond their experimental expediency or their geographic representation. Agriculture, grazing, mineral exploration, urbanization, tree encroachment, and the invasion of exotics have fragmented most native grasslands. These changes exceed those reported for any other major ecological community in North America (Bragg et al. 1996) and conditions are particularly acute in the tallgrass prairie where habitat loss is estimated to be between 80% to 100%.  Tallgrass Prairie Vegetation of the Great Plains has been subdivided into numerous grassland communities (Great Plains Flora Association 1977). The Great Plains Flora Association (1986) provides a general characterization including three zones: the short grass region of the west, the mixed grasses of the central region and the tallgrass region of the east (Great Plains Flora Association 1986). Tallgrass prairie is dominated by Andropogon gerardii (big bluestem), Schizachyrium scoparium (little bluestem), Panicum virgatum (switch grass), Sorghastrum nutans (indian grass), and Bouteloua curtipendula (sideoats grama) (Great Plains Flora Association 1977).  Konza Prairie Biological Station Konza is one of the National Science Foundation's Long Term Ecological Research areas (LTER) and has been the site of comprehensive ecological studies since 1972. The area remains mostly unaltered and is characteristic of the native tallgrass prairie region of the Great Plains. Konza currently encompasses 3,858ha in the Flint Hills of Kansas (Figure 1).  7  KONZALTER  Kansas  Scale (kms)  Figure  1. Map of Kansas  100  and the Konza  200  LTER.  Replicating a Random Neutral Landscape Model From 1995 through 1998, experimental landscapes were located within a Bromus  inermis  (brome) hayfield at Konza. The area of each experimental landscape was 2m X 2m, a size reported to be large enough to encompass a full range of tallgrass prairie plant associations (Barfhaefa/. 1995). Landscapes were subdivided into 400 (0.1m X 0.1m) plots. Plots of openhabitat were created by clearing all existing plants and spreading sterilized potting soil. Plots of brome were considered non-habitat (matrix).  Brome is an introduced species that is among the  most successful at colonizing and competing in native plant communities of the Great Plains (Mack and Contreras 1995, Blankespoor and May 1996), and is prevalent throughout the study area. Treatment levels were represented by the 21 classes of open-habitat  (i.e. 0%,  5%.. .100%) created by denuding the appropriate number of plots. For example, a treatment level of 25% open-habitat required 100 plots (0.25 X 400) of open-habitat. To keep the  8  proportions of open-habitat and matrix consistent throughout the study period, brome was regularly removed (weeded) from open-habitat plots. Random landscape patterns were created by randomly designating the plots to be denuded. The treatment levels (landscape habitat)  level of open-  were used to characterize various levels of fragmentation. This experimental fragmentation differs from the common notion of fragmentation,  where undisturbed areas define the habitat type. Thus, this research concentrates on species dispersal within the experimental landscapes. Inferences made here must be tempered by the recognition that the primary source of open-habitat species is the brome-dominated communities within, and beyond, these landscapes. Experimental Design  For each of the three replicates (A, B, C), landscapes were arranged in a systematic progression of the 21 levels of open-habitat ( (Figure 2). All landscapes were spaced 2m apart, and the total study area including all landscapes and interstitial areas, was 10m X 82m. The interstitial areas separating treatment levels and replicates were not maintained (i.e. they were not weeded or mowed).  9  Treatment Level 100 % Open-habitat  0% Open-habitat  *  82m  *  A  B  Figure  •  •  •  •  •  2. Experimental  levels (landscapes). shown in the  •  •  •  •  •  •  •  •  •  •  •  •  layout (not to scale) of the three replicates  Each landscape  is 2m X 2m. A single landscape  enlargement.  10  •  •  •  •  10m  (A, B, C) and 21 with habitat  treatment  (white) is  Field Methods When sampling, plots were delineated by positioning a grid. Within each plot, all plant species (or the nearest phylogenic level possible) were recorded. Plants were identified and named according to Flora of the Great Plains (1986) and species that could not be identified were coded as unknown.  Field notes were recorded on cards replicating the plot layout of landscapes  (a samplefieldcard is included in Appendix I). Sample Periods  Field sampling was completed from 1995 to 1998. To accommodate seasonal variations, whenever possible, landscapes were sampled biannually (spring and fall). The exceptions to this sampling routine were replicates B and C, which were not sampled in the fall of 1995. Spring samples were completed in June of each year while fall samples were completed in October, except for 1998 when the fall sample was completed in September.  RESULTS  Habitat Area The objective of this chapter is to characterize the spatial arrangement of open-habitat within the experimental landscapes. Over the course of this study, some plots of brome could not be maintained as a dense barrier, and many plots cleared to create open-habitat were often colonized by brome (Figure 3). These variations altered the random distribution of open-habitat within the experimental landscapes. At almost all levels of fragmentation, the number of Open-  11  habitat plots fell short of the design specifications, as illustrated in Figure 3. This trend is evident within each sampling period except in spring-1997 where data approaches the experimental design because sampling occurred soon after a weeding of brome from openhabitat plots (Figure 3).  Habitat Class Transformation Brome coverage was not consistent, some plots were densely covered while other plots were sparsely covered. To characterize these differences, observations of brome are further classified as having greater than 50% coverage, or less than 50% coverage, and are referred to as dense-brome or sparse-brome, respectively. For estimates of landscape open-habitat area, each plot is assigned the midpoint value of brome coverage. Specifically, open-habitat plots are assigned 100% open-habitat, sparse-brome plots are assigned 75% open-habitat and dense-brome plots are assigned 25% open-habitat. These transformed values of percent open-habitat are presented in Figure 4.  12  " 1995 100 T  • 1996 * 1997  x 1998  90 +  80 70 +  60 +  50  40  3  x 30  ft  20  10  X •  d A  M 10  8 %  o X  X  0 *  o X  X X  M 20  8 30  x  X X X  X  x X  X 40  50  60  70  80  90  100  Design Open-habitat (% bare)  Figure 3. The design percent and actual percent open-habitat for each experimental landscape for spring andfall from 1995 through 1998.  13  100 -i  I B  90 A  s  X  70  A  X  ¥  X  X  >x »  A  80  x  •i A  X  A  i A  X  X  X  X  X X  X  x  50  X  o o  8 §  X  •  x  0  •  Q  40  • a  °•  o  a  ^  X  «  X  x X X  • a  X  X  °  X  X  60  B  § 0  •  •  1995  30  1996 20  1997 1998  10 0 0  1  1  1  10  20  30  1  40  1  50  1  1  60  70  1  80  1  90  1  100  Design Open-habitat (% bare) Figure 4. Actual (transformed) coverage for each design percent for open-habitat including spring andfall samples distinguished by year. Plots with observations of dense-brome were assigned a open-habitat area of 0.25. Sparse-brome plots were assigned a value of 0.75.  14  Habitat Arrangement  Habitat Patches The shape, size, quality and configuration of habitat provides a template for species life history, such as dispersal, productivity and competitive capacity. Collectively, these elements define, in varying degrees, the abundance and spatial distribution of species. Here the spatial configuration of open-habitat is characterized by patchiness:  the relationship between the  number of patches and the total area within those patches. Patches are defined as adjacent plots of similar attributes, such as cover type or species observations. Patch metrics are calculated for each plot cover type and sample species. Adjacency is modeled according to the neighbour  rule  diagonal  (Stauffer and Aharony 1994) including the four adjacent plots, plus the nearest  four diagonal plots (Figure 5). Figure 6 shows examples of the distribution of open-habitat, sparse-brome and dense-brome coverage, and the transformed available open-habitat values.  • • BBSS! Nearest Neighbour Figure  5. Patch adjacency  Diagonal Neighbour  rules, nearest neighbour  15  and diagonal  neighbour.  Open-habitat  Patch Type  No. of Patches  No. of Plots  dense-brome sparse-brome open-habitat  26 0  312 88 0  45%  dense-brome sparse-brome open-habitat  2 11 0  248 152 0  55%  dense-brome sparse-brome open-habitat  16 2 6  169 224 7  65%  dense-brome sparse-brome open-habitat  25 1 13  105 270 25  75%  dense-brome sparse-brome open-habitat  23 3 16  83 213 104  85%  dense-brome sparse-brome open-habitat  3 2 13  3 199 198  35%  1  Figure 6. Adjacent neighbour patch metrics of six landscapes observed in 1996 (spring). Dark squares indicate dense-brome, light gray represents sparse-brome and white plots identify openhabitat (bare ground) for other species. 16  Habitat Patterns  Within landscapes, the random distribution of open-habitat was expected to remain constant. However, the clonal expansion of brome caused non-random open-habitat patterns. The non-random arrangement of open-habitat alters the relationship between habitat area and landscape characteristics, such as habitat dispersion and isolation. For example, landscapes with a clumped arrangement of habitat are more connected than patchy (dispersed arrangement of habitat) landscapes. Specifically, in patchy landscapes the percolating threshold is 54%, compared to a threshold of 45% in aggregated landscapes (With and King 1999a). Figure 7 presents eight maps of landscape 100A illustrating how brome coverage changed over the duration of the study. This effect was evident, to varying degrees, on most of the 63 landscapes. Non-random patterns became more apparent in each successive sample period and were most evident in landscapes with initially high levels of open-habitat availability. For example, when established, landscape 100A was expected to provide contiguous open-habitat. Figure 7 (Spring 1995), shows that this expectation was met. However, the seven maps depicting this landscape in successive years show the progressive invasion of brome. Further, the clonal expansion of brome was non-random, begiiining along the edges of the interstitial areas and eventually encompassing most of the landscape.  17  Figure 7. Changes in the abundance and spatial distribution of brome from spring, 1995 through the fall, 1998 on landscape (100A). The dark gray plots identify dense-brome plots, light gray plots represent sparse-brome plots and white represents open-habitat. Spring 1997 had no observations of sparse, or dense, brome.  18  To evaluate if these experimental landscapes approximated neutral landscapes, the number and arrangement of open-habitat plots (open-habitat patchiness) within these landscapes were compared to published values of patchiness generated with NL computer models. The collective shape of patchiness values is referred to as the spatial signature. The spatial signature of open-habitat, including all seasons and sample periods, is shown in Figure 8. Based on the open-habitat spatial signature for these experimental landscapes (Figure 8), fragmentation occurs below approximately 65% open-habitat plots. Above 65%, the landscapes become largely contiguous (few, very large patches). This interpretation is consistent with published results where connectivity is reported to occur between 0.6 in random for nearest neighbour patches, and 0.3 for aggregated, next-nearest neighbour patches (With and King 1999a). Overall, the arrangement of open-habitat plots is considered are reasonable representation of a random neutral landscape model.  19  50  -i  40 XL  BOcd  30  O O O O O  o  P-  <u  XL  OOOO  20  0900  O  •OOO oo O O O O OOOOO o o OOOO oo o  10  • O OOO 000 00 00 OOO  O  oo  «oo m o  o  o  OOOO 00  o  o o  OOO  ooooooo oo oo  £  2  oo  O  oo oo o o o O O 030 0 OO 0 0 o  c I—  OO O CO o  o o OOO  O  O OO  o  O O OOO  o  00  00  00  oo o  O O O  o  O  o  o  o  O O 00 O O O O  O O O  o  o  o o O OOOO o oo o o o o 00  O OOO  o  0 0  10  20  30  o o  o  40  50  o  o oo o  <3S0  OO  o  60  O  oo  70  o o  OO O OOOOO T  80  90  100  Open-habitat Plots (%) Figure 8. The patchiness (spatial signature) of open-habitat. Number of patches is plotted against plots (percent of landscape) for that cover type. This graph includes all landscapes all sample periods. The generalizedform of the spatial signature is the shaded area.  20  Sampling Seasons Landscape pattern, species presence, and composition, were predicted to vary seasonally. A species-specific analysis of the 1996 data indicates a seasonal effect. Of the 53 species (including Dense-brome and Sparse-brome) observed in the spring and fall, nine species were observed exclusively or extensively in spring, ten species occurred primarily in the fall, and 31 species occurred similarly in both sampling seasons. Three species were not classified because their identifications were uncertain. The number of species observations for each sample period is presented in Appendix II, and further summarized by seasonal occurrence in Appendix III. Seasonal variation was also evaluated using regression line analysis. Regression line lines for spring and fall samples are shown in Figure 9 and Figure 10, respectively. The results of these tests (Appendix IV) confirm that the landscape level of open-habitat varies significantly between spring and fall samples (p = 0.000). Interaction between the sampling season and percent open-habitat is significant at alpha 0.05. Spring samples more closely resemble the experimental design throughout most levels of coverage. Figure 9 and Figure 10 present the expected and actual percent open-habitat for spring and fall samples, respectively. Based on these analyses, spring and fall samples are treated separately throughout the remainder of this study.  21  •a i  a cu O -4—>  o  (Spring)  0  = 43.4 + 0.423X  (R =0.579)  1  1  1  1  1  1  i  i  i  i  10  20  30  40  50  60  70  80  90  100  Design Open-habitat (% Figure periods  9. The regression (Y=predicted  line of actual and design percent  actual open-habitat,  X=design  22  open-habitat  open-habitat  for all spring  percent).  sample  Figure 10. The regression line of actual and design percent open-habitat available for all fall sample periods (Y=predicted actual open-habitat, X=design open-habitat percent). Regression lines of the two sample seasons were statistically different.  23  Experimental Design  Treatment Replication To minimize potentially confounding effects (e.g. edaphic variations) associated with slope position, replicates were oriented parallel to slope contours. While the parallel arrangement of replicates may have reduced spurious effects between treatment levels, the arrangement may have also increased such interactions between replicates. Regression analysis was used to test for differences in open-habitat between the three treatments. Based on the values of open-habitat, the influence of slope position was insignificant (Appendix IV - B and Appendix IV - C). Accordingly, replicates are pooled in subsequent analyses. Actual Brome Coverage and Treatment Levels The experimental design included 21 treatment levels of open-habitat (Figure 2). However, the expansion of brome changed the representation of each treatment level. Figure 11 illustrates the number of open-habitat levels represented for each sample period of the study. Extreme levels of open-habitat (i.e. 100%, 95%, 90%, 5%, 0%) were least represented, and intermediate levels of open-habitat were best represented. Representation of treatment levels is important because the relationship between the landscape level of open-habitat and the species richness, abundance and distribution are not uniform. Representation of intermediate levels is particularly important because these values tend to represent percolating thresholds. Specifically, in random landscapes the percolating threshold occurs at 59% available open-habitat, assuming square plots and patch delineation according to Nearest Neighbour rule (With and King 1999a).  24  16 i  1995  1995  1996  1996  1997  1997  1998  1998  Sample Period  Figure periods  11. The number of 5% interval of the 4-year  open-habitat  study.  25  classes represented  during  the eight  sample  To better reflect the coarse method of depicting actual brome plot coverage (i.e. 0% open-habitat, 25% open-habitat and 75% open habitat) and improve representation of treatment levels, 10% open-habitat classes are adopted. The results of this classification scheme are illustrated in Figure 12. Although there is only a modest improvement in the number of openhabitat classes represented, the scheme does strengthen the representation within the various open-habitat classes.  26  160  • Spring LlFall 140  120 r/5  3" o  s I—I  100  80  o s-  S  60  40  20  0 0.0  1  I  i  T  T  T  T  1  1  9.5 19.5 29.5 39.5 49.5 59.5 69.5 79.5  T  89.5  100.0  Open-habitat Class Boundaries (%  Figure 12. Samples within each 10% open-habitat class for all years sampled. Spring and f samples are distinguished. This reclassification results in improved representation for most open-habitat classes. Exceptions are the terminal class (95%) for the fall samples and the class for both spring andfall samples. 27  CONCLUSION The expansion of brome altered the number and arrangement of open-habitat plots within landscapes, and the number and arrangement of landscapes representing each treatment class, as well as the arrangement of those landscapes. The number of plots with brome exceeded design specifications, yet coverage within plots was inconsistent. To accurately reflect the landscape level of open-habitat, a coarse classification scheme of open-habitat cover classes was applied. Landscape open-habitat was represented by the sum of the transformed plot open-habitat area. The clonal spread of brome also produced a clumped arrangement of open-habitat. This clumping was evident among plots and landscapes. Throughout the study period, clumping increased as brome continued to expand. Despite changes in the number and arrangement of plots with brome, the relationship between the number of patches and open-habitat areas (Figure 8) coincides with randomly arranged patches predicted by neutral landscapes. Another consequence of the expansion of brome coverage was the reduced representation of open-habitat coverage. The extreme treatment levels were poorly represented, although intermediate levels were well represented. Based on the open-habitat patch characteristics presented in Figure 8, the percolating threshold occurs below about 65% open-habitat, above this level open-habitat is extensively contiguous. Thus, the effects of fragmentation should be evident within the treatment levels represented in this study.  28  CHAPTER III SPECIES RICHNESS  INTRODUCTION Speciesrichness,a commonly used measure of diversity (Hunter 1996), is the number of different species within a region. The relationship between habitat area and species richness, often referred to as the species-area model, was introduced by Preston (1962). The species-area model was based on the observation that within a region, most species are relatively uncommon and few species are common. Consequently, as habitat area increases so does the number of species. The exact form of this relationship varies, nevertheless the implication is that for any particular area there is a predictable number of species. The theory of island biogeography builds on the species-area relationship and incorporates the influences of species extinction and habitat isolation; as habitat is isolated, immigration declines causing a decline in the number of species (MacArthur and Wilson 1967). Population size depends on area, and as area decreases, extinction rates increase. Although the model was derived from observations of oceanic islands, Preston (1962) and subsequently MacArthurand Wilson (1967), suggested that the theory was similarly applicable to natural  29  continental islands (Diamond 1975). The analogy between habitat fragmentation and natural continental islands is the theoretical foundation for the ominous predictions about mass extinction and global declines in biological diversity. Yet, this analogy may not be completely accurate. Landscapes are dynamic mosaics of heterogeneous types, and the elements that create this heterogeneity range from subtle (for example, the overstory age of a forest) to dramatic (such as fields, rock outcrops, water, etc). Often, biophysical similarities exist and differences are diminished through successions in plant communities (Stouffer and Bierregaard 1993). Landscape complexity is maintained by ongoing disturbances (Pickett and White 1985). This complexity obscures the relationship between species richness and area. The purpose of this chapter is to examine how the loss of open-habitat (open-habitat fragmentation) affects plant species richness in these experimental landscapes. The three cover types (open-habitat, sparse-brome and dense-brome) are regarded as a simplistic model of landscape heterogeneity. Species richness is reported in two ways;first,against each landscape level of open-habitat area (%) and second, against the number of plots within the three cover types.  METHODS  Sample Period Selection To simplify analyses required to report changes in species richness, abundance and spatial distribution throughout the entire study period, a single sample season is selected. The 1996 samples are chosen for two reasons. First, data for both the spring and fall seasons are  30  available for 1996 and second, the 1996 data provide the broadest range of open-habitat area values.  Species Richness For each cover type, plant species richness is to be plotted against the transformed landscape level of open-habitat to provide a consistent landscape metric. Species richness is also plotted against the untransformed proportion of plots within each cover type. Species richness within dense-brome plots is also plotted against the transformed proportion of dense-brome to illustrate the relationship between species richness and dense-brome area. The expectation was that for each cover type there would be a unique relationship between species richness and open-habitat area, as well as the number of cover type plots. Each relationship was expected to generally resemble the species-area model and subtle differences in these relationships would reflect the relative quality of these cover types to the species present. Specifically, open-habitat plots were expected to support the greatest number of species. Sparsebrome plots and dense-brome plots were expected to be progressively less diverse, reflecting the imposition of brome. Overall species richness was expected to provide the most comprehensive indicator of the effects of fragmentation.  RESULTS  Open-habitat Species Richness Open-habitat species richness is shown in Figure 13. Regression analysis (Appendix IV - F through Appendix IV -1) of these positive linear relationships reports coefficient of  31  determination (R ) values of 0.79^^) and 0.62(aii) for Figure 13A and Figure 13B, respectively. 2  Analysis of variance reports that the spring and fall regression lines equations were statistically significant, both at p = 0.000. The correlation between species richness and open-habitat is a positive logarithmic relationship (R (sprin ) 0.90, R (feii) = 0.93), in Figure 13C and Figure 13D, respectively. The 2  =  2  g  correlation between open-habitat species richness and open-habitat plots (%), for spring and fall data, were statistically significant at p = 0.000.  32  B  25  25 Y FaU)=-15.7 + 0.357A  y spm«)=-14.9 + 0.345J!f (tf=0.790) <D  20  20  15  15  10  10  '3 <U OH  GO  0.620)  r  (  (  o  l-l  <L> -O  s  0  10  20  30  40  50  60  70  80  90  11  100  Open-habitat Area (% 0 10  20  30  40  50  60  70  80  r~  90 100  D 25-1 25 y (3^=0.502+l-TaxgV+OaKJf (#=0.90)  <o  y  2CH  (FaU)=  0.314 + 2.430LogA- + 0.254JT (^ = 0.930)  20  '5  CU O.  GO  15  15  •8 l-i  <D  6  10  0  10 1  "i 0  1  0  1  n  11111 1  10 2 0 3 0 4 0 5 0 6 0 7 0 8 Plots 0 9 0 (%) 103  ~i  1  1-  0  10  20  30  40  50  60  70  80  90 100  1  in Open-habitat  Figure 13. Open-habitat species richness plotted against the transformed percent available open-habitat and the percentage ofplots classified as open-habitat for spring (A, C) andfall (B, D) samples.  33  Sparse-brome Species Richness Species richness in sparse-brome is also positively correlated with open-habitat area (R (spring)= 2  0.67, R (6ii) = 0.62) and sparse-brome patch (R ( pring) 0.30, R (ftii) = 0.35) (Appendix 2  2  =  2  S  IV - J through Appendix IV - M). These regression lines (p = 0.000) are statistically significant at p = 0.05. At 85% open-habitat area, sparse-brome species richness (Figure 14) is similar to open-habitat species richness (Figure 13). Although, sparse-brome species richness is less sensitive (i.e. slope is more gradual, raring) 0.21 and r^n) = 0.27) than open-habitat species =  richness (raring) = 0.345 and r(feii) = 0.357) to landscape level of open-habitat.  34  B 25  25  Y ^ = - 3 . 2 5 + 0.213^ (^=0.67)  Y  20  201  a.  15  15  <U  10  101  (M)  =-3.71+0.26fi\r  (#=0.62)  <U  'O  <U  in o  s  I  5 1 — 0 0 10 20 30 40 50 60 70 80 90 100 _j  j  —1  1  1  1  i  1  1  1  i  1  1  0 10 20 30 40 50 60 70 80 90 100  Ooen-habitat Area (%) D 25  251  Y (M5=-14.1+6.44Logr (#=0.35)  Y ^ = - 1 3 . 3 + 5 . 7 4 ^ (#=0.30)  20  20 CO  CD  '3  15  15  <a  • •  • «  10  10  o  s  5 —1 I  1  1  1  1  1  1  1  1  1  0 10 20 30 40 50 60 70 80 90 100  1  1  1  1  1  1  1  1  0 10 20 30 40 50 60 70 80 90 100  Plots (%) in Soarse-brome  Figure 14. Species richness in sparse-brome for spring (A, C) andfall (B, D) samples.  35  Dense-brome Species Richness In dense-brome, the relationship between species richness and open-habitat is more complex. The parabolic trend lines (Figure 15) indicate that species richness is lower at low and high proportions of open-habitat area, dense-brome area and number of dense-brome plots, than it is at intermediate proportions. Graphs for spring data (Figure 15 A, C, E) plotting dense-brome against open-habitat area (%), dense-brome area (%) and dense-brome plots (%) are statistically significant atp = 0.001,/? = 0.003 and p = 0.003, respectively. Coefficient of determination (R ) values were: 0.19, 0.151 and 0.15 for the open-habitat area (%), dense-brome area (%) and 2  dense-brome plots (%) regression lines. Fall data was not significant at p = 0.903, p = 0.388 and p = 0.361 for open-habitat area (%), dense-brome area (%) and dense-brome plots (%), respectively (Appendix IV - N through Appendix IV - S). The regression line for fall samples are not shown in Figure 15.  36  A  B  25 CU  25  F esprit = -10.20 + 0.562X- 0.005^ (#=0.19)  CO  • »—I  20  OH CO  15 1  15  10  10  O <L>  C+H  y (Fail) =  4.792 + 0.097Jf-0.00084^ (# = 0.003)  20  o tH CU  B  I  o  o  !1 1 1 1 1 1 1; 1 11| 1 Open-habitat Area (%  — i  0  10 20 30 40  (  D 40 50 60 70 80 90 100 10 20 30  0  50 60 70 80 90 100  25  25  y  CO  i^) = 3.00 + 0.186A'-0.0031A (# = 0.151)  y (F=u)= 5.687 + 0.123^-0.0017^ (#=0.000)  a  (Spl  20  H  2 0  0 o 00  0  0  •§  O  » o oo o O 0 0 0 0 O 00 0 04 0 0 o o oo ® 0 0 00 0 OOO 0 0 0 0 0 o oo  5i  5  0 00 0  0 coo  CU  O GO  15  15  10  10  C M  o cu  a  o> o  O O O O O •  5  5  0  r 10  0  20  30  40  1  50  1  60  O  OO  o o o  o  oo  o  OOO  O O O OO O O 00  O  o  O  o  OO  O  O  O  o  o  00  O  O  O O O  o  o  r-  1  1  70  80  90  100  0  10  20  30  40  50  60  70  80  90  100  Dense-brome Area (%  25 co CU  25  y (8,^=3.88 + 0.14^-0.00178^ (# = 0.15)  20  '5 CU CO  15  Y  20  =5.60 + 0.0959*-0.00IX  (# = 0.001)  1  15  1  C+1  O  2 cu  10  1 0  HD  5 1  1111  1111  —1——1  0  10  20  1  1  1  1  1  1  1  1  30  40  50  60  70  80  90  100  Plots (%) in dense-brome (——1  0  10  20  30  40  50  60  70  80  90  100  Figure 15. Species richness in dense-brome plots graphed against open-habitat area (%), dense-brome area (%) and the number of dense-brome plots (%). These represent spring (A, C, E) andfall (B, D, F).  37  Overall Species Richness Overall species richness includes the total number of species within landscapes. Overall species richness for spring and fall samples are plotted against open-habitat area (%) in Figure 16A and Figure 16B, respectively. Overall species richness is positively correlated with open-habitat area (R R  2  (feii)  (spring)  =  0.75,  = 0.57). Analysis of variance reports that the regression line were significant atp - 0.000  (Appendix IV - T and Appendix IV - U). Overall species richness reflects the contribution of species from each cover type.  38  A  0 "I 0  1  1  1  r  1  1  1  1  1  1  10 20 30 40 50 60 70 80 90 100 Open-habitat Area (%)  Figure 16. Two regression lines showing the total number ofspecies (overall species richne in relation to the transformed percent available open-habitat for spring (A) andfall (B) data  39  CONCLUSION For the open-habitat and sparse-brome types, species richness is correlated with the landscape level of open-habitat. Similarly, within landscapes, plots of open-habitat (100% openhabitat) are the most species rich, sparse-brome (75% open-habitat) supports fewer species than open-habitat and dense-brome (25% open-habitat) has the fewest number of species. Based on these observations, it appears that most species within sparse-brome occur within the openhabitat portions of those plots, while fewer species occupy the sparse-brome portions. Thus, sparse-brome and dense-brome may be more accurately regarded asfinerdepictions of discrete areas of open-habitat and dense-brome respectively, rather than a unique cover type. Areas of Dense-brome appear to be an effective barrier, or at least a very different, and inferior habitat for most species. Dense-brome has relatively fewer species than open-habitat; however, the specific relationship between species richness and open-habitat area, dense-brome area, or the number of dense-brome plots, is complex. Dense-brome plots support few species at low and high levels of open-habitat area, dense-brome area and the number of dense-brome plots. Relatively more species were observed at intermediate levels of dense-brome area. Thus, dense-brome appears to be an inferior alternate habitat that cannot support unique species without a constant source of immigrants.  40  C H A P T E R IV SPECIES A B U N D A N C E A N D DISTRIBUTION  INTRODUCTION Chapter III reported a relationship between species richness and open-habitat area that was consistent with the species-area model. It was noted that this model was based on the species-abundance curve, which empirically describes the observation of many uncommon, and few common species (Preston 1962). Thus, changes in species richness should also be evident in changes of species abundance. In ecological literature, the meanings of abundance and distribution vary (Gaston and Usher 1994, Gaston 1996). Generally, abundance refers to species density or frequency of occurrence, while spatial distribution can refer to the occurrence of a species (Collins and Glenn 1991), but often refers to the range of a species (Gaston 1996). In this thesis, abundance is defined by the number of 10cm X 10cm plots occupied by a species within a 2m X 2m landscape (including only landscapes with at least one observation). At each landscape level of openhabitat, species' abundance is reported by plot cover type.  41  Abundance characterizes populations without consideration of spatial distribution. Distribution is used here to describe the spatial arrangement of species observations within landscapes. In this thesis, distribution is characterized by species' spatial signatures. Species respond to habitatfragmentationdifferently, and the relationship between species abundance and distribution is a fundamental component of how resources are captured (Czaran 1992, Fahrig et al. 1994, Tilman 1994, van der Maarel et al. 1995, Jonsen and Fahrig 1997, Roslin and Koivunen 2001). Thus, distribution of habitat is an essential component in interpreting how habitat fragmentation affects species in these spatially altered landscapes. Changes in species abundance have been forecast by changes in species richness. Thefirstobjective of this chapter is to explore if open-habitatfragmentationaffects species abundance and spatial distribution. Specifically, species abundance is ranked at each landscape level of open-habitat, and these ranking are compared. The second objective of this chapter is to explore if biological guilds are associated with how species respond to open-habitatfragmentation.Species guilds are compared according to resilience to open-habitatfragmentation(resilience)  and tolerance of brome. Species are also  classified according to characteristics of spatial distribution (spatial signature).  Spatial  signatures are compared according to resilience and tolerance of brome. The spatial signatures of guilds are also compared.  42  METHODS  Species Abundance Ranking abundance provides a method to analyze and compare relative changes in community populations (Gaston and Usher 1994). Within each landscape, a frequency distribution of species' abundance was calculated. The mean species abundance is presented for each landscape level of open-habitat and plot cover type {rank  abundance).  Changes in species richness imply changes in species' abundance. Based on the changes in species richness reported in Chapter II, open-habitat fragmentation is expected to change species' abundance in these experimental landscapes. A decline in a species' population results in a leftward shift on the rank abundance graph.  Species Selection Twenty-four species are selected from 1996 data for further analysis. These species, presented in Table 1, represent 21 genera in 14 families. The species codes at the left of Table 1 are used to identify species throughout this document. All species are classified according to 19 biological traits, and species were selected to represent, as completely and uniformly as possible, these guilds. The selection criteria are briefly described in Table 2 and a detailed description of the guild criteria and results are provided in Appendices V and VI, respectively.  43  Table 1. List of sample species selectedfor the analysis of spatial occupancy.  Species Code Family SP2 Primulaceae SP11 Oxalidaceae SP12 Polygalaceae SP15 Scrophulariaceae SP17 Fabaceae SP18 Poaceae SP19 Euphorbiaceae SP20 Asteraceae SP21 Asclepiadaceae SP22 Convolvulaceae SP23 Fabaceae SP25 Fabaceae SP26 Caesalpiniaceae SP28 Fabaceae SP30 Solanaceae SP31 Asteraceae SP35 Caryophyllaceae SP37 Poaceae SP42 Asteraceae Brassicaceae SP43 Poaceae SP69 Euphorbiaceae S106 S110 Euphorbiaceae Euphorbiaceae Sill  Species Androsace Oxalis  Pursh  occidentalis  stricta  Polygonum  L.  aviculare  L.  Veronica  arvensis  Medicago  lupulina  Echinochloa Artemisia  (L.) Beauv. geyeri (Engelm.) Small var. ludoviciana Nutt.  Asclepias  verticillata  Euphorbia  Convolvulus Lespedeza  L. L.  crusgalli  L.  arvensis  L.  Maxim.  stipulacea  Trifolium  repens L.  Gleditsia  triacanthos  L.  Melilotus  officinalis  (L.)  Physalis  Lam.  P. Mill.  virginiana  Ambrosia  psilostachya  Arenaria  serpyllifolia  DC. L.  var.  asper  Ambrosia  artemisiifolia  L.  Capsella  bursa-pastoris  Digitaria  sanguinalis  Sporobolus  asper  Euphorbia  nutans  Acalypha  ostryifolia  Euphorbia  44  geyeri  (Michx.) Kunth  (L.) Medic. (L.) Scop.  Lag.  Riddell cyathophora Murr.  CU  R co  •2  .1  o  I  cu o c cu o o o o  s a  .2 cu  VH  CU  l>  CU co  '•a  o  cu  C  CU •>  CU  V  CO CO  09  |W  cu CO CO  Cd VH  CU  •a  •a  cu  CU  I:  OH  cu  13  I  i-a  co cu o, cd OH cu CO  OH  cu  >  .a .a  co cd  'it  fi. s  >  o  I*  cu co , CO* cu Ms cd "5 VH  CD  CU  •a  •s  cd  1  M VH  CO  o  "a cd IO  T3  <U O  '1  *Q cj •fx*  IP  s  IK  VH  OH CO  >  cu cu  "td  •SJ  'o ca  .9  00  o  R  R  s  K  R  6 55  a  U  o U  S J3  M»I  u  M  M  S S  cu o  O cu  H-»  M) CU  S3  ca  U  S  u  -~.  •»»  OH  1  Q<  -R -R -R  PH  6 ,  6 , 5b  bb  O O O o •  co CU  o ^=«  cj  &o & O  3  CJ  O  sI PH  CO &5  S  o  cu CO  O  o  tc  CO  ^  a  a a  o  o  ,1  O  £  o  to  o  R  VH  CU  CJ CJ  R  CO CO  , 1)  CO  cS  cj  'a  bo  CJ  5  CU  PH  R  o  CU  CU  "a cu £ a  R  cd  VH  VH  •  i  i  ^ y fe2 ^ ^ &5 H; U <J a< o<  Spatial Distribution of Species Patchiness has been shown to effectively characterize species' spatial characteristics (Fortin et al. 1999), thus spatial signature analysis should detect distributional changes resulting from open-habitat fragmentation. The tractable character of these experimental landscapes (for example, the signature of open-habitat presented in Figure 8) allows the spatial distribution of individual species to be compared with the spatial distribution of habitat.  Classifying Successional Status Throughout the process of establishment, species progress through phases that can be spatially characterized (Fortin et al. 1999). The colonizing phase exhibits numerous, small patches producing a linear relationship (roughly 1:1) (Figure 17A). As individual plants grow and compete for space with varying success, variation becomes apparent in the numbers and sizes of patches. Once established, patch characteristics vary among species, but each species shows less variability in their number and size than colonizing species (Figure 17B). As part of the spatial occupancy classification, species are classified according to successional status, colonizing ( Q species and established (E) species are distinguished.  46  B  CO  <D  30  30  25  25  o  "S  PH  20  20  H  CM  O  15  15  <U  £  1997  io  1998  10  5  5  20  40  60  80  100  0  20  40  60  80  100  Ambrosia psilostachya Plots (%)  D 30 CO  25  J3 o  20  cu  PH M-H  O  15  CU  30 25 20  i  15  X>  6  ' Habitat d a s 35  10  10 5  5  0  0 0  20  40  60  80  20  100  40  60  80  100  Ambrosia psilostachya Plots (%)  Figure 17. Distinguishing the effects of successional status from the varying proportions of open-habitat area for Ambrosia psilostachya (Species 31). Within the 75% and 85% openhabitat classes (combined), the spatial occupancy of Species 31 changedfrom a colonizing species inl997 (A) to an established species in 1998 (B). These effects were not uniform, a shown with the 1998 data in Figures C and D. The spatial signature evident at 35% openhabitat (C) is different from the signature at 85% open-habitat (D).  47  Spatial Signature Classification To detect the effects of open-habitat fragmentation, patch observations are further symbolized according to the open-habitat class of the landscape where it was observed (in Figures 18 and 19 the stylized symbols O, X, and O depict high, intermediate and low levels of open-habitat). A fragmentation effect {fragmentation effect or F) is apparent when a species' spatial occupancy changes progressively through the varying levels (O, X, and O symbols in Figures 18 and 19) of open-habitat (Figure 18EF and Figure 18CF). Species apparently unaffected (effect unapparent or U) by the area and arrangement of open-habitat are defined by the irregular changes in the spatial occupancy symbols (Figure 18EU and Figure 18CU). Species' spatial signatures are also classed according to their position relative to the spatial signature of open-habitat. In Figures 18 and 19, the spatial signature of open-habitat is shown as the grey area, this is a stylized version of the spatial signature illustrated in Figure 8. Established species were further subdivided as similar (S) or dissimilar (D), based on whether the shape of the spatial signature was similar to that of open-habitat. Similar and dissimilar species were further subdivided as above (A) or below (B), indicating whether spatial occupancy observations tended to occur above or below the spatial signature of open-habitat. Examples of these classes are provided in Figure 19. The spatial signatures of colonizing species were also classed according to their position relative to the spatial signature of open-habitat. However, establishment is also influenced by habitat area making subdivisions difficult. For example, where open-habitat area is relatively high (85%), the spatial occupancy of Ambrosia psilostachya (Species 31) progresses predictably, as shown in Figure 17D. In areas where there is little open-habitat, this progression is  48  dramatically retarded (Figure 17C). Colonizing species are subdivided into two broad classes: fragmented (F) and unapparent (U). Examples of colonizing/fragmented and colonizing/unapparent species are also provided in Figure 18. Species that cannot be classified for lack of observations are described as insufficient data (I).  49  Fragmentation Effect (F)  Fragmentation Unapparent (U) CO 25 JC  o 20 -  cS  OH  15 -  O u 10 <u  w  s  o  5 -  3  X  0  0 0 sd  10 2 0 3 0 4 0 5 0 6 0  70  8090  0  Plots Occupied (%)  CO  U  OH  O <u  X)  0*  a C  20  60 70 80 90  Signature Classification (EU) • clumped open-habitat area symbols • varied spatial occupancy  cu XI  "  %  JM 20 -  OH  *s  15  - j ^ B ^  1515  10  Xi E  s  I  50  cn 25  25  X!  I  30 40  Plots Occupied (%)  Signature Classification (EFS) • progression of open-habitat area symbols • varied spatial occupancy • signature shaped similar to open-habitat co  10 20  I  J ^ ^ ^ ,  10  5  0 0  10 20 30 40 50 60 70 80 90  0  10 20 30  40  50  60 70  80 90  ©  u  Plots Occupied (%)  Plots Occupied (%)  Signature Classification (CF) • progression of open-habitat area symbols • truncated spatial signature • linear 1:1 number/area ratio a  High Open-habitat Area  Signature Classification (CU) • irregular arrangement of open-habitat symbols • truncated spatial signature • linear 1:1 nunber/area ratio X Inter. Open-habitat Area Low Open-habitat Area  Figure 18. A stylized comparison of spatial signature classes. The generalizedforms of Openhabitat spatial signature are shaded, landscape Open-habitat symbols are simplified and only three levels of open-habitat are shown. 50  Established, Fragmented (EF) Dissimilar (D)  Similar (S)  co  25  <u 25  X! O  cu  3 20 0  f j 20 PH  OH  15  O  10  E  r  0  >  0 10 2 0 3 0 4 0 5 0 6 0 70 8090  0 10 2 0 3 0 4 0 5 0 6 0 70 8090  o  xs  Plots Occupied (%)  Plots Occupied (%)  Signature Classification (EFSA) • progression of open-habitat area symbols • varied spatial occupancy • signature shaped similar to open-habitat • more patchy spatial occupancy  Signature Classification (EFDA) • Progression of open-habitat area symbols • varied spatial occupancy • signature shape departs from openhabitat • more patchy spatial occupancy  <  cn  CD  25  25 CO  X  20  Pi  15  i-  10  o  43  fj  15 C+H  O  10  X  CU  E  I  20  OH  CD  co  15  CD  X  x  «  O  x  X —i  1  1  1  r  1  1  1  1  — i — i — i — i — i — i — i — i — i —  0 10 20 30 40 50 60 70 80 90  0 10 20 30 40 50 60 70 80 90  Plots Occupied (%)  Plots Occupied (%)  Signature Classification (EFSB) • progression of open-habitat area symbols • varied spatial occupancy • signature shaped similar to open-habitat • less patchy spatial occupancy  Signature Classification (EFDB) • progression of open-habitat area symbols • varied spatial occupancy • signature shape departs from openhabitat • less patchy spatial occupancy  O High Open-habitat Area  X Inter. Open-habitat Area  Low Open-habitat Area  Figure 19. A stylized comparison of the spatial signatures classes established, colonizing, fragmented and unapparent. The generalized form of Open-habitat are shaded. 51  RESULTS  Species Abundance Species abundance for each class of open-habitat is illustrated in Figure 20. Declines in species abundance, and a drop in the number of species is evident through declines in the landscape level of open-habitat. As populations of common species (high abundance classes) decline, their rankings shift leftward. For the lower abundance classes (rare species), bar lengths are shorter at low levels of open-habitat. At the lowest landscape level of open-habitat only rare species exist. This result is expected, and is consistent with the species richness analyses reported in Chapter III. Changes in the number of species within each abundance class were not uniform, but are roughly proportional to the initial number of species within the abundance class (i.e. the relative proportions of common and uncommon species remain similar at different levels of openhabitat). However, this analysis cannot discern whether species that were rare at high proportions of open-habitat were resilient to open-habitat fragmentation, or if the populations of species common at high levels of open-habitat declined with reductions in open-habitat. The effects of initial abundance and distribution are further considered in the analysis of spatial occupancy.  52  4 5 % Open-habitat  90  130  170  210  250  290  Number of Plots Occupied  Figure 20. Species abundance and richness for three levels of open-habitat.  53  Differences in the shapes of the rank abundance graphs were tested according to the Kolmogorov-Smirnov Two-Sample-Test (K-S Test). The results of these tests, found in Appendix VII, indicate that extensively (28/30), the rank abundance graphs of different levels of open-habitat are not statistically different in shape. These tests reflect the proportional changes in abundant and rare species, and understate the biological significance of these changes. For example, Figure 21 illustrates the range of abundance classes within each open-habitat category. Below 65% the range of abundance drops remarkably, from 14 classes at 65% to three and four classes (spring and fall, respectively) at 55% open-habitat. Thus, relatively abundant species are severely affected at intermediate levels of open-habitat.  54  300 i  — Spring ^-Fall  _B  250 -  200  150  100  50  0  "i  1  1  1  r  35%  45%  55%  65%  75%  85%  Open-habitat Class Figure 21. Maximum species abundance for each open-habitat area (%) class including all 1996 spring andfall data.  55  Resilience to Fragmentation and Biological Guilds Differences in biology determine species abundance, distribution and the capacity to exist in fragmented landscapes. To investigate the relationship between biological guilds and abundance, resilience values are compared between guilds. Resilience is measured using two methods:first,the minimum open-habitat class (MHC) occupied by a species, and as the minimum abundance open-habitat class (MAC) where at least 25% of the maximum species population exists. The MAC statistic provides a broader range of mean resilience values and is considered the most useful for subsequent analyses. Unless explicitly stated, resilience is described by the MAC statistic throughout this thesis. The MHC and MAC analyses are included in Appendix VIII - B and VIII - C. Open-habitatfragmentationinfluences species abundance across all guilds, however, the specific impacts vary (Appendix IX). The most resilient species are fall occurring (MAC at 45.0% open-habitat), although comparisons of resilience between seasonal categories suggest the differences are varied. Comparisons based on Student-t tests report (Appendix IX - B), springfall (p = 0.004), spring-spring and fall (p = 0.096), and fall-spring and fall (p = 0.058). Similarly, native species are resilient, although not significantly more than introduced species (p = 0.01). The grasses guild (other than brome) had a MAC value of 55.0% open-habitat, lower than all other plant forms. However, there are few samples of trees or grasses, and the statistics reflect this sampling artifact, tree - herb and forb (p = 0.486), tree - grass (p = 0.478), and grass - herb and forb (p = 0.494). Non-clonal species are less resilient (65.7% open-habitat) than clonal species (57.0% open-habitat in Appendix IX) at surviving infragmentedlandscapes, 56  although results are statistically insignificant (Appendix IX - B). To explore how initial abundance and spatial distribution influence resilience, species are classified according to Rabinowitz's (1981) topology of rarity (a classification of species distribution), and classes are compared. Based on this small sample of species, the capacity to exist in fragmented landscapes is less associated with local abundance than geographic abundance. Geographically rare species (including GRHCLC and GRHCLR species) are most successful at surviving (MAC 55.0% open-habitat in Appendix IX). However, local abundance is estimated from a very small sample and differences in resilience cannot be statistically distinguished among the different forms of rarity (Appendix IX - B).  Brome Tolerance, Biology Guilds and Resilience to Fragmentation Chapter III reported that dense-brome plots contributed to overall species richness. Based on results of the species-area models presented in Chapter III, species capable of tolerating brome are expected to exist most abundantly in fragmented landscapes. To explore the relationship between species tolerance and resilience, the propensity of species to inhabit brome occupied plots are compared. This propensity is measured as the mean number of plots concurrently occupied by a sample species and sparse-brome or dense-brome. Tolerance is presented for sparse-brome and dense-brome in Appendix VIII - D and Appendix VIII - E, respectively. Generally, the rankings of dense-brome tolerance reflect the rankings of resilience. Table 3, a summary of Appendix IX, reports the ranking of brome tolerance (most tolerant is one, least tolerant is four) and resilience (most resilient is one, least resilient is four). Specifically, four of the rankings of dense-brome tolerance (Seasonality,  57  Origin, Form, Dispersal and Photosynthetic Pathway C4) match the rankings of resilience. These results suggest that tolerance is a trait that enhances resilience in fragmented landscapes.  Table 3. Ranking of resilience and tolerance partitioned by biological attributes for 1996 sample species.  3 1 2 1 2 2 1 3 3 2 1 1 2 2  DenseBrome Tolerance 3 1 2 1 2 1 2 3 3 2 1 1 2 1  GCHCLC  4  1  GCHCLR GRHCLC GRHCLR  3 1 1  4 2 3  Biological Guild Spring Fall Spring and Fall Native Origin Introduced Plant Annual Characteristics Life Perennial Duration Biennial Tree Form Herb/Forb Grass Clonal Dispersal Non-clonal Photosynthetic Pathway C4 Seasonality  Rabinowitz's Topology of Rarity  Resilience (MAC)  GCHCLC - Geographically-Common, Habitat-Common, Locally-Common; GCHCLR Geographically-Common, Habitat-Common, Locally-Rare; GRHCLC - Geographically Rare, Habitat-Common, Locally-Common, GRHCLR - Geographically-Rare, Habitat-Common, Locally-Rare.  Ranking criteria: dense-brome tolerance (most tolerant is one, least tolerant is four), resilien (most resilient is one, least resilient is four).  58  Spatial Distribution of Species Differences in the tolerance of brome and resilience to open-habitat fragmentation indicate that species respond to open-habitat fragmentation differently. Based on these observations, variations in spatial distribution were expected. Spatial signatures are presented in Appendix X, a graphic summary is provided in Figure 22.  •  Fragmented/Similar  Hi Fragmented/Similar/Below  Insufficient Established  •  Fragmented/Similar/Above  •  Fragmented/Dissimilar/Below  ^  Fragmentation Unapparent  Colonizing 4  8  12  Number of Species Figure 22. Spatial signature classification of sample species.  The 24 sampled species occupy six of the eight broad spatial signature categories. Four of the species studied cannot be classified for lack of observations. Of those species further classified, 12 (60%) were colonizing and the remaining eight species (40%) are considered established. For colonizing species, seven are considered fragmented and five are classed as unapparent. Of the established species, seven are fragmented and one is unapparent. Of the established and fragmented species, seven are similar (to signature of open-habitat), two are similar-above, two are similar-below, one is similar-below, and two are dissimilar-below.  59  The classification of these spatial signatures are presented in Appendix XI, and a summary of spatial signature classifications is presented in Appendix XII, which reports the expected and actual observations according to each combination of biological guild and signature class. These results indicate that species spatial occupancy was influenced by open-habitat fragmentation.  Evaluating Spatial Signature Classifications Spatial signature classifications are map truthed. Observations of each species are plotted on landscape maps that distinguished the open-habitat, sparse-brome and dense-brome cover types, and the classification are considered in the context of the relationship between openhabitat and species locations. Sample maps for species Medicago lupulina, Lespedeza stipulacea and Sporobolus asper are included in Appendix XIII. This survey indicates that signature classifications generally reflect changes in the abundance of sample species throughout the full range of open-habitat levels. However, two observations warrant further explanation. First, maps indicate that species' spatial distributions vary dramatically, within and between the various treatment levels. When graphed as signatures, these variations are tempered by the aggregation of observations from numerous landscapes. Second, an underlying assumption in the classification of signatures is that similarities in the signatures of open-habitat and species would reflect similarities in their distribution, that is, species observations are expected to occur within open-habitat. While this assumption often holds true, there are exceptions, and species observations often occur throughout the various cover types.  60  Spatial Signature and Resilience to Fragmentation Species' capacity to exist within fragmented landscapes, whether in association or independent of brome, was expected to manifest through distinct characteristics in spatial occupancy. The relationship between species resilience and spatial signature is compared using the mean resilience values of the different signature classes (Appendix XIV). It was expected that species classified as unapparent would exist with greater abundance in landscapes of low open-habitat. Measures of resilience suggest that, for both colonizing and established species, this expectation is met. While open-habitat fragmentation does influence species' ability to colonize (Figure 17), these results indicate that both colonizing and established species are affected by open-habitat fragmentation. However, comparisons based on the Student-t test are not statistically significant, with colonizing (p = 0.182) and established (p = 0.592) (Appendix XIV - B). Comparisons of subordinate classifications (such as between EFS and EFSA) show similar results.  Spatial Signature and Brome Tolerance Chapter III concluded that at intermediate levels of open-habitat, dense-brome enhanced species richness by providing alternate habitat for some species. These results are supported by non-spatial comparisons between brome tolerance and resilience. The spatial characteristics of tolerant species are described by the tolerance within each spatial signature classification (Appendix VIII - D and Appendix VIII - E). The mean tolerance is compared between signature classes, a summary of those results are presented in Appendix IX.  61  The results of these comparisons vary. For colonizing species, there is a notable difference in brome tolerance between fragmented and unapparent species (26.7 and 88.8 observations/landscape, respectively) (Appendix XIV). Unapparent species are much more tolerant than fragmented species. Other analyses indicated that brome tolerant species are most resilient, and this analysis further suggests that based on spatial occupancy, tolerant and intolerant species cannot be clearly distinguished. However, these results are averaged across a range of open-habitat levels so potential changes associated with the availability of open-habitat cannot be detected.  Spatial Signature and Biological Guilds Differences in biology determine species' capacity to exist in fragmented landscapes, these differences were expected to produce variations in spatial signatures. The 20 species classified represented 18 of the 19 biological guilds considered (C3 and non-clonal are not explicitly presented in Appendix XII, although they are included in expected value calculations). In 16 of these categories, the abundance of fragmented species meets or exceeds expectations. However, differences between the actual and expected number of species is often small. Nevertheless, the negative effects of open-habitat fragmentation are pervasive in the experimental landscapes.  62  CONCLUSION Changes in rank abundance reflect observed changes in species richness. At different levels of open-habitat, the relative proportions of common and rare species remain similar, generally reflecting predictions of the species-area model. Ranking abundance does detect the level of open-habitat fragmentation that most severely affects common species. However, the analysis cannot indicate which species are extirpated. Specifically, the analyses did not predict whether species that were initially rare would exist in, or expire from, these fragmented landscapes. Rank abundance analyses do provide insight into how grassland communities are affected by open-habitat fragmentation. Changes in the ranges of occupied rank abundance categories indicate that species are affected most dramatically at intermediate levels of open-habitat, yet numerous species exist at lower levels of open-habitat. This resilience is attributed, in part, to species' capacity to colonize sparse-brome and dense-brome plots. This conclusion is based on results from the dense-brome species-area model and comparisons of dense-brome tolerance and resilience. Changes in species abundance resultingfromopen-habitatfragmentationare categorized according to their signatures. Casual comparisons between species observation maps and spatial signatures indicate that spatial signatures are responsive to changes in habitat distribution, and reasonably depict changes in spatial distribution. However, spatial signature classification does not predict species locations. Although statistically unsubstantiated, the classification system was able to distinguish which sample species existed in these experimentally fragmented landscapes. As expected,fragmentedspecies are least successful and unapparent species are  63  most successful in fragmented landscapes. Open-habitat fragmentation affects some species' capacity to colonize. However, openhabitat fragmentation affects both colonizing and established species. In these randomly structured grassland landscapes, open-habitat fragmentation constrains spatial distribution, reduces species abundance, and ultimately reduces species richness.  64  CHAPTER V CONCLUSION  The landscape level of open-habitat influenced the capacity of species to exist in these experimental landscapes. This trend was apparent in three assessments, species richness, abundance and spatial occupancy. Species richness was correlated with open-habitat area, and generally consistent with the species-area model. Although overall species declines were tempered by the colonization of alternative habitat, landscapes at low levels of open-habitat were species impoverished. Overall species abundance was also correlated with open-habitat area, although the relationship between abundance and open-habitat was not proportional. Through intermediate proportions of open-habitat (55% and 65%), abundant species were affected dramatically. These levels of open-habitat were roughly consistent with the percolating threshold, thus open-habitat connectivity appeared to influence species abundance. Variations in species abundance further indicate that species respond to changes in the landscape level of open-habitat differently.  65  Comparisons between the spatial signatures of 24 species indicate that spatial distributions were affected by open-habitatfragmentation.Although not all species were influenced, or were influenced in the same way, the general character of these changes could be coarsely categorized. Despite the limitations of this simplistic classification scheme, patch characteristics reflected the relative influence of open-habitat fragmentation. The location of open-habitat was thought to influence the location of species in fragmented landscapes. That is, species were expected to occur extensively within plots of openhabitat. However, map analyses suggested that species often occur with sparse-brome and dense-brome. Thus, species' distributions likely reflect a balance between available openhabitat, initial location, and the capacity and form of dispersal. Brome tolerance appeared to be a valuable trait toward survival in the experimentally fragmented landscapes. Species richness was correlated with the landscape level of open-habitat, although at intermediate levels of open-habitat the number of species inhabiting dense-brome increased. Despite the relatively small size of these experimental landscapes, logistic and experimental challenges were encountered. The grassland communities studied were responsive, nevertheless the short study period meant that the ecological responses reported for 1996 were amidst rapid change, and this change continued throughout the study period. Similarly, only a portion of the 1996 data collected could be spatially analyzed. Certainly more sample species would have enhanced the interpretative and statistical confidence of these analyses. Similarly, the results reported here are specific to these experimental prairie landscapes, more general predictions of the effects of habitatfragmentationcannot be inferred.  66  The experimental creation of fragmented habitats (creating habitat through disturbance) differed from the typical process of fragmentation where habitat remains unaltered while the disturbed surroundings represent non-habitat. This process contrasts the predicament of species inhabiting remnant patches of habitat embedded within a disturbed landscape. Thus, extending the observations for these experimental landscapes to other fragmented landscapes is tenuous. In Chapter I, fragmentation was described as the loss, dispersion and isolation of habitat. While variations in the levels of open-habitat were clearly reported, changes in the dispersion and isolation of these areas were collectively expressed as changes in distribution (spatial signature). Although the expansion of brome imposed some variation in distribution of openhabitat, changes to isolation and dispersion could not be characterized. The relative consequences of these elements of fragmentation would be valuable knowledge in landscape management. Despite limitations in its design, analyses and techniques, this research may provide an alternate approach for future research into the effects of habitat fragmentation. The ability to predict consequences based on the arrangement of habitat within a landscape is essential to the management of habitat fragmentation. The capacity to predict the effects of various habitat arrangements requires a theoretical foundation. This thesis presented the island biogeography theory as the historical basis for concerns regarding habitat fragmentation, and applied percolation theory in the analysis of these experimental landscapes. Results of species richness were also cast in the context of island biogeography theory. However, species richness is an expression of changes in the abundance and distribution of species. This empirical research lacked a theoretical explanation for observed changes in abundance and distribution. Such an  67  explanation might be provided in the core-satellite hypothesis, metapopulation theory, intermediate disturbance hypothesis or the habitat niche hypothesis. While future work would benefit from an increased sample size, expanded scale, variation in ecotypes, as well as the distinction of area loss, dispersion and isolation, the need to provide a theoretical explanation for observations reported within fragmented landscapes is paramount.  68  LITERATURE CITED Bartha, S., S. L. Collins, S. M. Glenn, and M. Kertesz. 1995. Fine-scale spatial organization of tallgrass prairie vegetation along a topographic gradient. Folia Geobotanica & Phytotaxonornica 30:169-184. Blankespoor, G. W., and J. K. May. 1996. Alien smooth brome (Bromus inermis Leyss.) in a tallgrass prairie remnant: seed bank, seedling establishment, and growth dynamics. Natural Areas Journal 16:289-294. Bunnell, F. L. 1999. What habitat is an island? Pages 1-31 In J. A. Rochelle, L. A. Lehmann, and J. Wisniewski, editors. Forest Fragmentation: Wildlife and Management Implications. Koninklijke Brill NV, Boston. Bunnell, F. L., and J. F. Johnson, editors. 1998. Policy and Practices for Biodiversity Managed Forests: the Living Dance. UBC Press.  in  Cissel, J. H., F. J. Swanson, and P. J. Weisberg. 1999. Landscape management using historical fire regimes: Blue River, Oregon. Ecological Applications 9:1217-1231. Collins, S. L., and S. M. Glenn. 1991. Importance of Spatial and Temporal Dynamics in Species Regional Abundance and Distribution. Ecology 72:654-664. Czaran, T. B. S. 1992. Spatiotemporal dynamic models of plant populations and communities. Trends in Ecology & Evolution 7:38-42. Daily, G. C , P. R. Ehrlich, and N. M. Haddad. 1993. Double keystone bird in a keystone species complex. Proceedings of the National Academy of Sciences of the United States of America 90:592-594. Dale, V. H., S. M. Pearson, H. L. Offerman, and R. V. O'Neill. 1994. Relating patterns of landuse change to faunal biodiversity in the Central Amazon. Conservation Biology 8:10271036. Diamond, J. M. 1975. The island dilemma: lessons of modern biogeographic studies for the design of nature reserves. Biological Conservation 7:129-146. Ehrlich, P. R. 1992. Population biology of checkerspot butterflies and the preservation of global biodiversity. Oikos 63:6-12. Fahrig, L. 1997. Relative effects of habitat loss and fragmentation on population extinction. Journal of Wildlife Management 61:603-610.  69  Fahrig, L., D. P. Coffin, W. K. Lauenroth, and H. H. Shugart. 1994. The advantage of longdistance clonal spreading in highly disturbed habitats. Evolutionary Ecology 8: 172-187. Fahrig, L., and G. Merriam. 1994. Conservation of fragmented populations. Conservation Biology 8:50-59. Fortin, M-J., S. Payette, and K. Marineau. 1999. Spatial vegetation diversity index along a postfire successional gradient in the northern boreal forest. Ecoscience 6:204-213. Gaston, K. J. 1996. The multiple forms of the interspecific abundance-distribution relationship. Oikos 76:211-220. Gaston, K. J., and M. B. Usher. 1994. Rarity, 1st edition. Chapman & Hall, New York. 205. Great Plains Flora Association, editor. 1977. Atlas of the Flora of the Great Plains, 1st edition. The Iowa State University Press, Ames. 600. Great Plains Flora Association, editor. 1986. Flora of the Great Plains. University Press of Kansas, Lawrence. Gustafson, E. J., and G. R. Parker. 1992. Relationships Between Landcover Proportion and Indices of Landscape Spatial Pattern. Landscape Ecology 7:101-110. Harrison, S., and J. F. Quinn. 1989. Correlated environments and the persistence of metapopulatins. Oikos 56:293-298. Hessburg, P. F., B. G. Smith, S. D. Kreiter, C. A. Miller, R. B. Salter, C. H. McNicoll, W. J. Harm, and T. M. Quigley. 1999a. Historical and current forest and range landscapes in the interior Columbia River basin and portions of the Klamath and Great Basins. Part 1: Linking vegetation patterns and landscape vulnerability to potential insect and pathogen disturbances. U S Forest Service General Technical Report PNW 0: 1-357. Hessburg, P. F., B. G. Smith, and R. B. Salter. 1999b. Detecting change in forest spatial patterns from reference conditions. Ecological Applications 9:1232-1252. Hessburg, P. F., B. G. Smith, and R. B. Salter. 1999c. Using estimates of natural variation to detect ecologically important change in forest spatial patterns: A case study, Cascade Range, eastern Washington. U S Forest Service Research Paper PNW 0:1-65. Hunter, M. L. 1996. Fundamentals of Conservation Biology. Blackwell Science Inc., Cambridge, Mass. 482. Jonsen, I. D., and L. Fahrig. 1997. Response of generalist and specialist insect herbivores to landscape spatial structure. Landscape Ecology 12: 185-197. MacArthur, R. H., and E. 0. Wilson. 1967. The Theory of Island Biogeography. Princeton University Press, Princeton. Mack, A., and T. Contreras. 1995. Page 244 In A. Joern and K. H. Keeler, editors. The Changing Prairie: North American Grasslands. Oxford University Press, New York. Maurer, B. A., and S. G. Heywood. 1993. Geographic range fragmentation and abundance in neotropical migratory birds. Conservation Biology 7:501-509. 70  Mclntyre, N. E., and J. A. Wiens. 1999. Interactions between habitat abundance and configuration: experimental validation of some predictions from percolation theory. Oikos 86:129-137. Milne, B. T., M. G. Turner, J. A. Wiens, and A. R. Johnson. 1992. Interactions Between the Fractal Geometry of Landscapes and Allometric Berbivory. Theoretical Population Biology 41:337-353. O'Neill, R. V., S. J. Turner, V. I. Cullinan, D. P. Coffin, T. Cook, W. Conley, J. Brunt, J. M. Thomas, M. R. Conley, and J. Gosz. 1991. Multiple Landscape Scales an Intersite Comparison. Landscape Ecology 5:137-144. Palmer, M. W., and T. A. Maurer. 1997. Does diversity beget diversity? A case study of crops and weeds. Journal of Vegetation Science 8:235-240. Pearson, S. M., M. G. Turner, R. B. Gardner, and R. V. O'Neill. 1996. An organism-based perspective of habitat fragmentation. Pages 77-95 In R. C. Szaro and D. W. Johbston, editors. Biodiversity in managed landscapes: theory and practice. Oxford University Press, New York. Pickett, S. T. A., and D. S. White, editors. 1985. The Ecology of Natural Disturbance and Patch Dynamics. Academic Press, New York. Power, M. E., D. Tilman, J. A. Estes, B. A. Menge, W. J. Bond, L. S. Mills, G. Daily, J. C. Castilla, J. Lubchenco, and R. T. Paine. 1996. Challenges in the quest for Keystones: Identifying keystone species is difficult-but essential to understanding how loss of species with affect ecosystems. Bioscience 46:609-620. Preston, F. W. 1962. The canonical distribution of commonness and rarity. Ecology 43:185215:410-432. Robinson, G. R., R. D. Bolt, M. S. Gaines, S. P. Hamburg, M. L. Johnson, B. S. Pitch, and E. A. Martinko. 1992. Diverse and contrasting effects of habitat fragmentation. Science 257:524-526. Roslin, T., and A. Koivunen. 2001. Distribution and abundance of dung beetles in fragmented landscapes. Oecologia 127:69- 77. Rudis, V. A. 1995. Regional forest fragmentation effects on bottomland hardwood community types and resource values. Landscape Ecology 10:291-307. Simms, P. L. 1988. Grasslands. In M. G. Barbour and W. D. Willings, editors. North American Terrestrial Vegetation. Cambridge University Press, New York. Stauffer, D., and A. Aharony. 1994. Introduction to Percolation Theory, 2nd edition. Taylor & Francis Ltd, London, U .K. 181. Stouffer, P. C , and R. 0. Bierregaard, Jr. 1993. Spatial and temporal abundance patterns of ruddy quail-doves (Geotrygon montana) near Manaus, Brazil. Condor 95:896-903. Symstad, A. J., D. Tilman, J. Willson, and J. M. B. Knops. 1998. Species loss and ecosystem functioning: Effects of species identity and community composition. Oikos 81:389-397. 71  Tilman, D. 1994. Competition and biodiversity in spatially structured habitats. Ecology 75:2-16. Tilman, D. 1996. Biodiversity: Population versus ecosystem stability. Ecology 77:350-363. Tilman, D., D. Wedin, and J. Knops. 1996. Productivity and sustainability influenced by biodiversity in grassland ecosystems. Nature 379:718~ 720. Turner, I. M., and R. T. Corlett. 1996. The conservation value of small, isolated fragments of lowland tropical rain forest. Trends in Ecology & Evolution 11:330-333. Turner, M. G., Robert V. O'Neill, Robert H. Gardner and Bruce Milne. 1989. Effects of changing spatial scale on the analysis of landscape pattern. Landscape Ecology 3:153-162. Turner, M. G., Y. Wu, W. H. Romme, and L. L. Wallace. 1993. A landscape simulation model of winter foraging by large ungulates. Ecological Modelling 69:163-184. van der Maarel, E., V. Noest, and M. W. Palmer. 1995. Variation in species richness on small grassland quadrats: Niche structure or small-scale plant mobility? Journal of Vegetation Science 6:741-752. Wallin, D. O., Frederick J. Swanson, B. Marks, John H. Cissel, and J. Kertis. 1996. Comparison of managed and pre-settlement landscape dynamics in forests of the Pacific Northwest, USA. Forest Ecology & Management 85:291-309. Weaver, J. E., and F. W. Albertson. 1956. Grasslands  of the Great Plains.  Johnsen Publishing  Company, Lincoln, Nebraska. 395. Wiens, J. A. 1995. Habitat fragmentation: Island v landscape perspectives on bird conservation. Ibis 137:S97-S104.' Wiens, J. A., R. L. Schooley, and R. D. Weeks; Jr. 1997. Patchy landscapes and animal movements: do beetles percolate. Oikos 78:257-264. Wilcox, B. A. 1980. Insular ecology and conservation. Pages 95-117 In M. E. Soule and B. A. Wilcox, editors. Conservation Biology: An Evolutionary Ecological Perspective. Sinauer Associates, Inc., Sunderland, Massachusetts. With, K. A. 1997. The application of neutral landscape models in conservation biology. Conservation Biology 11:1069-1080. With, K. A.; Sean J. Cadaret, and C. Davis. 1999. Movement responses to patch structure in experimental fractal landscapes. Ecology 80:1340-1353. With, K. A., and T. 0. Crist. 1996. Translating across scales: Simulating species distributions as the aggregate response of individuals to heterogeneity. Ecological Modelling 93:125-137. With, K. A., and A. W. King. 1997. The use and misuse of neutral landscape models in ecology. Oikos 79:219-229. With, K. A., and A. W. King. 1999a. Dispersal success on fractal landscapes: A consequence of lacunarity thresholds. Landscape Ecology 14:73-82.  72  With, K. A., and A. W. King. 1999b. Extinction thresholds for species in fractal landscapes. Conservation Biology 13:314-326.  73  APPENDIX I  SAMPLE OF FIELD DATA CARD Date  Plot  Observers  Treatment  1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Each cell represents a 0.1m X 0.1m plot -within a landscape (experimental unit).  74  19  20  -a  CO  o H • — 00  ~« 2;  "«  cu  CN  CO  CO «  Si  on  a »  CN  to  cc  «5  cu to  £  CO  2  —  o  CN  43 cu  v©  "3 2;  CO  vo  011  CC  Z  o  a  £  *o  CN  CN  CS  w cc  Q  Z W  Pi PM  n o M  VO  ON  H  > OS  cu  IT)  2  s£  H  •5 a CN i - i CO  CU  o  sj  ON  cc  CC  CJ  •5  •«  •I  H  •Si  I  cc  u Q  UJ  CO  U W PN  CC  es  co •« «  CJ  cu  CO  '5  cu  O  CU  a.  "s  "5b  cc  to 51 co  cj  §  s s c  S  o  S  CO  cu  CU  a cu  to  CL  a -^s  cu  &  S4,  52 c? •Q cj  ,1  8 CU  0  s  •2 ca  Si s:  *  -S  -C cu  CU  "S,  cj  O  cu  CO  st  sa,  ro  o  TT  o  CN  t--  cn  Tf  o  O  o  o  o r- o  o  ©  O  TT o  O cn  CN o  o  o  CN o  cn © TT  o  O  o  o  o  Tf  o  m ro  o  CN CN o  o  CN  o  o  m  O  oo  o  Tf  o  r- o  ©  cn  CN o  o  TT o  o  o  ©  o  o  O  o  o  o  o  o  o  o  o  CN  o  o  -  NO  U U o  o  o  o  o o ©  CN o  o CN  NO  o o CN ro  cn o o  cn  o o NO  o o O  NO  r~- o Tf  ©  00 ©  o o o  CO  ro  o CN  o  ON  r- o  00  O  O o  Draba  00  (L.) Webb ex Prantl reptans (Lam.) Fern.  ©  ON  sophia  o  CN  Descurainia  o  rcn  bursa-pastoris (L.) Medic.  cn  oo  Capsella  Tf  Tf  Vernonia  SO CN cn  Torr.  Tragopogon  CN  baldwinii  dubius Scop.  Taraxacum  cn  Leontodon  spp. (1) spp. (1) spp. (1)  Lactuca  CN  SP81 SP61 SP58 SP29 SP55 SP89 SP43 SP4 SP5  o  Kuhnia  Tf  CN  spp. rigidus  CN  cn  o  eupatoriodes (L.)  00 00 NO cn CN TT ON CN  (Pursh) Dunal  cn  altissimum L.  o  o  Eupatorium  cn © O CN  Muhl. ex Willd.  o  strigosus  Tf  annuus (L.) Pers.  ©  Erigeron  O  canadensis (L.) Cronq.  ro  Conyza  oo  CN cn  vulgare (Salvi) Ten.  r- o o o  (L.) Hill undulatum (Nutt.) Spreng.  ON  altissimum  ON  Cirsium  r—t  Brassicaceae  CN  o  rigidus (Cass.) Desf.  Carduus  NO  NO  annuus L.  spp. (1) spp. (1)  ON  Helianthus  ericoides L.  Aster  i—i  squarrosa  ludoviciana Nutt.  Artemisia  Asteraceae cont.  Code Spring Fall Spring Fall Spring Fall Spring Fall Total 1995 1995 1996 1996 1997 1997 1998 1998 2949 SP20 SP93 SP70 SP48 SP49 SP51 SP40 S103 SP39 SP24 SP90 S147 SP57 S135 ON  Grindelia  Species  Genus  Family  o  CN oo  00  i—t  o  NO  o o o  Tf  o o o  ro  o o oo  o  NO  - - •  alpinum L.  \media (L.) Vill.  Cerastium Stellaria  o  o  o O N  o o o r- o 1—1 cn o C N  o o r-  o  © O N  o o o o  in O C N  cn o  cn  o O  C N  o O  o  o o © C N V O  o  o  oo o in O o o C N O N  o oo cn m o  in oin o C N  o o  o O  o o  o  r-  cn o o o o  o o o o ©  o o o o o  O N V O  o o o o o o o o  cyathophora Murr.  Euphorbia  1—1  o  C N C N  ©  o  o in oo in  in  oo  C N  Pursh  O N  6058  C O  o  C N  ©  O N  V O  marginata  00  cn  var. geyeri  o  .—i in  1361  oo r-cn vo T  SP8  monanthogynus Michx.  cn  geyeri (Engelm.) Small  oo  Croton  o  SllO S116  o  ostryifolia Riddell  in  Acalypha  (Mackenzie) Crins acuminatus Torr. & Hook, ex Torr.  cn  inops ssp. heliophila  C N  C N  2663 1521  o  Spp. (1) Unknown brevior (Dewey) Mackenzie  o  S136 SP19  C N  o  dentata Michx.  O N  O  S118  o o  o  V O  Euphorbiaceae  o  C N  vo  Cyperus  C N  O N  V O  Carex  00  C N  cn o  Ipomoea  cn in cn o cn cn C N  Cyperaceae  in  cn  arvensis L.  C N  Convolvulus  m  Unknown  m  C N  SP35 SP6 SP13 SP22 SP62 SP94 SP85 S134  O N  Convolvulaceae  serpyllifolia L.  o  C N  Var. perfoliata  Arenaria  Caryophyllaceae  in o C N  C N  perfoliata (L.) Nieuwl.  SP47 SP26 S127  o  cn  Triodanis  arvense L.  Thlaspi  Code Spring Fall Spring Fall Spring Fall Spring Fall Total 1995 1995 1996 1996 1997 1997 1998 1998 SP10 C N  triacanthos L.  Spp. (1)  Species  Lepidium  Genus o  00  Gleditsia  JT  co  Caesalpiniaceae Campanulaceae  Brassicaceae cont.  Family  --fi  r-  00 00 C N  O N V O  1—1  V O  oo  ON  V O  ©  C N  — .1  H  t-  nuttallii (DC.) B.L. Turner dubium Sibthorp pratense L. repens L.  Unknown amplexicaule L.  azurea Michx. ex Lam.  Spp. (1) drummondii Regel trionum L. spinosa L.  Tri folium E-H"  Unknown Lamium Salvia CO  ci  o 00 cn © CN  O  o o o o in  o o CN  Tf  cn CN CN  Tf  o  rCN  o ©  m o O  as  SO as in o © CN  o CN SO as ©  o  m  -  Tf  Ti-  ro  »-H  CO  SO oo in  Os  4461  o o  r~~  o o ©  cn  oo in  o  1440  CS  2791 1670  3656  2900 5809  O  ON  oo oo  o ON  ro  o o  o so CN CN  o O ro  o O  o CN ro  Tf Tf Tf  NO  CN  ON  ©  CN  ©  o o o o o o Tf  r-  ro  o o ON  ro  o O o o  !-ON  ©  o CN  o o o Os  ON NO  o © ©  o cn  NO  o  in  o  ©  lanceolata L.  o o o o  ©  r-  o o o o  Tf  Plantago  SO 00 cn  .—1  o  co cn CN  CN CN  o  stricta L.  o o o o o o  o  r—1  Oxalis  o  o  CN .—H TIcn r- CN .—i  rO  in  Sida  o o o  1223  o o o o  in  in  Hibiscus  CN  CN  O  r—1  Allium  o  o  © ro  o  Oxalidaceae Plantinagaceae  o h  Mimosa  s c officinalis (L.) Lam.  |  Melilotus  5  lupulina L.  o  Medicago  -  SP23 SP17 SP28 SP46 S130 SP36 SP25 SP71 SP9 S125 SP65 S146 S114 S123 SPll SP50  00 ro  00  O  Lespedeza  Tf  Desmodium  NO  o o  CN  Liliaceae Malvaceae  1 r  o  SP64 SP67 S115 S122 S121  CN  Spp.(l) Spp. (2) Spp. (3) illinoense Gray sessilifolium (Torr.) Torr. & Gray stipulacea Maxim.  in  CN ro  Labiatae Lamiaceae  1  nutans Lag.  Code Spring Fall Spring Fall Spring Fall Spring Fall Total 1995 1995 1996 1996 1997 1997 1998 1998 1015 S106 o  l-H  Fabaceae  i  Euphorbiaceae cont.  0  Species  3 > t  Family co  ON  m  %  1—1  O  Os  ©  o CN ro  in  O  .—H oo  Tf  r-  Tf  m  CN  m  ro  as  Tf  CN  CN  00  o  ro  fN  O  ON  00  OO  O  O  CO  CN  in  oo  o  CN  o CN  o  o  CN  CN  o  O  o  a.  ON  o  o O ON  o  ON  CN  CN CN  in  O  o  o  o  o  o  o  ©  o  CN  ro  r-  CN  CO  CN  CO  VO  CN  r-  m  ON  O  -3-  ^r  ON  o  m  o  CN  o  o  o  o  m  o  o  ©  o  o o  o  ©  o  in  o  CN  o  o  o o o  o  o  o  o  o  o  o o ON  VO  to  CN  tN CN  CN  CN  NO  r-  CN m  m  o CN  © o in  © CN  ^r o o  o CN  o © o  23686  2058  i—i  CN  NO ON  in  ©  oo  o  O  o  o  o  o  ON  o  o  o  CN  CO CN  m oo ro CN  1  § m ro  o  1  ^ o  CQ 00  S112 SP2 SP34 SP15 SP14  ON  o  ON  S131  m  o  so  Dichanthelium oligosanthes (J.A. Schultes) Gould oleracea L. Portulaca Portulacaceae occidentalis Pursh Androsace Primulaceae circaezans Michx. Galium Rubiaceae arvensis L. Scrophulariaceae Veronica  o  o  ON  Bouteloua Bromus  o  i—i  s  Poaceae  o  *t  Polvealaceae  ©  I—1  Tridens Polygala Polygonum Rumex Alopecurus  7785 00  i—i i—i  Setaria Sporobolus  CN  2311 2498 SP18 2112 crusgalli (L.) Beauv. S117 capillare L. SP54 compressa L. S126 pratensis L. S137 Spp. (1) SP37 asper var. asper (Michx.) Kunth S133 cryptandrus (Torr.) Gray SP82 flavus (L.) A.S. Hitchc. SP45 incarnata L. SP12 aviculare L. S141 altissimus Wood SP66 brachystachus Bieb. S142 pratensis L. curtipendula (Michx.) Torr. S107 13470 20561 7773 22940 inermis Leyss.  sanguinalis (L.) Scop.  Digitaria  Poaceae continued  Code Spring Fall Spring Fall Spring Fall Spring Fall Total 1995 1995 1996 1996 1997 1997 1998 1998 SP69  in  Echinochloa Panicum Poa  Species  Genus  Family  m  r-  ro CN  O  o ©  O  o O  CN  m  NO  o o ©  o o o  o o o o  o in  m in o  CN  in  ro  CN  ON  o o  ro CN CO  O CN  CN NO NO  CO  ON  in  •1  d, oo  ON  Nees  T t  o O  in 00  in  o o  NO © ON  o  o o T t  o  ©  o o o  ON o ©  ON CM CM o  tN NO CM o o  o  o  o T t  m  CM  cn  ©  Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown  1—1  Tribulus  o  Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown  oo  o cn  cn ©  O  NO o ©  o O o o o o o  o o o i—I  ©  o o o o o  in o  © 1—1  Zygophyllaceae Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown  o  o  NO  oo CM  Spp. (2)  o  cn  T t  terrestris L.  T t  T t  1—1  tN tN tN T t  rcn  o  NO O o o  o o o o o  o T t  o o o  o o o o o  o o o o o  o o  <N T t  o o  o  o  .—I  o  o o o o o o o  -  o O o o T t i—<  o o o o  1662  cn  rafinesquii Greene  o  bicolor Pursh  ©  Viola  00  Os r - i n ON  Violaceae  r-  stricta  o  Lag. & Rodr.  o  bracteata  in  Verbena  O  cn  Spp. (1)  T t  Phyla  o  Rusby  S113 SlOl SP30 S149 SP68 SP78 SP83 SP16 SP91 S105 SP88 SP52 SP72 SP73 SP74 SP77 S108 S2 S3 S4  o  physalifolium  virginiana P. Mill.  pumila Nutt.  heterophylla  Code Spring Fall Spring Fall Spring Fall Spring Fall Total 1995 1995 1996 1996 1997 1997 1998 1998 SP76 CM  NO ON  Verbenaceae  Physalis  Solanaceae  Spp. (2)  Species 00  tN  Solanum  Genus  m CN  ON CM i n CM  o o  NO o NO  cn o  T t T t  o o O o  in  cn  1676  ON i — i CM O CM oo NO  m  <Z3  as  Family  cn  cn o o o o O o  o i—<  o o o o o o  o o o o o o o  90  APPENDIX in SEASONAL OCCURRENCE OF 1996 SPECIES Fall Species  Resilient Species  Spring Species  Unclassified  Galium circaezans  Lepidium Spp. (1)  Bromus inermis  Viola bicolor  Tragopogon dubius  Veronica Spp. (1)  Croton onanthogynus  Viola rafinesquii  Digitaria sanguinalis  Veronica arvensis  Ruellia humilis  Euphorbia Spp. (3)  Aster ericoides  Androsace occidentalis  Oxalis stricta  Euphorbia nutans  Arenaria erpyllifolia  Polygonum aviculare  Unknown  Capsella bursa-pastoris  Medicago lupulina  Acalypha stryifolia  Thlaspi arvense  Echinochloa crusgalli  Euphorbia cyathophora  Veronica Spp. (2)  Euphorbia geyeri  Hibiscus trionum  Unknown  Artemisia ludoviciana  Panicum capillare  Asclepias verticillata Convolvulus arvensis Lespedeza stipulacea Erigeron strigosu Trifolium pratense Gleditsia triacanthos Melilotus officinalis Taraxacum Spp. (1) Bromus inermis Physalis virginiana Ambrosia artemisiifolia Trifolium pratense Sporobolus asper var. asper Ambrosia artemisiifolia Plantago lanceolata Poa compressa Helianthus annuus Asclepias syriaca Convolvulus arvensis Apocynum cannabinum Tridens flavus Verbena stricta  Species classified as fall were observed exclusively or predominantly in the fall sample. Similarly, spring species were observed primarily in spring samples. Resilient species were commonly observed in both seasons. Unclassified species were observed in the fall of1996 a not observed in 1997 and could not be confirmed as fall samples. 81  APPENDIX IV SUMMARY TABLE OF REGRESSION LINE MODELS REGRESSION LINE RESULTS FOR FIT OF REGRESSION LINE MODEL SPRING/FALL  Regression Line Model = bo + biXi + D 2 X 2  Spring Fall  +  D3X1X2  X, % Open-habitat % Open-habitat  X2  1 0  X1X2  % Open-habitat 0  Spring Regression Line = 43.444 + 0.42283 Xi Fall Regression Line = 53.948 + 0.2287 X, Regression line Coeffecients Predictor Constant Designed Open-habitat Xi Dummy X2 Interaction X1X2  Coef 53.948 0.229 -10.504 0.194  Analysis of Variance Source Regression Line Residual Error Total  Df 3.000 458.000 461.000  /-test (Spring Fall Interaction) 0.194/0.037 t (obs) '(crit 0.975,458)  SE Coef 1.586 0.027 2.147 0.037  T 34.020 8.430 -4.890 5.280  P 0.000 0.000 0.000 0.000  SS MS F 51439.000 17146.000 121.011 64894.000 142.000 116333.000  5.285 1.960  S2  P 0.000  APPENDIX IV - B REGRESSION LINE RESULTS FOR FIT OF REGRESSION LINE MODEL SPRING  Regression line Model = bo + biXi + b2X2 + b Xx % O-habitat Replicate A % O-habitat Replicate B % O-habitat Replicate C  3  X + b4XiX + b XiX X,X XX x3 X 0 0 0 0 % O-habitat 0 0 1 0 % O-habitat 1 0 3  2  5  3  t  2  2  3  Regression Line A = 45.043 + 0.396Xt Regression Line B = 44.141 + 0.418Xi Regression Line C = 41.148 + 0.455Xi Regression line Coeffecients Predictor Constant % Open-habitat Dummy X2 Dummy X3 Interaction X1X2 Interaction X1X3 Analysis of Variance Source Regression line Residual Error Total Mest (X1X2 Interaction) t (obs) t (crit 0.975,246)  /-test (XiX Interaction) t (obs) 3  t (crit 0.975,246)  Coef 45.043 0.396 -3.895 -0.902 0.059 0.021  SE Coef 2.322 0.040 3.284 3.284 0.056 0.056  T 19.40 9.97 -1.19 -0.27 1.05 0.38  P 0.00 0.00 0.24 0.78 0.30 0.71  Df 5 246 251  SS 41495.10 29893.90 71389.00  MS 8299.0 121.5  F 68.29  1.050 1.960 0.38 1.96  83  p 0.00  APPENDIX IV - C REGRESSION LINE RESULTS FOR FIT OF REGRESSION LINE MODEL - FALL  Regression Line Model = bo + biXi + bzX2 + b3 + b4XiX2 + bsXiX3 Xi X2 X3 X1X2 X 1 X 3 Replicate A % O-habitat 0 0 0 0 Replicate B % O-habitat 0 1 0 % O-habitat Replicate C % O-habitat 1 0 % O-habitat 0 Regression Line A = 51.529 + 0.233Xi Regression Line B = 44.553 + 0.191Xi Regression Line C = 52.612 + 0.261Xi Regression line Coeffecients Predictor Constant % Open-habitat Xi Dummy X2 Dummy X3 Interaction X1X2 Interaction X1X3 Analysis of Variance Source Regression line Residual Error Total Mest (X1X2 Interaction) t (oh) t (crit 0.975,204)  /-test (X1X3 Interaction) t(obs) t (crit 0.975,204)  Coef 51.5290 0.2328 1.0880 6.9760 0.0286 -0.0423  SE Coef 2.711 0.046 4.142 4.142 0.0709 0.0709  T 19.01 5.02 0.26 1.68 0.40 -0.60  P 0.00 0.00 0.79 0.09 0.69 0.55  Df 5 204 209  SS 11075.00 33795.00 44870.90  MS 2215.0 165.7  F 13.37  0.400 1.960  -0.06 1-96  84  p 0.00  APPENDIX IV - D REGRESSION LINE RESULTS FOR FIT OF REGRESSION LINE MODEL - SPRING  Regression line Model = b + biXi + b X + b X + b X +b XiX + b XiX + b XiX4 X1X4 X!X XiX x X4 x x, 0 0 1 0 0 % % O-hab. 1995 Data 0 0 0 1 0 % O-hab. % 1996 Data 0 0 0 0 1 % O-hab. % 1997 Data 0 0 0 0 0 0 % O-hab. 1998 Data 0  2  2  2  3  3  4  4  5  Analysis of Variance Source Df 7 Regression line 233 Residual Error 240 Total /-test (XiX Interaction) 2  t (obs) t (crit 0.975,204)  SE Coef 2.293 0.035 2.637 2.804 3.277 0.0412 0.0451 0.0460  T -2.37 6.96 -0.11 0.31 0.32 1.00 0.55 -1.49  P 0.02 0.00 0.91 0.75 0.75 0.32 0.59 0.14  ss  MS 475.3 9.1  F 51.98  3327.11 2130.56 5457.67  1.000 1.960  /-test (XiX Interaction) 3  t(obs) t (crit 0.975,204)  /-test (X1X4 Interaction) t (obs) t (crit 0.975,204)  0.55 1.96 -1.49 1.96  85  6  3  2  3  Regression Line 1995 = -5.72+0.656Xi Regression Line 1996 = -4.54 + 0.2697Xi Regression Line 1997 = -4.38 + 0.1766Xi Regression Line 1998 = -5.42 + 0.245Xt Regression line Coeffecients Predictor Coef -5.4250 Constant 0.2450 % Open-habitat Xi -0.3010 Dummy X2 0.8790 Dummy X 3 1.0420 Dummy X 4 0.0411 Interaction X1X2 0.0247 Interaction X1X3 -0.0684 Interaction X1X4  2  7  3  APPENDIX IV - E REGRESSION LINE RESULTS FOR FIT OF REGRESSION LINE MODEL - FALL  Regression line Model = b + biXi + b X + b X + b4X +b XiX + b XiX + b XiX 0  1995 1996 1997 1998  Data Data Data Data  2  2  3  3  4  x, % O-hab.  x0  x0  % O-hab. % O-hab. % O-hab.  0 1 0  1 0 0  2  3  5  2  X4  1 0 0 0 0  6  3  x,x 0  2  2  2  Analysis of Variance Source Regression line Residual Error Total /-test (XiX Interaction) t (obs) 2  t (crit 0.975, 189)  /-test (X]X Interaction) t (obs) 3  t (crit 0.975,189)  /-test (X1X4 Interaction) t (obs)  Coef SE Coef 3.159 -9.8410 0.044 0.2640 3.784 4.5080 7.6500 3.637 -17.38 6.143 -0.0067 0.0601 0.0018 0.0552 0.0821 0.2281  Df Df 7 7 189 196  T -3.20 5.94 1.19 2.10 -2.83 -0.11 0.03 2.78  SS MS SS MS 322.4 3356.97 3356.97 322.4 9.0 1692.56 3949.53  -0.110 1.960 0.03 1.96 2.78  t (crit 0.975,189)  1.96 86  P  0.00 0.00 0.24 0.04 0.005 0.91 0.97 0.01  F 36.00  X1X3  4  X1X4  0 % 0 % O-hab. 0 % O-hab. 0 0  Regression line Equation (1995) = -5.33 + 2.63Xi Regression line Equation (1996) = -2.19 + 0.266Xi Regression line Equation (1997) = -27.22 + 0.492Xi Regression line Equation (1998) = -9.84 + 0.264Xi Regression line Coeffecients Predictor Constant % Open-habitat Xi Dummy X Dummy X3 Dummy X4 Interaction X i X Interaction X1X3 Interaction X1X4  7  p 0.00  O-hab. 0 0 0  APPENDIX IV - F REGRESSION LINE ANALYSIS: OPEN-HABITAT, OPEN-HABITAT AREA (%), SPRING Species Richness = - 14.9 + 0.345 Open-habitat (%) Predictor Constant Open-habitat  Coef -14.931 0.34468  SEE = 2.384  R = 79.7% 2  SE Coef 1.273 0.02227  T -11.73 15.48  p 0.000 0.000  R (adjusted) = 79.4% 2  Analysis of Variance Source Regression line Residual Error Total  DF 1 61 62  SS 1361.5 346.8 1708.3  MS F 1361.5 239.50 5.7  87  P  0.000  APPENDIX IV - G REGRESSION LINE ANALYSIS: OPEN-HABITAT, OPEN-HABITAT AREA (%), FALL Species Richness = - 15.7 + 0.357 Open-habitat (%) Predictor Constant O-habitat  Coef SE Coef -15.712 1.955 0.35689 0.03518  SEE = 3.317  R - 62.8% 2  T -8.04 10.15  p 0.000 0.000  R (adjusted) = 62.2% 2  Analysis of Variance Source Regression line Residual Error Total  DF 1 61 62  SS 1132.7 671.3 1804.0  MS 1132.7 11.0  F 102.93  88  P  0.000  APPENDIX IV - H REGRESSION LINE ANALYSIS: OPEN-HABITAT, OPEN-HABITAT PLOTS (%), SPRING Species Richness + 1 = 0.502 + 3.99 Log Open-habitat Plots (%) + 0.208 Percent+1 SE Coef 0.3501 0.7915 0.03052  Predictor Constant Log(10) O-hab. O-hab. + 1  Coef 0.5024 3.9859 0.20764  SEE = 1.637  R - 90.6% 2  T 1.44 5.04 6.80  P 0.156 0.000 0.000  R (adjusted) = 90.3% 2  Analysis of Variance Source Regression line Residual Error Total Source Log(10) O-hab. O-hab. +1  DF 2 60 62 DF 1 1  SS 1547.47 160.85 1708.32  MS 773.74 2.68  Seq SS 1423.41 124.06  89  F p 288.62 0.000  APPENDIX IV -1 REGRESSION LINE ANALYSIS: OPEN-HABITAT, OPEN-HABITAT PLOTS (%), FALL Species Richness + 1 = 0.314 + 2.43 Log Open-habitat Plots + 0.254 Open-habitat Plots (%) +1 Predictor Constant Log O-hab. Percent+  Coef 0.3141 2.4346 0.25447  SE Coef 0.2638 0.3680 0.04195  R = 93.3%  SEE = 1.419  2  T 1.19 6.62 6.07  P 0.238 0.000 0.000  R (adjusted) = 93.1 % 2  Analysis of Variance Source Regression line Residual Error Total Source Log O-hab. Percent+  DF 1 1  DF 2 60 62  SS 1683.19 120.81 1804.00  MS 841.60 2.01  Seq SS 1609.11 74.08  90  F 417.9c  APPENDIX IV - J REGRESSION LINE ANALYSIS: SPARSE-BROME, OPEN-HABITAT AREA (%), SPRING Species Richness = - 3.25 + 0.213 Open-habitat (%) Predictor Constant Actual  Coef -3.248 0.21317  SEE = 2.027  R = 67.5% 2  SE Coef 1.082 0.01893  T -3.00 11.26  p 0.004 0.000  R (adjusted) = 67.0% 2  Analysis of Variance Source Regression line Residual Error Total  DF 1 61 62  SS 520.75 250.52 771.27  MS 520.75 4.11  F 126.80  91  p 0.000  APPENDIX IV - K REGRESSION LINE ANALYSIS: SPARSE-BROME, OPEN-HABITAT AREA (%), FALL Richness = - 3.71 + 0.266 Open-habitat (%) Predictor Constant Actual SEE = 2.461  Coef -3.710 0.26593  SE Coef 1.451 0.02610  R = 63.0% 2  T -2.56 10.19  p 0.013 0.000  R (adjusted) = 62.4% 2  Analysis of Variance Source Regression line Residual Error Total  DF 1 61 62  SS 628.92 369.49 998.41  MS 628.92 6.06  F 103.83  p 0.000  APPENDIX IV - L REGRESSION LINE ANALYSIS: SPARSE-BROME (SB), PLOTS (%), SPRING Species Richness = - 13.3 + 5.74 log SB Plots (%) Predictor Constant Log O-hab.  Coef -13.277 5.735  SEE = 2.942  SE Coef 4.139 1.081  R = 31.6% 2  T -3.21 5.30  p 0.002 0.000  R (adjusted) = 30.4% 2  Analysis of Variance Source Regression line Residual Error Total  DF SS 1 243.42 61 527.85 62 771.27  MS 243.42 8.65  F 28.13  93  p 0.000  APPENDIX IV - M REGRESSION LINE ANALYSIS: SPARSE-BROME (SB), PLOTS (%), FALL Species Richness = - 14.1 + 6.44 log SB Plots (%) Predictor Constant Log O-hab.  Coef SECoef -14.121 4.269 6.438 1.101  SEE = 3.238  R = 35.9% 2  T p -3.31 0.002 5.85 0.000  R (adjusted) = 34.9% 2  Analysis of Variance Source Regression line Residual Error Total  DF 1 61 62  SS 358.67 639.74 998.41  MS 358.67 10.49  F 34.20  p 0.000  APPENDIX IV - N REGRESSION LINE ANALYSIS: DENSE-BROME, OPEN-HABITAT AREA (%), SPRING Species richness = - 10.2 + 0.562 Open-habitat (%) - 0.00464 Open-habitat (%)  2  Predictor Constant O-habitat O-habitat  Coef -10.241 0.5620 -0.004636  2  SEE = 2.043  R = 21.5% 2  SE Coef 4.003 0.1385 0.001153  T -2.56 4.06 -4.02  p 0.013 0.000 0.000  R (adjusted) = 18.9% 2  Analysis of Variance Source Regression line Residual Error Total  DF 2 60 62  Source O-habitat O-habitat  Seq SS 1.279 67.417  2  DF 1 1  SS 68.696 250.384 319.079  MS 34.348 4.173  95  F 8.23  P 0.001  APPENDIX IV - O REGRESSION LINE ANALYSIS: DENSE-BROME, OPEN-HABITAT AREA (%), F A L L Species Richness = 4.79 + 0.097 Open-habitat (%) - 0.00084 Open-habitat (%)  2  Predictor Constant O-habitat O-habitat  Coef 4.794 0.0965 -0.000844  2  SEE - 2.464  R = 0.3% 2  SE Coef 5.847 0.2148 0.001913  T 0.82 0.45 -0.44  p 0.415 0.655 0.661  R (adjusted) = 0.0% 2  Analysis of Variance Source Regression line • Residual Error Total Source O-habitat O-habitat  2  DF 1 1  DF 2 60 62  SS 1.237 364.192 365.429  MS 0.618 6.070  Seq SS 0.056 1.181  96  F 0.10  p 0.903  APPENDIX IV - P REGRESSION LINE ANALYSIS: DENSE-BROME (DB), DB AREA (%), SPRING 2  Species Richness = 3.90 + 0.186 DB (%) - 0.00315 DB (%) Predictor Constant DB DB  Coef 3.8954 0.18586 -0.0031484  SE Coef T p 0.9239 4.22 0.000 0.06015 3.09 0.003 0.0009035 -3.48 0.001  SEE = 2.091  R = 17.8%  R (adjusted) = 15.1%  2  2  2  Analysis of Variance Source Regression line Residual Error Total Source DB DB 2  DF 1 1  DF 2 60 62  SS 56.841 262.239 319.079  MS 28.420 4.371  Seq SS 3.766 53.075  97  F 6.50  p 0.003  APPENDIX FV - Q REGRESSION LINE ANALYSIS: DENSE-BROME (DB), DB AREA (%), F A L L Species Richness = 5.69 + 0.123 DB (%) - 0.00175 DB (%)  2  Predictor Constant DB DB  Coef SE Coef T p 5.687 1.394 4.08 0.000 0.12253 0.08884 1.38 0.173 -0.001746 0.001267 -1.38 0.173  2  SEE - 2.429  R = 3.1% 2  R (adjusted) = 0.0% 2  Analysis of Variance Source Regression line Residual Error Total  DF 2 60 62  DF 1 1  Seq SS 0.158 11.208  Source DB DB 2  SS 11.365 354.063 365.429  MS 5.683 5.901  98  F 0.96  p 0.388  APPENDIX IV - R REGRESSION LINE ANALYSIS: DENSE-BROME (DB), PLOTS (%), SPRING Species Richness = 3.88 + 0.140 DB Plots (%) - 0.00178 DB Plots (%)  2  Predictor Constant DB DB  Coef 3.8827 0.13982 -0.0017786  2  SEE = 2.091  R = 17.8% 2  SE Coef 0.9294 0.04536 0.0005117  T 4.18 3.08 -3.48  p 0.000 0.003 0.001  R (adjusted) = 15.0% 2  Analysis of Variance Source Regression line Residual Error Total Source DB DB 2  DF 1 1  DF SS 2 56.673 60 262.406 62 319.079  MS 28.337 4.373  F p 6.48 0.003  Seq SS 3.833 52.841  99  APPENDIX IV - S REGRESSION LINE ANALYSIS: DENSE-BROME (DB), PLOTS (%), F A L L Species Richness = 5.60 + 0.0959 DB Plots ( % ) - 0.00102 DB Plots (%)  2  Predictor Constant DB DB  Coef 5.600 0.09587 -0.0010232  2  SEE = 2.426  R = 3.3% 2  SE Coef 1.406 0.06695 0.0007157  T 3.98 1.43 -1.43  p 0.000 0.157 0.158  R (adjusted) = 0.1 % 2  Analysis of Variance Source Regression line Residual Error Total Source DB DB 2  DF 1 1  DF 2 60 62  SS 12.210 353.218 365.429  MS 6.105 5.887  F 1.04  Seq SS 0.177 12.033  100  P 0.361  APPENDIX IV - T REGRESSION LINE ANALYSIS: O V E R A L L , OPEN-HABITAT AREA (%), SPRING Overall Species Richness = - 4.55 + 0.270 Open-habitat (%) Predictor Constant O-habitat  Coef SE Coef -4.546 1.131 0.26970 0.01979  SEE = 2.119  R = 75.3% 2  T -4.02 13.63  p 0.000 0.000  R (adjusted) = 74.9% 2  Analysis of Variance Source Regression line Residual Error Total  DF 1 61 62  SS 833.60 273.83 1107.43  MS 833.60 4.49  F 185.70  101  P 0.000  APPENDIX IV - U REGRESSION LINE ANALYSIS: O V E R A L L , OPEN-HABITAT AREA (%), F A L L Overall Species Richness = - 2.29 + 0.268 Open-habitat (%) Predictor Constant O-habitat  Coef SECoef -2.290 1.646 0.26785 0.02962  SEE = 2.793  R = 57.3% 2  T p -1.39 0.169 9.04 0.000  R (adjusted) = 56.6% 2  Analysis of Variance Source Regression line Residual Error Total  DF SS 1 638.04 61 475.90 62 1113.94  MS 638.04 7.80  F p 81.78 0.000  102  APPENDIX V SAMPLE SPECIES SELECTION  Seasonal Duration Based on spring and fall (1996) observations, all species are classified as seasonal (spring or fall) or as seasonally resilient. Specifically, species that are observed exclusively during a particular season are classified accordingly. Species that are observed relatively evenly in the two sampling seasons are considered seasonally resilient while species that are observed in both sampling seasons, but disproportionately so, are further evaluated on 1997 observations. The list of classified species is presented in Appendix IV. Between the spring and fall samples of 1996, the number of species codes included on field cards increased from 77 to 136. Of the 59 potentially new species, 12 species were actually observed for the first time that fall. To ensure that the seasonal classification of these species are not simply an artifact of changes in sampling, a further analysis based on 1997 data is necessary. Of these 12 new species observed, seven species are observed exclusively in fall samples, two species are observed in both the spring and fall samples, while three species are not observed in the spring only.  Phenology Species are selected to represent ranges of origin, life duration, form, reproductive characteristics and photosynthetic pathway. More specifically, species are classified as native or introduced; annual, perennial or biennial; tree, herb/forb or grass; clonal; and grasses are further categorized as C3 or C4. Species information was compiled from a variety of sources. These  103  sources include Flora of the Great Plains , Wildflowers and weeds of Kansas , the 1  2  Konza on-line Information database , and the U.S. Department of Agriculture PLANTS 3  online database . 4  Rarity To include a range of abundance and distribution characteristics, species rarity was considered. The classification of rarity developed for this work was based on Rabinowitz's  5  typology of rare species. Rabinowitz's three classification criteria; Geographic Range, Habitat Specificity and Local Population Size, were all considered. According to the classification scheme, each element was further dichotomized as Common or Rare. Estimates of Geographic Range are based on observations reported in Atlas of the Flora of the Great Plains . Specifically, for each species, the percentage of Great Plains (U.S.) 6  counties with an observation is calculated. Species that are reported in less than 25% of the total number of counties are considered Rare. Local population size is based on the number of plots occupied within landscapes within the 75% and 85% open-habitat Classes. For each landscape, species that occupied less than 25% (100 plots) of all plots are classified as Rare, while species that occupy more than 25% of the plots are considered Common. Species classified as Locally Rare on all of the designated landscapes are classified as Locally Rare.  104  Habitat specificity was a qualitative classification based on information provided by a local authority , as well as habitat descriptions provided in Flora of the Great Plains and 7  1  Wildflowers and weeds of Kansas . Based on these sources none of the sampled species could 2  be considered rare. Practically, each species is classified according to regional distribution and local abundance. Four combinations are possible: regionally common - locally common, regionally common - locally rare, regionally rare - locally common and regionally rare - locally rare. Seven geographically rare while habitat common species are observed. Rabinowitz questioned whether such species even exist, considering whether their seeming absence is a consequence of ecological processes or of observational oversights. Here geographic rarity was an expression of distribution within the Great Plains region of the U.S., a range much contracted from Rabinowitz's notion of global distribution. It is reasonable to assume that the classification of these species reflect methodological deviationsfromRabinowitz's format. Brome Tolerance Brome tolerance is the propensity of species to occupy plots concurrently occupied by brome. More specifically, the total number of plots occupied by the two species is calculated respective of seasonal duration (i.e., spring and fall species are calculated according to season). Seasonally resilient species are quantified according to observations of the two seasons combined.  105  90  X  fc  X  X  X  oo X  XX X  X  X X  X  X  X  X  X  X X  X  X  X  X  X  X  X  X  X  X  X  X  X  fc CO  fO  rN fc  X  TH  i—I  fc  CO  CO  o CM  rs fc  H  tf  Q Z W  PH OH  I—4  H U W -J  fc  CO  co fc U  fc fc  CM  NO UO  NO ON O NO  ro  CN uo  CM  ro uo  NO  oo  T t  T t  UO  uo  oo SO  ro uo  X  X  X  X  X  X  uo NO uo  CO  oo  CO fc X  X X  oo  X  X  NO ON  NO  ro  UO  NO  SO  CM  X X  fc  X  X  X  CO  fn> X  xi X  fc  X  NO Tt NO  NO NO  as  ON  NO  X  X  oo  o  X  X X  CM  X  X  X  UO  T t  fc 00  CO  ro  cd  >  '1  (U  fc  oo  lu • MM  U  *> , ,'9to  =1 1  .If PH  CCJ CO PH  CD  u  a xi  x>ViVi 1-4  X  <a  ro  u  o  PH  P3  op  + H  o  o  .a  *  Q  00 K-1  1/1  o PH  o o  lu  PH  c CD  T t  o  CD  PH  PH  O  O  8  a  PQ  ro  Vi  Vi  CD  I  'o cu PH  <L> CO  '5 PH  vi  S3 1  3  CO o CO "o  PH  N  § o  c/3  las  >  u fc o  uo uo  Tt  CO  3  CN NO  uo  CO  as  uo  ro  X  CO  voo  TT"  <D CJ  Ofi  2 o •S 2  2H  CU  £ O tf  P  PQ  O  ro  \x\  PH  CZ)  CM  X  •*  PH  X  X  X  X  C N  X  C M  X  rn T-H  O N  V O  V O  O N  in  m  CZ)  X  X  PH  X  X  V O  X  oo  O N  X  X X  PH  X  X  X  m  cz) in l-H PH  X  X X  X  X  in  o  oo rn  m vo  V O  PH  X  X  X  X  X X  X  <3 ^ cu  a I  R O  S  vo  CZ)  CM  «  © .ts C -©  in  wo o ro  S  V O  X  CH  C*>  ^  CZ)  CM  X  PH  X  X  vo  X  in od  C N V O  cz)  ro  X X  PH  X  X X  X  X  X X  X X  C N O N  in  CZ!  oo  X  i—i  PH  X  oo  vo  cz)  o\ so  X  X  PH  cz)  X X  PH  X X  X X  X  X  X  X  X  O N  X  V O  vd m  X  O N  ^  i  VO  rO N  ^I sf © -i^  O N  X  3§  V O V O  s S  cz) IT,  CM TS  cu  s  X  X  PH  cz)  vo  in  o  a  e o  Id  T5  CD O  w  > '1  CU  CD OH  U cd  o JU  T3  cz)  s  CO  .2.  CO  v  3  PL,  a  cz)  PH  Q  B  Cd  PQ co • SH  to w  —i  o PH  _C  cd  PH PH  ' CD o O , O O  O  &  St cu  co CO CO  13  a  cu  'o CD  I  PH O O  •* 3 cu a CO  CZ)  cu  co Q  CU  -3 s u  I  CL  CO  •ts o-" •° ^  O  >o cu  VH  c  5  O  s o  K ca "=> -t: © ^  cd  VH  © i  cd  H-»  CD  "5  8s  <u p  VH  <D O cd  -t->  o  CO  cd <U  00  cd ^3  JH  S  B g S o  PQ  H  i  s  cu" CJ C3  © ©  CJ  o  APPENDIX V n K - S TEST FOR OPEN-HABITAT AREA Percent Available Open- labitat 55% 65% 75% 85% >  *  ^ -e  *- O  85% 75% 65% 55% 45% 35%  1.00  0.018 0.075 0.012 0.054 1.00  1.00 1.00  0.998 0.988 0.894 0.54  0.998 1.00 0.187  45%  35%  1.00  0.956  0.398  1.00  0.998  0.493  0.998  0.988  0.398  0.988  0.597  0.988 0.038  0.894  0.14  The p-values for the K-S Test at each combination of percent available openhabitat class for 1996 data. Values at the top of the matrix (italicized) represent spring samples, bottom values represent fall samples.  108  APPENDIX VIII  CLASSIFICATION OF SPATIAL SIGNATURES  ,,-•„  Species Code  -•'•''/ Established  Insufficient Colonizing data:  U ; Androsace occidentalis  SB  X X X  Medicago lupulina. X  Echinochloa crusgalli  X  Euphorbia geyeri var. geyeri Artemisia ludoviciana  X X  Asclepias verticillata Convolvulus arvensis  X X  Lespedeza stipulacea  X  Trifolium repens Gleditsia triacanthos  X  Melilotus officinalis  X  Physalis virginiana  X  Ambrosia psilostachya  X X X  Sporobolus asper var. asper Ambrosia artemisiifolia Capsella bursa-pastoris  DA  X  Polygonum aviculare  Arenaria serpyllifolia  DB  X  Oxalis stricta Veronica arvensis  SA  s  X X  Digitaria sanguinalis  X  Euphorbia nutans  X  Acalypha ostryifolia  X  Euphorbia cyathophora TOTAL  X 4  7  5  2  2  1  2  0  1  (F) Fragmented; (U) Fragmentation Unapparent; (S) Similar; (SA) Similar Above; (SB) Similar Below; (DA) Dissimilar Above; (DB) Dissimilar Below.  109  APPENDIX VII! - B  SPECIES RESILIENCE (MHC) .•}"•• Species Code  Lowest Openhabitat Class Occupied  Established  Colonizing  u  u -, .-.:S';; SA  Androsace occidentalis Oxalis stricta Polygonum aviculare Veronica arvensis Medicago lupulina. Echinochloa crusgalli Euphorbia geyeri var. geyeri Artemisia ludoviciana Asclepias verticillata Convolvulus arvensis Lespedeza stipulacea Trifolium repens Gleditsia triacanthos Melilotus officinalis Physalis virginiana Ambrosia psilostachya Arenaria serpyllifolia Sporobolus asper var. asper Ambrosia artemisiifolia Capsella bursa-pastoris Digitaria sanguinalis Euphorbia nutans Acalypha ostryifolia Euphorbia cyathophora  75 35 45 65 35 35 35 35 35 35 35 35 45 35 35 35 85 35 35 85 35 35 35 35  SB  DB  DA  35 45 35 35 35 35 35 35 35 35 45 35 35 35 35 35 35 35 35 35  MHC, (lowest open-habitat class reporting species observation) partitioned by signature category. (F) Fragmented; (U) Fragmentation Unapparent; (S) Similar; (SA) Similar Above; (SB) Similar Below; (DA) Dissimilar Above; (DB) Dissimilar Below.  110  APPENDIX VIII - C SPECIES RESILIENCE (MAC) Established Lowest Open- Colonizing habitat Class with >25%of F .. F u. Maximum S SA SB DB DA Abundance Androsace occidentalis 75 Oxalis stricta 75 75 Polygonum aviculare 75 75 Veronica arvensis 65 Medicago lupulina. 65 65 Echinochloa crusgalli 75 75 Euphorbia geyeri 75 75 Species Code  ;  var. geyeri Artemisia ludoviciana Asclepias verticillata Convolvulus arvensis Lespedeza stipulacea Trifolium repens Gleditsia triacanthos Melilotus officinalis Physalis virginiana Ambrosia psilostachya Arenaria serpyllifolia Sporobolus asper var. asper Ambrosia artemisiifolia Capsella bursa-pastoris Digitaria sanguinalis Euphorbia nutans Acalypha ostryifolia Euphorbia cyathophora  U  35  35 45 75 55 75 75 75 65 35 85 35  45 75 55 75 75 75 65 35 35 65  65 85 55 55 35 35  55 55 35 35  MAC, (lowest open-habitat class with at least 25% of maximum number of species observation, partitioned by signature category. (F) Fragmented; (U) Fragmentation Unapparent; (S) Similar; (SA) Similar Above; (SB) Similar Below; (DA) Dissimilar Above; (DB) Dissimilar Below.  I l l  APPENDIX VIII - D SPARSE BROME TOLERANCE Sparsebrome  Species Code  Established  Colonizing F  F SA SB DB DA  U S  1 33 7 18 325 1289 1197  Androsace occidentalis Oxalis stricta Polygonum aviculare Veronica arvensis Medicago lupulina. Echinochloa crusgalli Euphorbia geyeri var. geyeri Artemisia ludoviciana Asclepias verticillata Convolvulus arvensis Lespedeza stipulacea Trifolium repens Gleditsia triacanthos Melilotus officinalis Physalis virginiana Ambrosia psilostachya Arenaria serpyllifolia Sporobolus asper var. asper Ambrosia artemisiifolia Capsella bursa-pastoris Digitaria sanguinalis Euphorbia nutans Acalypha ostryifolia Euphorbia cyathophora Frequency  of occurrence  (U) Fragmentation Below;  33 7 325 1289 119 7 160 141 19 522 31 31 75 55 160 142 61 105 244 91 237  of species within sparse-brome  Unapparent;  (DA) Dissimilar  160 141 19 522 131 31 75 55 160 2 142 61 0 105 244 91 237  Above;  U  (S) Similar;  (SA) Similar  (DB) Dissimilar  Ill  Below.  plots.  (F)  Above;  Fragmented;  (SB)  Similar  APPENDIX Vin - E DENSE BROME TOLERANCE Species Code  Densebrome  F Androsace occidentalis Oxalis stricta Polygonum aviculare Veronica arvensis Medicago lupulina. Echinochloa crusgalli Euphorbia geyeri var. geyeri Artemisia ludoviciana Asclepias verticillata Convolvulus arvensis Lespedeza stipulacea Trifolium repens Gleditsia triacanthos Melilotus officinalis Physalis virginiana Ambrosia psilostachya Arenaria serpyllifolia Sporobolus asper var. asper Ambrosia artemisiifolia Capsella bursa-pastoris Digitaria sanguinalis Euphorbia nutans Acalypha ostryifolia Euphorbia cyathophora  0 12 3 2 118 284 337 92 94 8 155 35 5 26 34 54 0 94 19 0 13 92 85 207  Established  Colonizing U  F S SA SB DB DA  U  12 3 118 284 337 92 94 8 155 35 5 26 34 54 19 19 13 92 85 207  Frequency of occurrence of species with dense-brome plots. (F) Fragmented; (U) Fragmentation Unapparent; (S) Similar; (SA) Similar Above; (SB) Similar Below; (DA) Dissimilar Above; (DB) Dissimilar Below.  113  O  CU  w B  es  co  o o  00 o CN  l>  NO  UO I o" OS CO Os Os Os o  Tf  CO  ON  ft <u •.ft  CO  Tt  CN UO  loo  uo NO  INO °0  oo  o  ON CO UO  uo o  oo  ©  Tt  CO  CN  ON CO CN  oo T t  CO  ©  © NO uo uo uo  CO CO  ©  CO  Tt  CO  CN  NO  oo NO T t  NO CO  uo ©  CO  NO CN CN  00  00  co|  CO  uo  oo r-  NO  uo  «©  CO  .jo  a or S  Tf  00 CM  CN UO  © ©  r--  ca  ©  ©  CO  CO  3  CO  NO UO  NO UO  «©  Tf*  co ©  I 00 CO CN d l r-- CO  I  col  t>  CO  Tf  CO  00 uo  f-H  00  Os  o  ICS NO o  00 T t Os r-' ON uo CO NO CO 00 ON 00  uo C N , CN o uo uo Os\ 00 © NO oo UO © CO 0 O uo uo  i ON  S  uo  r--  CO  Tf  UO CM  C  CU  uo  uo uo uo 00  loo  cu  c «s I.  ON  1—I  CN CN CO  as  o - <3  <3  a,  NO  NO  CU  a  UO CN  i  ON NO  CU  UO  I  NO ON  E *.  © uo  ON  I CO  CO  ©I  ON  00 00 CN o ©  UO  CO CN  sf ©  ON  00  ON  S  CN  .© •  sf ^ © o-  ft CO  <u  T t  o © ©  T t  UO  uo  I ON I  © uo  NO  ON NO CN  od|  Tt  ON CN  |NO  SO  ©J  © UO  CO CN Tt |  Tt  CN NO  uo  O CO CO  CO  UO  <  O CO  fc © I © ©  ©I © uo I Os uo oo rNO  o o © ©  CM  o uo © o  CM NO  UO  co|  uo| 1—I  ON  00 uo :  © ©  <  ON  00 NO  f c  00  I© ©J co  uo uo UO uo' uo uo NO uo  Tt  NO  S £  o  8  <=>  ^H  u S o o o  »©  ca  ii  o  Si  fc  '1 (H  13  s O  co  Vi  CD  s © S S  X  o  P-i  tf u X u o o  *3  a  " 5b.If _©  , pco VH  "© ft  .3 .SP  Q o  '(H  o  PH  O  lu OH  Vi  1 o Vi  a  co  CD  o  m  M U  I a_ i  .o u 8 © ca  o  CD  Vi tH  "a  NJ  ^ 8" r' U  & Qsi  Vi  OH  sf ©  S  O  -t->  ©  O ©  5  <3 ^ ©0  APPENDIX IX - B STATISTICAL COMPARISON OF BIOLOGICAL TRAITS  Comparison Spring-Fall Spring-Spring and Fall Fall-Spring and Fall Native-Introduced Annual-Perennial Annual-Biennial Perennial-Biennial Tree-Herb/Forb Tree-Grass Herb/Forb-Grass Clonal-Non-Clonal GCHCLC-GCHCLR GCHCLC-GRHCLC GCHCLC-GRHCLR GCHCLR-GRHCLC GCHCLR-GRHCLR GRHCLC-GRHCLR GCHCLC  Resilience T(obs) P 0.004 4.33 1.76 0.096 2.02 0.058 2.79 0.01 0.394 0.87 0.234 1.25 1.63 0.128 0.71 0.486 0.87 0.478 0.494 0.69 1.25 0.226 0.56 0.58 65535 NA 0.318 1.11 0.514 0.614 1.12 0  - Geographically-Common;  Geographically-Common; Habitat-Common;  0.276 1  Habitat-Common;  Habitat-Common;  Locally-Common;  Sparse Tolerance T(obs) P 0.022 3.309 1.117 0.28 0.624 0.5 0.724 0.357 1.087 0.29 0.664 0.446 0.644 0.475 0.602 0.531 0.584 0.619 1.56 0.134 0.902 0.125 19.962 3.8E-11 NA 65535 2.234 0.066 7.297 0.00006 2.021 0.058 0.498 0.636  GRHCLR  Locally-Rare;  115  T(obs) 1.084 0.818 0.253 0.529 1.5 0.683 0.57 0.84 0.78 0.835 0.379 6.833 65535 1.141 3.255 2.98 0.118  Locally-Common; GRHCLC  - Geographically-Rare;  Locally-Rare.  Dense Tolerance P 0.358 0.426 0.804 0.602 0.152 0.512 0.58 0.414 0.516 0.414 0.708 1.8E-05 NA 0.306 0.006 0.008 0.91 GCHCLR  - Geographically  Rare;  Habitat-Common;  APPENDIX X  SPATIAL SIGNATURE OF SPECIES  Patchiness of Oxalis stricta Spring/Fall 1996 30 25  f  20  O Habitat • Habitat A Habitat X Habitat ac Habitat O Habitat  t3 15 <L>  I m  5 0  25  50  Class 35 Class 45 Class 55 Class 65 Class 75 Class 85 100  75  Percent of Total Area in Patches  30  I  Patchiness of Androsace occidentalis Spring 1996  25  I  20  o  15  o ° A x * o  OH  10 5  Habitat Habitat Habitat Habitat Habitat Habitat  Class 35 Class 45 Class 55 Class 65 Class 75 Class 85  0 25 50 75 Percent of Total Area in Patches  116  100  APPENDIX X SPATIAL SIGNATURE OF SPECIES  Patchiness of Polygonum  aviculare  Spring/Fall 1996 30  i  25  09 o  £  « Habitat Class 35  20  D Habitat Class 45 15  A Habitat Class 55 x Habitat Class 65  10  * Habitat Class 75 o Habitat Class 85  5 H  25  50  75  100  Percent of Total Area in Patches  tches  Patchiness of Veronica arvensis Spring 1996 30 -| 25 -  Habitat Class 35 Habitat Class 45 Habitat Class 55 Habitat Class 65 Habitat Class 75 Habitat Class 85  20 o 15 -  PH  u <U -O  10 50 0  25  50  75  Percent of Total Area in Patches  117  100  APPENDIX X SPATIAL SIGNATURE OF SPECIES  Patchiness o f Echinochloa crusgalli S p r i n g / F a l l 1996 30 24  CJ  o  o ° * * * °  18  PH  12 6  CO •  AA  xo  20  40  Habitat Habitat Habitat Habitat Habitat Habitat  60  Class Class Class Class Class Class 80  35 45 55 65 75 85 100  Percent o f T o t a l A r e a i n Patches  Patchiness of Medicago lupulina Spring/Fall 1996  30 25 B  • Habitat • Habitat A Habitat X Habitat XHab itat o Habitat  20  cd PH  o  15  1  10  (L>  x  Class Class Class Class Class Class  35 45 55 65 75 85  XBt 25  50  75  Percent of Total Area in Patches  118  100  APPENDIX X SPATIAL SIGNATURE OF SPECIES  Patchiness of Echinochloa crusgalli Spring/Fall 1996 30 24 « ° * * * °  18  PH  12 6  Habitat Habitat Habitat Habitat Habitat Habitat  Class Class Class Class Class Class  35 45 55 65 75 85  0 20  40  60  80  100  Percent of Total Area in Patches  Patchiness of Medicago  lupulina  Spring/Fall 1996 .  30 25 CA  • D A x X o  B 20 PH  15 A  1  10 x  Habitat Habitat Habitat Habitat Habitat Habitat  Class Class Class Class Class Class  35 45 55 65 75 85  XH  25  50  75  Percent of Total Area in Patches  119  100  APPENDIX X SPATIAL SIGNATURE OF SPECIES  Patchiness of Euphorbia Spring/Fall 1996  geyeri  x  J3  3  Habitat • Habitat A Habitat X Habitat X Habitat o Habitat o  PN CM  O ti  o  x  25  50  Class 35 Class 45 Class 55 Class 65 Class 75 Class 85  1  100  Percent of Total A r e a in Patches 75  Patchiness of Artemisia ludoviciana Spring/Fall 1996  30  CO  I PH  25 20  o Habitat Class 35  A  • Habitat Class 45  .  15  A Habitat Class 55 x Habitat Class 65  10 A  ac Habitat Class 75 o Habitat Class 85  25 50 75 Percent T o t a l A r e a in Patches  120  100  APPENDIX X SPATIAL SIGNATURE OF SPECIES  Patchiness of Asclepias verticillata Spring/Fall 1996  :rof:Patches  30  H |  i  25 -  •  o  • Habitat Class 35  20 15 10 -  nl  • Habitat Class 45 o  L  0  A Habitat Class 55  •  x Habitat Class 65 x Habitat Class 75 o Habitat Class 85 i  25 50 75 Percent of Total Area in Patches  100  Patchiness of Convolvulus arvensis Spring/Fall 1996  O Habitat Class 35 • Habitat Class 45 A Habitat Class 55 X Habitat Class 65 X Habitat Class 75 o Habitat Class 85 —i  25  1  1  1  50  75  100  Percent of Total Area in Patches  121  APPENDIX X SPATIAL SIGNATURE OF SPECIES  Patchiness of Lespedeza stipulacea Spring/Fall 1996  30 25 CO  CM  • • A x X o  20 15 10  x  f°  X  5  Habitat Habitat Habitat Habitat Habitat Habitat  Class Class Class Class Class Class  35 45 55 65 75 85  X  0  25  50  100  75  Percent o f Total Area in Patches  30  Patchiness of Trifolium repens Spring/Fall 1996  25 u o 20  • o A X X o  ts  OH  o 15 A S3 10  25  50  Habitat Habitat Habitat Habitat Habitat Habitat  Class Class Class Class Class Class  75  Percent of Total Area in Patches  122  35 45 55 65 75 85  100  APPENDIX X  SPATIAL SIGNATURE OF SPECIES  Patchiness of Gleditsia triacanthos Spring/Fall 1996 30 Vi  1 u  H  25  • Habitat Class 45  20  A Habitat Class 55  15  x Habitat Class 65 10 5 0  i  x Habitat Class 75 o Habitat Class 85  100  25 50 75 Percent of Total Area in Patches  Patchiness of Melilotus officinalis Spring/Fall 1996  30 25  %  20 -|  —  15 10 5  -I  o Habitat • Habitat A Habitat X Habitat X Habitat o Habitat  Class 35 Class 45 Class 55 Class 65 Class 75 Class 85  0 0  25 50 75 Percent of Total Area in Patches  123  100  APPENDIX X SPATIAL SIGNATURE OF SPECIES  Patchiness of Physalis virginiana Spring/Fall 1996 30 25  CD  <• Habitat Class 35 • Habitat Class 45  f £03 20 A  A Habitat Class 55  PH  15 J  x Habitat Class 65  10  X Habitat Class 75 o Habitat Class 85  25 50 75 Percent of Total Area in Patches  100  Patchiness of Ambrosia psilostachya Spring/Fall 1996  30 25  "I  20  • Habitat Class 35  cd  PH  ^ H  o t*  ,  • Habitat Class 45  -  15  A Habitat Class 55  u XI  10  n  X Habitat Class 65  1  x Habitat Class 75 o Habitat Class 85 20  40 60 80 Percent of Total Area in Patches  124  100  APPENDIX X  SPATIAL SIGNATURE OF SPECIES  Patchiness o f Arenaria serpyllifolia S p r i n g 1996  30 25 <L>  • Habitat Class 35  S 20 a  PH U-t  ° Habitat Class 45  15  o  A  Habitat Class 55  * Habitat Class 65  10  * Habitat Class 75 • Habitat Class 85  5 0 25  50  75  100  Percent o f T o t a l A r e a i n Patches  Patchiness o f Sporobolus asper var. asper Spring 1996  30 -j m  25 1  o Habitat Class 35  * ° 15 I  o Habitat Class 45  £  2  0  Js  A  A Habitat Class 55  10 H  X Habitat Class 65  A*  X Habitat Class 75 o Habitat Class 85 25 50 75 Percent o f Total A r e a in Patches  125  100  APPENDIX X SPATIAL SIGNATURE OF SPECIES  Patchiness of Ambrosia artemisiifolia Spring/Fall 1996 30  o Habitat Class 35  25  • Habitat Class 45  20  A Habitat Class 55  O  15  X Habitat Class 65  1  io  3 OH  «4H  b  X Habitat Class 75  4  t  o Habitat Class 85  25  50  75  100  Percent of Total Area in Patches  Patchiness of Capsella bursa-pastoris Spring 1996  30 en  25  Cu>  o  20  o Habitat Class 35  A  a Habitat Class 45  PH <HH  O Ii  15 A  A Habitat Class 55  CD  X}  x Habitat Class 65  10  * Habitat Class 75 o Habitat Class 85  25  50  75  Percent of Total Area in Patches  126  100  APPENDIX X  SPATIAL SIGNATURE OF SPECIES  Patchiness of Digitaria sanguinali s Fall 1996  30 25 20  o Habitat Class 35  PH  O  15  X  X  • Habitat Class 45  „x  A Habitat Class 55  10  x Habitat Class 65 X Habitat Class 75 o Habitat Class 85 25  50  75  100  Percent of Total Area in Patches  Patchiness of Euphorbia nutans  Fall 1996 30 to  s  I  -  o Habitat Class 35  25  o Habitat Class 45  20  A Habitat Class 55  PH  o  x Habitat Class 65  15  * Habitat Class 75  10  o Habitat Class 85  5 0  25  50  75  Percent of Total Area in Patches  127  100  APPENDIX X SPATIAL SIGNATURE OF SPECIES  Patchiness of Acalypha Fall 1996  ostryifolia  30 25  • Habitat Class 35  2CS 20 Pw o f-l  15  u  • Habitat Class 45 A Habitat Class 55  •  X Habitat Class 65  10  X Habitat Class 75 o Habitat Class 85  25  50 Total Patch Area  100  75  Patchiness of Euphorbia cyathophora Fall 1996 • Habitat Class 35 • Habitat Class 45 A Habitat Class 55 x Habitat Class 65 x Habitat Class 75 o Habitat Class 85 — i  25  1  1  50  75  Percent of Total Area in Patches  128  100  PH - H  PM oo  c« - H  <rt  CO  fc O Z  <  U HH fc HH  CO CO  u  u  u  u  o  c/l - H  PJ  PJ  PJ  feci  PJ  PJ  w  u  u  PM C\ tn - H  o  PJ  PJ  UJ  u  u  ^H  i-H  y  UJ  PJ  PJ  u  a a-  u UJ  u  o  u  o  u  s; o  u  u  u  u  o  a  HH CO  CM VO  CJ  H  CM O VI <*»  <S  u  o  u u  u  CO  u  o  u  u  u  O  u  is.  sf o o £ £  PJ  U  <a  ca  w  u  s; o  5?  o o  PJ  u  st I  o u  UJ  UH  *H  C/5  a ft  o  UJ  <a  o <3  w  u  u  -SS  O  CM CS VI <N  <  UJ  CM t -  fc tf  fc  PJ  o  U  Z O  PJ  o  < H  PJ  ci  ^H  PJ  w  PM 90 Vi rs  c/l «§  PJ  UJ  VI CS  Fall Spring and Fall Origin Native  PH PH  H  u  UJ  u  Spring  Z W  HH  u  PJ  u u  Seasonality  HH Q  o  u  PJ  Un-  Species Code  X X  u  -s* co  -SS  s  -S3  C3N  ca «©  ^a > O  3  I  s IT  u u u  u  u u  sf S! O  m U  C3  Tf  u  GRHCLC GRHCLR  PH  o u  GCHCLR  <  u  GCHCLC  -  o  Rabinowitz's Topology of Rarity  PM  w  CJ  Clonal  -J fc  U  PJ  PJ  Photo. Pathway  CJ\  UJ  PJ  Grass  PM  UJ  PJ  u  UH  PJ  u u  PJ  Tree  PH CO  PJ  u  BJ  Herb/Forb  w  CM r~ Vi  u  Introduced  HH  U  ^H  Annual  w  PJ  PJ  Life Duration  CO  PJ  u  PH <•>«  VI  UJ  u u  O  PH  w  o  Form  CN  t/i  u  Biennial  PH  u o  Perennial  ei  Plant Characteristics  VI  o ^ S g ca  i  <S  -ss r~  PL - H  PH  PH  eu  PH  PH  PH  PH  PH  PH  PH  PH  P  P  p  P  P  P  PH  PH  PH  cu  P  I  <s  Vi  PH  PH  P -  -R  PL, - H Vi <S  Pk  PH  PH  P-  PH  e  -1 o  PL <S Vi ^ H  PH  PL t 1/5  r>  P  PH  oo  PH  PH  PH  P PH  p  P  P  P  PH  PH  P H  P-  Co  If ^ cu  R  vj  R  o  PH  R  c\  PH  &o vo  PH  P H  PL - H  PH  P H V5 SO f S  PH  PH  PL oo  P  P  P  PL I--  PH  PH  PH  PH  PH  GO «N PH  PH  P -  PH  P  P  PH  I  PH  PH  PH  P  o  PH  PH  cu G  y PH  PH  P.  PH  PH  PH  PH  PH  P -  PH  P  P  P  P  P  P  P  P  P  P  P  P  Form  Grass  Perennial  Plant Characteristics  Introduced  Annual  Life Duration  Origin  Spring and Fall  Fall  Seasonality  Spring  Native  . —:—a—_  Species Code  cu  J3  S  IS  cu  -R  9  •S3 cu  — A  O  13 R « R O  -R  1  •E3 Q R O  la sf ©  U  R  £  o  P  P  P  cu  o eq  cc GRHCLR  P  GRHCLC  PL O  if  I a e o  O KI  PH  PH  GCHCLR)  PH  GCHCLC  PH  Rabinowitz's Topology of Rarity  PH  2  C4  PL vo Vi <s  a  a  PH  PH  PH  PH  PH  Photosynthetic Pathway  PH  Clonal  PH <S Vi CM  o  KI 13 R CS  PH  PH  PH  Herb/Forb  1—1  : 1  PH  PH  Tree  e  PL o\ Vi *  • '  PH  sf  PL  c/5  P-  P -  Biennial  PL'  R ©  C3  5b O  Cj  O  cu K )  a  R 53 R O  £'  5s  v^ CJ  5o  s cu s > O -©  -2 S  cu CS  CO  fo  t—*•  "5 co s:  sf o  I s: o s;  Of  y  I o o  I-H "CS  s: 0-5 s: cu s; sf o o  73  Vi  PH  - H  Vi f S  PH  Vi  PH  oo  Vi  Vi  Vi  Vi  - H  Vi  Vi  Vi  Q  Q  Q  Q  r-  Q  Q Q Q  Q  fi  Vi  Vi  Vi  Vi  Vi  Vi  ^H  *1  Vi  Vi  Vi  Vi  aa  Vi  Vi  Vi  "Ti  Vi «H  PH  Vi  Vi  Vi  SP19  Vi  Vi  13  o o  Vi  e cs s: o  Q Q  Vt PH  ^i yi  Vi  a  Vi  9 o  CS  cs cs -s:  I  cs o cs o u  GRHCLR  GRHCLC  GCHCLR  GCHCLC  s? k s -"*=  to  ^  So o o o  ?> o  "CS K CS  cs  I  Rabinowitz's Topology of Rarity  u  Clonal Photosynthetic Pathway  Grass  Herb/Forb  Tree Form  Perennial  Biennial  Plant Characteristics  Annual  Life Duration  Origin  Fall Spring and Fall Seasonality  — • »  Species Code  •  Spring  Native Introduced  ,  sf U  g  CO  O U e< ^£  to  -S*  ^  1?  s f Co o  S S  a3 cj  I  * 53  -52  cj  cu "O  I  CS cj o •o K I is" —* "CS ca K 0=1 cs cj s: sf o o  CU  5 £  3 O  a  cfl  <5  U  I K  13  -I  CS o o K ] cu" 13 iL. K CS CS  cu CO  Tr  CU  S  s ©  CS  CO  60  a  >-l  e CS  cu  > O  CS cj  J3  c^  Co O  u ts  so  sf o  c« CS CU  U  I-  3 •** cs  CQ  A  u o o  C8  to  tL,  is  I SO  O [CD O  G  •5b 00  %•» CS  CM CU CO •V  Q  "is o c5 CL,  'S  .£ "3 -p  a  cd o 3 Ct, H Cc,  to  C J  CS Cj  SL.  CS  >W  co cu  cj  O O cu K I CO 13 I s: cs  cn cu  PH»J  U Co  c3 00  to  o U  I  §  > O  SL.  CS  3 B -l •i s CO  CJ  fie] S£  f5  3> U  00  2  iB  C/3  CQ  Q  PH  UH U4  2  CD Vi UH UH  I  UH  cu  1  czi UH W  3) U  Vi ©  s  ©  U  3)  U  CA  CN  u-  UH  u  u  PH  Vi  tn  .22 '  '3 v  "3  .a  cc  .£  UH  "a'  S  u  es  CO  c  T PH  Vi  CS .  o  p u  13  s  U  "so es h< 9 -*•» es  •a  i-e  CJ  '1  a  l l PH  .Sf • co  u  CQ  CO  is es a  c  •5b  co CU  "O O  CL, a} CL Vi | b |CZ>  Q (19  o UH  -3 U a  Vi  I  <U CU  a  CO  cd c o  3  S * "S £ OH OH  P5 3 II & S  „  CO  1 S3  E u  H O  OH  i i PH  P H  u  PH  oo  u  P J  O PO U  <  CS PH  00 PH  00  sf -st o st  u  85  53  st o  P J  P J  o  ca  Pu pq  UL.  GO  -  CD  Cj  00  0*3 PH  o  aI ca  ca PH  00  3 t o  PH  H, ^ £ ca <0 st g 5  00  TS CU  3 B'  Pu  Pu  U  PH  00  U  o rU 00  D  P-l  PH  C« •  '3  oo  00  Pu W  PH  cu  CL  P J  00  p. P J  CO  "3. s  P u U  Pu  U  PH  o  es CO C M  O  c o • M  Pu  PH  U  U  PH  00  •*#  CQ Q  es ' •<AS u PH  CQ  Q  Pu PJ  Pu W  00  CS  ca Q  Pu UJ  O  cu u 3 es fi  ,  3  T3  <0  O  C3  .Sf to  tH  g  Cj  a  cu  PH  T3  es  a  on  h-1 cu  •X • hs o TS  fi  C  c  to  00  '5b  3  a  o  loo^  PU  a  to  cu  ca i O vJ  I  ~ o  CA CU  I §>  cu wcu  I  §> i I  o £ 3 cS eu eu  S .2 cu U  CJ  H  ^ St  »5  <3  s >>> - ^ Cc  -  ca  co Cj  ^  o" & H 0$  —.  "a a o  c  39  « s 2 O .5  s S  —s  O •§  ^  O -~ CO  1  CO  o  •a pa 2 c/5  Q -J H H  &  HI  a  eg  Q •"i  -J  u  ca CJ  o  o -J o  O o5  •a  I-H  P5  s: cu  Q  a  Z  X X H H  Q  zfc fc fc <  «: z o H H  H  CJ O s: o  "5 « a -g S §8 g S *  £ S  fc PS  H  5 s:  o  5J  DJD  co co «< HJ U  SHH  CJ  O »-H  "a st a s: o  I  •9 •H s S 2  a  s:  so  o  CJ  Z  a  sa'-xs  H H  CO  <  Sf O  H «< fc co  §"•1  S cf S  O  H H  C3  -S -a  o  CJ  ca CH  O "8  •H C j bo -sr ^ .s:  c  '5b  'f •  J  o  o  Q  tin  88  a  'i  s  OH N  ca  co  „  ca  ca*  j£  S u  It C3  toe Cj  CJ CJ  I C J |  CJ -S  O  go"  OH  50  O  H  cs a CD o CJ a a  3  APPENDIX XIII  LANDSCAPE MAPS OF SPECIES LOCATIONS  Landscape maps of species in six landscapes. Each cell represents a plot, dense-brome (dark), sparse-brome (light) and open-habitat (white). Medicaeo luvulina (U-Species 17), Lespedeza stivulacea (S-Species 23) and Sporobolus asper var. asper (D-Species 37).  136  APPENDIX XIV SUMMARY OF RESILIENCE Brome Tolerance Spatial Signature Classification Colonizing  F U  Established F U  s SA SB DB  Sparse-brome Mean SD 85.57 88.847 123.60 74.691 675.00 868.327 669.00 746.705 522.00 NA 178.00 207.889 142.00 NA  Dense-brome Mean SD 26.71 33.654 88.80 72.358 151.50 187.383 215.50 171.827 155.00 NA 76.50 58.690 NA 19.00  Resilience MHC SE Mean 37.86 4.88 NA 35.00 NA 35.00 NA 35.00 NA 35.00 NA 35.00 NA 35.00  MAC SD Mean 63.57 15.736 49.00 19.494 70.00 7.071 60.00 21.213 55.00 NA 70.00 7.071 NA 35.00  (C) Colonizing; (E) Established; (F) Fragmented; (U) Fragmentation Unapparent; (S) Similar; (SA) Similar Above; (SB) Similar Below; (DA) Dissimilar Above; (DB) Dissimilar Below.  137  APPENDIX XIV - B SUMMARY OF RESILD2NCE  Comparison F-U F-U S-SA S-SB SA-SB DA-DB  Colonizing  Established  (C) Colonizing; Similar; Dissimilar  Resilience Brome Tolerance MAC MHC Sparse- jrome Dense- jrome T(obs) T(obs) T(obs) T(obs) P P P P 1.435 0.182 0.778 0.454 2.013 0.072 1.291 0.226 NA 2.43 0.592 65534 0.354 1.006 0.54 0.649 0.632 0.333 NA 0.007 0.994 0.356 0.756 65535 1.732 0.879 NA 65536 0.99 0.144 0.909 0.015 0.192 0.051 NA 65537 0.82 0.161 0.898 0.287 NA NA NA NA NA NA NA NA  (E) Established;  (SA) Similar  Above;  (F) Fragmented;  (SB) Similar  Below;  (U) Fragmentation (DA) Dissimilar  Below.  138  Unapparent; Above;  (DB)  (S)  APPENDIX XIV - C COMPARISON OF BIOLOGICAL TRAITS  Comparison Spring-Fall Spring-Spring and Fall Fall-Spring and Fall Native-Introduced Annual-Perennial Annual-Biennial Perennial-Biennial Tree-Herb/Forb Tree-Grass Herb/Forb-Grass Clonal-Non-Clonal GCHCLC-GCHCLR GCHCLC-GRHCLC GCHCLC-GRHCLR GCHCLR-GRHCLC GCHCLR-GRHCLR GRHCLC-GRHCLR  Resilience T(obs) P 0.004 4.33 0.096 1.76 0.058 2.02 0.01 2.79 0.394 0.87 0.234 1.25 0.128 1.63 0.486 0.71 0.478 0.87 0.494 0.69 0.226 1.25 0.56 0.58 NA 65535 0.318 1.11 0.614 0.514 0.276 1.12 1 0  Sparse Tolerance Dense Tolerance T(obs) T(obs) P P 3.309 0.022 1.084 0.358 1.117 0.28 0.818 0.426 0.5 0.624 0.253 0.804 0.357 0.724 0.529 0.602 1.087 0.29 1.5 0.152 0.446 0.664 0.683 0.512 0.475 0.644 0.57 0.58 0.531 0.602 0.84 0.414 0.619 0.584 0.78 0.516 1.56 0.134 0.835 0.414 0.125 0.902 0.379 0.708 19.962 3.8E-11 6.833 1.8E-05 NA NA 65535 65535 2.234 0.066 1.141 0.306 7.297 0.00006 3.255 0.006 2.021 0.058 2.98 0.008 0.498 0.636 0.118 0.91  GCHCLC - Geographically Common, Habitat Common and Locally Common; GCHCLR - Geographically Common, Habitat Common and Locally Rare; GRHCLC Geographically Rare, Habitat Common and Locally Common; GRHCLR Geographically Rare, Habitat Common and Locally.  139  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0090270/manifest

Comment

Related Items