Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Tolerance analysis of habitat loss for multispecies conservation in western Riverside County, California, USA

Tolerance analysis of habitat loss for multispecies conservation in western Riverside County,... International Journal of Biodiversity Science and Management 2 (2006) 87–96 Tolerance analysis of habitat loss for multispecies conservation in western Riverside County, California, USA 1,2,3 1,2 2 2,4 Xiongwen Chen , Bai-Lian Li , Thomas Scott and Michael F. Allen Department of Botany and Plant Sciences, University of California, Riverside Center for Conservation Biology, University of California, Riverside Center of Forestry & Ecology, PO Box 1927, Alabama A & M University, Normal Department of Plant Pathology, University of California, Riverside Key words: Ecological safety, multiple species habitat conservation plans, habitat loss, reserve network system, tolerance analysis SUMMARY Understanding the structure of a reserve network and the tolerance of habitat loss within a reserve network system is important for regional species conservation management. To date, there exist only theoretical models of design using randomized spatial structure in which to test alternatives. The Western Riverside Multiple Species Habitat Conservation Plan provides a complex reserve pattern in which to hypothesize and test outcomes for very different species and groups of organisms. This study uses data analysis and computer simulation to investigate (i) the relationship between species richness and habitat quantity and (ii) tolerance of a reserve network to habitat loss. The results indicated exponential decay between species richness and habitat quantity for plants, reptiles, mammals, birds and all species. The current network of habitats for each species group has low tolerance for loss, especially for mammals. A conceptual model of an ecological safety net between habitat loss and connected networks has an ‘S’ shape, which means high tolerance for species conservation even under possible habitat loss. Reserve areas with viable buffers and corridors are essential to preserve biodiversity and to increase the ecological safety net of the entire habitat network at risk in areas such as western Riverside County of California. INTRODUCTION The objective of the Endangered Species Act is to accounted for 68% of the population variation preserve biodiversity and provide for the persist- (Wiegand et al. 2005). Over 1200 species are listed ence of species within the borders of the USA. As and there are many more candidates; only a few the human population grows and habitat is increas- species have been removed from the list due to ingly fragmented, species continue to be added to recovery (Stein et al. 2000). With this increasing threatened and endangered species lists. In the number, new approaches to species protection are USA, computer simulation suggests that habitat loss needed. Multiple species habitat conservation Correspondence: Xiongwen Chen, Center of Forestry & Ecology, PO Box 1927, AlabamaA&M University, Normal, AL 35762, USA. Email: xiongwen.chen@email.aamu.edu 87 Tolerance analysis of habitat loss Chen et al. plans (MSHCP) are increasingly replacing indivi- natural conservation will increase dramatically dual species habitat conservation plans (HCP), in under the projected three- to four-fold increase in part due to the growing numbers of species of con- numbers of people over the next 40 years, based cern within a region (e.g. Akcakaya 2000; Scott and on current projections (http://www.dof.ca.gov). Sullivan 2000). These community-scale planning Second, many rare species depend on private land efforts represent a unique experiment in species for their survival, which results in delays and reorga- protection, requiring new scientific approaches for nization of housing projects. Nationwide, more monitoring, management and decision-making. than 90% of federally listed species have some One of the difficulties in utilizing the MSHCP habitats on non-federal land, and 37–50% depend approach is that there are few data on which to entirely on non-federal land (US General Account- make decisions about either land areas to be placed ing Office 1994; Stein et al. 1995). This concern led in these plans or in developing survival and recov- to the western Riverside County Multiple Species ery goals for the species within the plans. Single Habitat Plan, which seeks the purchase of an species HCPs are generally focused around ‘charis- additional 64,000 ha of private land to add to an matic’ megafauna or a few plant species that are existing 146,000 ha of protected lands, creating readily recognized and generally well quantified. a preserve network. Following input by various However, in MSHCPs, even if quantitative data on stakeholder groups, 146 species were designated population parameters exist, they generally come as target species. Unfortunately, little data exist on from a few experiments on a small proportion of the vast majority of species, and existing data are the species in a few already protected areas, such as limited to location records, with only a few studies university field stations. More commonly, no data on dispersal or reproductive capacity. The assump- exist (particularly on species such as insects or tion is that purchasing and organizing the land into plants that are largely ignored). At best, limited dis- a reserve network will protect these species into the tribution data exist, but these are of variable quality future. This assumption will be explicitly tested because they are often derived from old historical over the next century. collections, often with outdated names and places, The range of most of these species is undergoing or come from quick surveys conducted on private fragmentationbysuburbandevelopment,evenwith lands for potential development purposes. Our protections provided by the MSHCP. Further, the quandary, as conservation biologists, has been to invasion of exotic species is dramatically changing either pursue purely theoretical studies by ran- localized species composition and subsequent eco- domizing patterns in computer simulations (e.g. system function (e.g. fire cycle). Most of the native Plotnick and Gardner 2002; With and King 1999) habitat is undergoing invasion by annuals domi- or small-scale model organisms such as beetles nated by European grasses or exotic black mustard (Gering et al. 2003). Alternatively, we can try to (Brassica nigra). Stylinski and Allen (1999) found model, based on the data available, hypothetical that nearly 60% of the cover on previously disturbed patterns that can then be explicitly re-evaluated by sites consisted of exotic annual species, with no evi- subsequent monitoring and experimental assess- dence of recovery to native shrublands even 40 years ments of populations and distributions as MSHCPs following disturbance. The increasing population are developed. pressures are creating additional changes in envi- Southern California represents a test case for ronmental conditions of wildland ecosystems. Vast these efforts. The region contains some of the most areas of land are exposed to low levels of diverse geological and biological landscapes in atmospheric nitrogen deposition, with elevated N North America. It is one of 25 hotspots of global deposition downwind of large and expanding met- biodiversity (Myers et al. 2000). Currently, there are ropolitan centres or large agricultural operations 102 state or federally listed species within a six- (Fenn et al. 2003). Elevated CO appears to favour county area. However, the region has been under- preferential growth of exotic grasses in nearby going changes in several ways. First, there is a major deserts (Smith et al. 2000). Global climate change, conflict between the habitat requirements of many with increased frequency of extreme weather events, species and the expanding human population with is shifting distribution ranges at large scales, which is needs for land, water, resources and recreation. intertwined with habitat fragmentation (Jump and The conflict between human development and Peñuelas 2005). The combination of environmental 88 International Journal of Biodiversity Science and Management Tolerance analysis of habitat loss Chen et al. change pressures – including greater productivity Current studies on the impact of habitat loss with elevated CO and N deposition, exotic grass generally focus on single species (or few species) invasion and increasing ignition events – may within test locales (e.g. Estrada et al. 2002). These decrease fire return interval, affecting wildland vege- bottom-up studies are useful to single species con- tation composition (e.g. Smith 2000; Mouillot et al. servation. However, the general study of habitat loss 2002; Fenn 2003). on many species and an entire habitat network in As a result, this region is undergoing major any region is lacking. To facilitate development in shifts in land use and land cover. There are more the western third of the County, and simultaneously endangered and threatened species in southern provide for the protection of the large number of California than any other area of similar size in the listed or other species of concern, Riverside County continental USA. Without care, once-extensive nat- undertook to create the Western Riverside Multiple ural communities, many of which are unique to the Species Habitat Conservation Plan (MSHCP), as ecoregion, will be reduced to isolated remnants. It part of the Integrated Planning processes. In is thus essential to manage available resources to on-going efforts to evaluate the MSHCP, we orga- sustain biodiversity in remaining habitats while pro- nized over 35,000 records of rare, concerned, viding opportunities for other appropriate uses of threatened and endangered species and their the land. Vulnerability analysis of large-scale land- habitats in western Riverside County. The goal of scape change could be very helpful for strategic this study is to evaluate the tolerance of possible management and adaptation (e.g. Roland 1993; habitat loss due to multiple causes to numerous Bergin et al. 2000; Kelley and Adger 2000). Highly species and the entire reserve network. optimized tolerance emphasizes robustness trade- offs which underlie resilience in different eco- systems (Moritz et al. 2005). Evaluation of tolerance MATERIALS AND METHODS to possible habitat loss will be useful, particularly Study area and species dataset in land acquisition for reserves, to manage the cur- rent available habitats and to restore the disturbed The study area is the planning area of the Multiple sites. Species Habitat Conservation Plan in western Figure 1 The location of study area (MSHCP study area) in USA International Journal of Biodiversity Science and Management 89 Tolerance analysis of habitat loss Chen et al. Table 1 General information of multispecies dataset Modelling in western Riverside County For this study, we evaluate the tolerance of possible Plants Reptiles Mammals Birds habitat loss on species richness and the entire Species number 57 16 14 51 reserve network in this area using two approaches. included First, we evaluated the relationship between species richness and number of habitat patches Federal or state listed 18 298 occupied by multiple species. We constructed the threatened and relationship between species richness rank and endangered species occupied patch (location) number based on record Species of special 313 2 31 information, such as how many habitat patches concern exist for any one species. These relationships were then combined to create a multiple species ‘habitat patch’. If most of these multispecies habitat patches Riverside County of California, USA (Figure 1). have a high species number, habitat loss will greatly The total area is about 5000 km . Minimum average impact species richness. In contrast, if most patches air temperature is about 3°C and the maximum is have a low species number, habitat loss will have less around 45°C. Annual precipitation averages affect on regional species richness, but it may result between 300–400 mm (http//rcip.org/mshcpdocs in species loss or the local extinction of an indivi- /vol1/mshcpvol1toc.htm). Lowland valleys occur dual population. at elevations below 600 m, and mountainous areas Our second approach was to assess the tolerance extend to over 3000 m. The species records (includ- of habitat loss on the entire MSHCP reserve. Here, ing rare, concerned, threatened and endangered) all habitat patches are considered as a complex of plants, reptiles, mammals and birds and their network for plants, reptiles, mammals and birds. locations in western Riverside County, obtained The characteristic of the whole network is usually from the Center for Conservation Biology database, evaluated by the diameter (e.g. Albert et al. 1999), were organized from numerous sources. Initial dis- such as r = d , where n is the total patch n ∑ tributions were provided to the public (http:// number, and d is the distance between any two ecoregion.ucr.edu/). Subsequently, we have que- patches. This formula is widely used in studies of ried numerous museum records and integrated data landscape ecology (Forman and Godron 1986). sources, including the California Natural Diversity Given the same meaning here, we use the total dis- Database, Fish and Wildlife Service records, and tance (D) between all distribution patches, instead others that were made available to us. In order to of diameter, to describe the network as follows: understand the original biodiversity pattern and dis- tribution in this area without human disturbance, D = d ∑ l i =1 record information was collected from these inde- pendent sources for over 100 years: we assumed (for where m is the total time for the distance measure- this study) that such data can describe well the origi- ment between any two patches (e.g. for four nal rare, concerned, threatened and endangered patches, m = 6). With habitat loss, D will decrease. If species (herein multispecies) and their distributions D decreases dramatically, this means that these in this area prior to the most recent human popula- patches are isolated from each other. For the pur- tion expansion. The summary of record information poses of this study, we assume that the tolerance of is in Table 1. All the locations were georeferenced by the whole network for habitat loss is low. The possi- latitude and longitude using ArcGIS. Eight species ble mechanisms for loss are: (i) patch isolation, with vague locations in their records were excluded. where it is difficult for species to migrate back to We recognize that this dataset does not represent all previous habitats (except for some birds); (ii) iso- species and does not represent the current distribu- lated habitat patches are more threatened by tion of species. However, it does integrate those surrounding land-use change; and (iii) isolated included for coverage in the MSHCP and it repre- habitat patches usually host low numbers of species. sents all data records prior up to 1995, when the Hence, these patches are critical and cannot be lost planning efforts commenced and the most recent from the network. In this study, we estimated D by majorgrowthinsuburbandevelopment began. simulations of random habitat loss of 1%, 5%, 10%, 90 International Journal of Biodiversity Science and Management Tolerance analysis of habitat loss Chen et al. Figure 2 The relationship between species richness and number of occupied patches for MSHCP-covered plants, reptiles, mammals, birds and all species 20%, 30% and 50% of the total number of patches RESULTS for different species groups. D is the averaged total Relationship between rank of species distance of the network for all simulated times. The richness and habitat patch number simulation number for the case of habitat loss of 1% is equal to the total habitat number for plants, Because of the complex spatial structure of reptiles, mammals and birds. In order to achieve biodiversity in this region, the covered species were stable variance of D, the random simulation is run widely distributed. The overall relationship 100 times, when habitat loss is more than 1% of the between rank of species richness and habitat patch total points for each species group. number in this area is characterized by an International Journal of Biodiversity Science and Management 91 Tolerance analysis of habitat loss Chen et al. exponential decay. For some species groups (e.g. highest tolerance to habitat reduction if that loss is plants and reptiles), only slight differences exist in less than 5%. A linear relationship exists between the relationship between rank of species richness percentage of habitat loss and percentage of D and habitat number (Figure 2a–e). Habitat patches loss, (y = 1.519x + 2.9154 (r = 0.9945, p < 0.01). with high species richness are very limited. For For both reptiles and birds, the percentage of D example, for plants, reptiles and mammals there loss changes little when the percentage of habitat are only 2–5 occupied habitats with species richness loss is between 5% and 10%. For all occupied more than 5; while for birds it is even less. For all habitats of each species group, the percentage of species under consideration in this area, there are D loss can reach 50% when the percentage of only around 10 (or less) occupied habitats with habitat loss is over 30%. Spatial location of habitats species richness greater than 5. For most individual is important for the integrity of the entire habitat habitats, species richness is limited, and more than network. From a regional perspective, even a slight half of the total number of patches comprises habi- loss of some habitat patches at critical locations tats with only one and two species (Figure 3). In may affect the sensitivity and the tolerance of the order to maintain higher biodiversity at regional overall preserve network. Maintaining preservation level, more habitats and a preservation network are in strategic areas will be critical for the whole net- needed, especially for isolated small patches. work. Keeping the total habitat area above at least 70% of its original covered area may be important for sustaining the integrity of the regional habitat Tolerance of reserve network for habitat network and for biodiversity conservation. These loss results demonstrate that choosing the proper loca- The reserve network of each species group has low tions and size of preserved areas is crucial to create tolerance for habitat loss. Further, with any habitat a functional reserve system. loss, the total distance between occupied habitats increases (Figure 4). For different species groups, DISCUSSION the change of total distance following habitat loss varies. Rare mammals have the lowest tolerance of In the study area, most location records contain few habitat loss in comparison with other groups. A rare species and most of the covered species are nonlinear relationship exists between habitat loss habitat-specific, so that most of the habitat patches (x) and percentage of D loss (y), y = 18.698Ln(x) are occupied by only 1–2 species. In order to pre- + 2.382 (r = 0.9732, p < 0.01). Plant species have serve habitat for a large number of these species, Figure 3 The percentage of points occupied with species richness of less than 3 in all locations of MSHCP-covered species of plants, reptiles, mammals, birds and all species 92 International Journal of Biodiversity Science and Management Tolerance analysis of habitat loss Chen et al. Figure 4 The relationship between habitat loss and the total distance between occupied points for MSHCP-covered species of plants, reptiles, mammals and birds ultimately be a critical limiting factor, and patterns are not well understood for the covered species, but dispersal ability appears to be relatively high. For example, Dipodomys stephensi (Merriam), a small mammal, disperses over 400 m, likely farther along dirt roads (Price et al. 1994). Larger animals, such as cougars, bobcats and coyotes, move over long dis- tances, especially dispersing juveniles (Beier 1995; Riley et al. 2003). Both plants (Ferguson 1999) and Figure 5 The conceptual model with ecological safety in tolerance of habitat loss rodents (Metcalf and Nunney 2001) show a low degree of population genetic separation, indicat- large core areas need to be protected and a reserve ing connected populations. However, even with network must be integrated between these core high dispersal ability, species cannot persist outside areas. The reasons that most species have restricted of suitable patches. distributions are due to complex geological and Currently, habitat networks are not well orga- climate constraints, and intrinsically high levels of nized. Some resource-rich habitats are not shared local disturbances (e.g. fire, drought, landslide, among species because of numerous issues ranging and flooding). Thus, environmental resources used from vegetation structure, to predators, to pollina- by species can be divided into discrete ‘mosaic tors. In order to preserve these species, the current elements’ that represent the different needs of reserves have to function as a single network and each species (Hutchinson and MacArthur 1959). In mechanisms have to be available for migration arid and semi-arid areas, such as western Riverside under conditions of global climatic change, local County, physical conditions are highly variable in environmental pollution and habitat fragmenta- space and time. Dispersal distances may or may not tion. Many of these species exhibit metapopulation International Journal of Biodiversity Science and Management 93 Tolerance analysis of habitat loss Chen et al. dynamics and these species will be lost when there the isolated habitats need special consideration are no linkages between them (e.g. Armsworth for protection. 2002). Finally, dispersal networks are dynamic, so Our methodology may be used to provide a that constructing adequate buffer areas around measure of ecological safety (Chen 2002) for a current locations will be necessary. habitat network system at regional level. Theoreti- The overall spatial structure of the MSHCP has cal research shows the existence of a common the potential to be highly connected, based on the destruction threshold for all species; when habitat structure of the proposed area to be incorporated loss reaches certain value, a sudden collapse in (Allen et al. 2003). The intent of the plan is to create biodiversity may occur (Loehle and Li 1996; Solé linkages between core areas such that genetic et al. 2004). For example, Gibbs (1998) studied the constraints are not created, and species can move in distribution of five species of woodland amphibians response to environmental change (Allen et al. along a forest fragmentation gradient. His results 2003). Of the 76,000 ha available for the analysis, indicated that some species (such as red-spotted only 64,000 ha will actually be integrated into the newts and wood frogs) were absent from different reserve network. The exact structure will be based portions of the gradient when forest cover was on a combination of available lands, habitat condi- reduced below a threshold of between 30% and tions and species distributions that remain to be 50%. Our results, based on the historical dataset determined. An additional concern is that many of of all species, provide a benchmark for habitat the ‘linkages’ are ‘constrained’. The definition of conservation and reserve construction. This ‘constrained’ used by the authors of the MSHCP approach can also be used to assess the ecological is ‘a constricted connection expected to provide for safety of a regional reserve network under the movement of identified Planning Species between pressure of land-use change. Core Areas, where options for assembly of the Western Riverside County poses an important connection are limited due to existing patterns of test of the MSHCP concept. Public and quasi-public use’ (http://rcip.org/mshcpdocs/vol11/mshcpv lands recognized as protected lands, constitute less ol11toc.htm). Thus, care is needed in the selection than 30% of the land area and would be badly frag- of lands incorporated into the MSHCP to ensure mented upon unchecked suburban development. that ‘constrained’ is not too constrained, and link- The additional 64,000 ha of private land, if prop- ages remain a priority for land acquisition in the erly chosen, may provide a connected network for reserve network. the species of concern. The current plan delineates The relationship between habitat loss and the 76,000 ha within which 64,000 ha will be purchased loss of integrity (e.g. total distance) of the network dependent on several criteria, including acceptable suggests that the responses of tolerance in habitat habitat, willing sellers and conservation value, loss for different species groups vary by group. But including connectivity. It remains important that for each species group, the habitat network will ecological factors, such as providing linkages lose about half of integrity (i.e. total distance D) between core reserves as well as economic areas, be when habitat loss is over 30%. There is a linear included in lands chosen for the MSHCP. response for habitat loss of plants, but no linear In conclusion, across western Riverside County, responses for reptiles, birds and mammals were as well as in many regions across the USA, most found. The occupancy network of mammals is species of rare, concerned, threatened and endan- more sensitive to habitat loss than for other species gered status have small habitats, and few patches groups. This outcome may suggest that an optimal with high species richness occur. The tolerance of habitat network would provide for little change of habitat loss in a network differs for plants, reptiles, integrity (such as total distance) and only a small mammals and birds. Habitats of mammals appear degree of habitat loss. The conceptual model to be very sensitive to the loss of a small patch should be S-shaped (Figure 5). This means that quantity. Thus, an optimal conceptual model of a the reserve network would be vulnerable under reserve network with high tolerance of habitat even a small amount of loss of habitats. With loss is proposed. Protected areas integrating the increasing loss of habitats, the integrity (i.e. total individual core preserves should be maintained or distance) of the network will decrease greatly. In constructed in order to achieve an ecologically order to create a safety net within the MSHCP areas, sound habitat network. 94 International Journal of Biodiversity Science and Management Tolerance analysis of habitat loss Chen et al. True for their help in organizing the record infor- ACKNOWLEDGEMENTS mation and Vanessa Isabel Rivera Del Río for Figure This research was partially supported by the UC 1. We also acknowledge numerous herbaria and Agricultural Experiment Station, Center for Con- museums for their response, with special thanks to servation Biology of UCR, County of Riverside, the the UCR herbarium, Rancho Santa Ana herbarium, California Department of Fish and Game and the Dickey Collection (University of California Los School of Agricultural and Environmental Sciences Angeles), Yale Peabody Museum, Los Angeles at Alabama A & M University. Special thanks to County Museum, the Museum of Vertebrate Zool- Tracy Tennant, Robert Cox, Adam Malisch, Monica ogy (University of HCalifornia Berkeley), CalFlora Swart, Edith Allen, Anthony Celis, Kathleen (University of California Berkeley), Riverside Fleming, Andrew Sanders, Steven Boyd and Chris Municipal Museum, and San Jose State University. REFERENCES Akcakaya HR. Conservation and management for Forman RTT and Godron M. Landscape Ecology. New multiple species: integrating field research and York: Wiley; 1986 modeling into management decisions. Environ- Gering JC, Crist TO and Veech JA. Additive partition- mental Management 2000;26:75–83. ing of species diversity across multiple spatial Albert R, Jeong H and Barabási A-L. Diameter of the scales: implications for regional conservation of world-wide web. Nature 1999;401:130–1 biodiversity. Conservation Biology 2003;17:488–99 Allen MF, Scott TA, Allen EB, Boyd S, Minnich R, Gibbs JP. Distribution of woodland amphibians along Malcolm J, Nunney L, Redak R and Reznick D. a forest fragmentation gradient. Landscape Ecology Scientific Review Panel Review of: Final Draft Western 1998;13:263–8 Riverside County MSHCP Document. http://reposito- Hutchinson GE and MacArthur RH. A theoretical eco- ries.cdlib.org/ccb/; 2003 logical model of size distributions among species. Armsworth PR. Recruitment limitation, population American Naturalist 1959;93:117–25 regulation, and larval connectivity in reef fish Jump AS and Peñuelas J. Running to stand still: adap- metapopulations. Ecology 2002;83:1092–104 tation and response of plants to rapid climate Beier P. Dispersal of juvenile cougars in fragmented change. Ecology Letters 2005;8:1010–20 habitat. Journal of Wildlife Management 1995;59: Kelley PM and Adger WN. Theory and practice in 228–37 assessing vulnerability to climate change and Bergin TM, Best LB, Freemark KE and Koehler KJ. facilitating adaptation. Climatic Change 2000;47: Effects of landscape structure on nest predation in 325–52 roadsides of a Midwestern agroecosystem: a Loehle C and Li B-L. Habitat destruction and the multiscale analysis. Landscape Ecology 2000;15: extinction debt revisted. Ecological Applications 131–43 1996;6:784–9 Chen X. Study on the ecological safety of the forest Metcalf AE, Nunney L and Hymann BC. Geographic in Northeast China under climate change. Interna- patterns of genetic differentiation within the tional Journal of Sustainable Development and World restricted range of the endangered Stephens’ Ecology 2002;9:49–58 kangaroo rat Dipodomys stephensi. Evolution Estrada A, Mendoza A, Castellanos L, Pacheco R, Belle 2001;55:1233–44 SV, García Y and Muñoz D. Population of the Black Moritz MA, Morais ME, Summerell LA, Carlson JM Howler monkey (Alouatta pigra) in a fragmented and Doyle J. Wildfires, complexity, and highly landscape in Palenque, Chiapas, Mexico. American optimized tolerance. Proceedings of National Journal of Primatology 2002;58:45–55 Academy of Sciences of USA 2005;13:17912–17 Fenn ME, Baron JS, Allen EB, Rueth HM, Nydick KR, Mouillot F, Rambal S and Joffre R. Simulation climate Geiser L, Bowman WD, Sickman JO, Meixner T change impacts on fire frequency and vegetation and Johnson DW. Ecological effects of nitrogen dynamics in a Mediterranean-type ecosystem. deposition in western United States. BioScience Global Change Biology 2002;8:423–37 2003;53:404–20 Myers N, Mittermeier RA, Mittermeier CG, da Fonseca Ferguson NJ. Demographic and genetic variation in GAB and Kent J. Biodiversity hotspots for conserva- Dodecahema leptoceras (Gray) Rev. and tion priorities. Nature 2000;403:853–8 Hardham (endangered species). University of Plotnick RE and Gardner RH. A general model for California, Riverside: PhD Thesis; 1999 simulating the effects of landscape heterogeneity International Journal of Biodiversity Science and Management 95 Tolerance analysis of habitat loss Chen et al. and disturbance on community pattern. Ecological Stein BA, Breden T and Warner R. Significance of Modelling 2002;147:171–97 federal lands for endangered species. In LaRoe ET Price MV, Kelly PA and Goldinggay RL. Distances (ed.), Our living resources: a report to the nation on the moved by staphens kangaroo-rat (Dipodomys distribution, abundance, and health of U.S. plants, stephensi Merriam) and implications for conserva- animals, and ecosystems. Washington DC: U.S. tion. Journal of Mammalogy 1994;75:929–39 National Biological Service; 1995:398–401 Riley, SPD, Sauvajot RM, Fuller TK, York EC, Kamradt Stein BA, Kuter LS and Adam JS. Previous Heritage: DA, Bromley C and Wayne RK. Effects of urbaniza- The Status of Biodiversity in the United States. Oxford tion and habitat fragmentation on bobcats and University Press; 2000 coyotes in southern California. Conservation Biology Stylinski CD and Allen EB. Lack of native species 2003;17:566–76 recovery following disturbance in southern Roland J. Large-scale forest fragmentation increase California shrublands. Journal of Applied Ecology the duration of tent caterpillar outbreak. Oecologia 1999;36:544–54 1993;93:25–30 US General Accounting Office. Endangered Species Act: Scott TA and Sullivan JE. The selection and design of information on species protection on nonfederal lands. multiple-species habitat preserves. Environmental Washington DC: US General Accounting Office; Management 2000;26:37–53 1994 Smith SD, Huxman TE, Zitzer SF, Charlet TN, Wiegand T, Revilla E and Moloney KA. Effects of Housman DC, Coleman JS, Fenstermaker LK, habitat loss and fragmentation on population Seemann JR and Nowak RS. Elevated CO increase dynamics. Conservation Biology 2005;19:108–21 productivity and invasive species success in an arid With KA and King AW. Extinction thresholds for ecosystem. Nature 2000;408:79–82 species in fractal landscapes. Conservation Biology Solé RV, Alonson D and Saldana J. Habitat frag- 1999;13:314–26 mentation and biodiversity collapse in neutral communities. Ecological Complexity 2004;1:65–75 96 International Journal of Biodiversity Science and Management http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png International Journal of Biodiversity Science & Management Taylor & Francis

Tolerance analysis of habitat loss for multispecies conservation in western Riverside County, California, USA

Download PDF
10 pages

Loading next page...
 
/lp/taylor-francis/tolerance-analysis-of-habitat-loss-for-multispecies-conservation-in-VmBvug7lZ2

References (36)

Publisher
Taylor & Francis
Copyright
Copyright Taylor & Francis Group, LLC
ISSN
1745-1604
eISSN
1745-1590
DOI
10.1080/17451590609618101
Publisher site
See Article on Publisher Site

Abstract

International Journal of Biodiversity Science and Management 2 (2006) 87–96 Tolerance analysis of habitat loss for multispecies conservation in western Riverside County, California, USA 1,2,3 1,2 2 2,4 Xiongwen Chen , Bai-Lian Li , Thomas Scott and Michael F. Allen Department of Botany and Plant Sciences, University of California, Riverside Center for Conservation Biology, University of California, Riverside Center of Forestry & Ecology, PO Box 1927, Alabama A & M University, Normal Department of Plant Pathology, University of California, Riverside Key words: Ecological safety, multiple species habitat conservation plans, habitat loss, reserve network system, tolerance analysis SUMMARY Understanding the structure of a reserve network and the tolerance of habitat loss within a reserve network system is important for regional species conservation management. To date, there exist only theoretical models of design using randomized spatial structure in which to test alternatives. The Western Riverside Multiple Species Habitat Conservation Plan provides a complex reserve pattern in which to hypothesize and test outcomes for very different species and groups of organisms. This study uses data analysis and computer simulation to investigate (i) the relationship between species richness and habitat quantity and (ii) tolerance of a reserve network to habitat loss. The results indicated exponential decay between species richness and habitat quantity for plants, reptiles, mammals, birds and all species. The current network of habitats for each species group has low tolerance for loss, especially for mammals. A conceptual model of an ecological safety net between habitat loss and connected networks has an ‘S’ shape, which means high tolerance for species conservation even under possible habitat loss. Reserve areas with viable buffers and corridors are essential to preserve biodiversity and to increase the ecological safety net of the entire habitat network at risk in areas such as western Riverside County of California. INTRODUCTION The objective of the Endangered Species Act is to accounted for 68% of the population variation preserve biodiversity and provide for the persist- (Wiegand et al. 2005). Over 1200 species are listed ence of species within the borders of the USA. As and there are many more candidates; only a few the human population grows and habitat is increas- species have been removed from the list due to ingly fragmented, species continue to be added to recovery (Stein et al. 2000). With this increasing threatened and endangered species lists. In the number, new approaches to species protection are USA, computer simulation suggests that habitat loss needed. Multiple species habitat conservation Correspondence: Xiongwen Chen, Center of Forestry & Ecology, PO Box 1927, AlabamaA&M University, Normal, AL 35762, USA. Email: xiongwen.chen@email.aamu.edu 87 Tolerance analysis of habitat loss Chen et al. plans (MSHCP) are increasingly replacing indivi- natural conservation will increase dramatically dual species habitat conservation plans (HCP), in under the projected three- to four-fold increase in part due to the growing numbers of species of con- numbers of people over the next 40 years, based cern within a region (e.g. Akcakaya 2000; Scott and on current projections (http://www.dof.ca.gov). Sullivan 2000). These community-scale planning Second, many rare species depend on private land efforts represent a unique experiment in species for their survival, which results in delays and reorga- protection, requiring new scientific approaches for nization of housing projects. Nationwide, more monitoring, management and decision-making. than 90% of federally listed species have some One of the difficulties in utilizing the MSHCP habitats on non-federal land, and 37–50% depend approach is that there are few data on which to entirely on non-federal land (US General Account- make decisions about either land areas to be placed ing Office 1994; Stein et al. 1995). This concern led in these plans or in developing survival and recov- to the western Riverside County Multiple Species ery goals for the species within the plans. Single Habitat Plan, which seeks the purchase of an species HCPs are generally focused around ‘charis- additional 64,000 ha of private land to add to an matic’ megafauna or a few plant species that are existing 146,000 ha of protected lands, creating readily recognized and generally well quantified. a preserve network. Following input by various However, in MSHCPs, even if quantitative data on stakeholder groups, 146 species were designated population parameters exist, they generally come as target species. Unfortunately, little data exist on from a few experiments on a small proportion of the vast majority of species, and existing data are the species in a few already protected areas, such as limited to location records, with only a few studies university field stations. More commonly, no data on dispersal or reproductive capacity. The assump- exist (particularly on species such as insects or tion is that purchasing and organizing the land into plants that are largely ignored). At best, limited dis- a reserve network will protect these species into the tribution data exist, but these are of variable quality future. This assumption will be explicitly tested because they are often derived from old historical over the next century. collections, often with outdated names and places, The range of most of these species is undergoing or come from quick surveys conducted on private fragmentationbysuburbandevelopment,evenwith lands for potential development purposes. Our protections provided by the MSHCP. Further, the quandary, as conservation biologists, has been to invasion of exotic species is dramatically changing either pursue purely theoretical studies by ran- localized species composition and subsequent eco- domizing patterns in computer simulations (e.g. system function (e.g. fire cycle). Most of the native Plotnick and Gardner 2002; With and King 1999) habitat is undergoing invasion by annuals domi- or small-scale model organisms such as beetles nated by European grasses or exotic black mustard (Gering et al. 2003). Alternatively, we can try to (Brassica nigra). Stylinski and Allen (1999) found model, based on the data available, hypothetical that nearly 60% of the cover on previously disturbed patterns that can then be explicitly re-evaluated by sites consisted of exotic annual species, with no evi- subsequent monitoring and experimental assess- dence of recovery to native shrublands even 40 years ments of populations and distributions as MSHCPs following disturbance. The increasing population are developed. pressures are creating additional changes in envi- Southern California represents a test case for ronmental conditions of wildland ecosystems. Vast these efforts. The region contains some of the most areas of land are exposed to low levels of diverse geological and biological landscapes in atmospheric nitrogen deposition, with elevated N North America. It is one of 25 hotspots of global deposition downwind of large and expanding met- biodiversity (Myers et al. 2000). Currently, there are ropolitan centres or large agricultural operations 102 state or federally listed species within a six- (Fenn et al. 2003). Elevated CO appears to favour county area. However, the region has been under- preferential growth of exotic grasses in nearby going changes in several ways. First, there is a major deserts (Smith et al. 2000). Global climate change, conflict between the habitat requirements of many with increased frequency of extreme weather events, species and the expanding human population with is shifting distribution ranges at large scales, which is needs for land, water, resources and recreation. intertwined with habitat fragmentation (Jump and The conflict between human development and Peñuelas 2005). The combination of environmental 88 International Journal of Biodiversity Science and Management Tolerance analysis of habitat loss Chen et al. change pressures – including greater productivity Current studies on the impact of habitat loss with elevated CO and N deposition, exotic grass generally focus on single species (or few species) invasion and increasing ignition events – may within test locales (e.g. Estrada et al. 2002). These decrease fire return interval, affecting wildland vege- bottom-up studies are useful to single species con- tation composition (e.g. Smith 2000; Mouillot et al. servation. However, the general study of habitat loss 2002; Fenn 2003). on many species and an entire habitat network in As a result, this region is undergoing major any region is lacking. To facilitate development in shifts in land use and land cover. There are more the western third of the County, and simultaneously endangered and threatened species in southern provide for the protection of the large number of California than any other area of similar size in the listed or other species of concern, Riverside County continental USA. Without care, once-extensive nat- undertook to create the Western Riverside Multiple ural communities, many of which are unique to the Species Habitat Conservation Plan (MSHCP), as ecoregion, will be reduced to isolated remnants. It part of the Integrated Planning processes. In is thus essential to manage available resources to on-going efforts to evaluate the MSHCP, we orga- sustain biodiversity in remaining habitats while pro- nized over 35,000 records of rare, concerned, viding opportunities for other appropriate uses of threatened and endangered species and their the land. Vulnerability analysis of large-scale land- habitats in western Riverside County. The goal of scape change could be very helpful for strategic this study is to evaluate the tolerance of possible management and adaptation (e.g. Roland 1993; habitat loss due to multiple causes to numerous Bergin et al. 2000; Kelley and Adger 2000). Highly species and the entire reserve network. optimized tolerance emphasizes robustness trade- offs which underlie resilience in different eco- systems (Moritz et al. 2005). Evaluation of tolerance MATERIALS AND METHODS to possible habitat loss will be useful, particularly Study area and species dataset in land acquisition for reserves, to manage the cur- rent available habitats and to restore the disturbed The study area is the planning area of the Multiple sites. Species Habitat Conservation Plan in western Figure 1 The location of study area (MSHCP study area) in USA International Journal of Biodiversity Science and Management 89 Tolerance analysis of habitat loss Chen et al. Table 1 General information of multispecies dataset Modelling in western Riverside County For this study, we evaluate the tolerance of possible Plants Reptiles Mammals Birds habitat loss on species richness and the entire Species number 57 16 14 51 reserve network in this area using two approaches. included First, we evaluated the relationship between species richness and number of habitat patches Federal or state listed 18 298 occupied by multiple species. We constructed the threatened and relationship between species richness rank and endangered species occupied patch (location) number based on record Species of special 313 2 31 information, such as how many habitat patches concern exist for any one species. These relationships were then combined to create a multiple species ‘habitat patch’. If most of these multispecies habitat patches Riverside County of California, USA (Figure 1). have a high species number, habitat loss will greatly The total area is about 5000 km . Minimum average impact species richness. In contrast, if most patches air temperature is about 3°C and the maximum is have a low species number, habitat loss will have less around 45°C. Annual precipitation averages affect on regional species richness, but it may result between 300–400 mm (http//rcip.org/mshcpdocs in species loss or the local extinction of an indivi- /vol1/mshcpvol1toc.htm). Lowland valleys occur dual population. at elevations below 600 m, and mountainous areas Our second approach was to assess the tolerance extend to over 3000 m. The species records (includ- of habitat loss on the entire MSHCP reserve. Here, ing rare, concerned, threatened and endangered) all habitat patches are considered as a complex of plants, reptiles, mammals and birds and their network for plants, reptiles, mammals and birds. locations in western Riverside County, obtained The characteristic of the whole network is usually from the Center for Conservation Biology database, evaluated by the diameter (e.g. Albert et al. 1999), were organized from numerous sources. Initial dis- such as r = d , where n is the total patch n ∑ tributions were provided to the public (http:// number, and d is the distance between any two ecoregion.ucr.edu/). Subsequently, we have que- patches. This formula is widely used in studies of ried numerous museum records and integrated data landscape ecology (Forman and Godron 1986). sources, including the California Natural Diversity Given the same meaning here, we use the total dis- Database, Fish and Wildlife Service records, and tance (D) between all distribution patches, instead others that were made available to us. In order to of diameter, to describe the network as follows: understand the original biodiversity pattern and dis- tribution in this area without human disturbance, D = d ∑ l i =1 record information was collected from these inde- pendent sources for over 100 years: we assumed (for where m is the total time for the distance measure- this study) that such data can describe well the origi- ment between any two patches (e.g. for four nal rare, concerned, threatened and endangered patches, m = 6). With habitat loss, D will decrease. If species (herein multispecies) and their distributions D decreases dramatically, this means that these in this area prior to the most recent human popula- patches are isolated from each other. For the pur- tion expansion. The summary of record information poses of this study, we assume that the tolerance of is in Table 1. All the locations were georeferenced by the whole network for habitat loss is low. The possi- latitude and longitude using ArcGIS. Eight species ble mechanisms for loss are: (i) patch isolation, with vague locations in their records were excluded. where it is difficult for species to migrate back to We recognize that this dataset does not represent all previous habitats (except for some birds); (ii) iso- species and does not represent the current distribu- lated habitat patches are more threatened by tion of species. However, it does integrate those surrounding land-use change; and (iii) isolated included for coverage in the MSHCP and it repre- habitat patches usually host low numbers of species. sents all data records prior up to 1995, when the Hence, these patches are critical and cannot be lost planning efforts commenced and the most recent from the network. In this study, we estimated D by majorgrowthinsuburbandevelopment began. simulations of random habitat loss of 1%, 5%, 10%, 90 International Journal of Biodiversity Science and Management Tolerance analysis of habitat loss Chen et al. Figure 2 The relationship between species richness and number of occupied patches for MSHCP-covered plants, reptiles, mammals, birds and all species 20%, 30% and 50% of the total number of patches RESULTS for different species groups. D is the averaged total Relationship between rank of species distance of the network for all simulated times. The richness and habitat patch number simulation number for the case of habitat loss of 1% is equal to the total habitat number for plants, Because of the complex spatial structure of reptiles, mammals and birds. In order to achieve biodiversity in this region, the covered species were stable variance of D, the random simulation is run widely distributed. The overall relationship 100 times, when habitat loss is more than 1% of the between rank of species richness and habitat patch total points for each species group. number in this area is characterized by an International Journal of Biodiversity Science and Management 91 Tolerance analysis of habitat loss Chen et al. exponential decay. For some species groups (e.g. highest tolerance to habitat reduction if that loss is plants and reptiles), only slight differences exist in less than 5%. A linear relationship exists between the relationship between rank of species richness percentage of habitat loss and percentage of D and habitat number (Figure 2a–e). Habitat patches loss, (y = 1.519x + 2.9154 (r = 0.9945, p < 0.01). with high species richness are very limited. For For both reptiles and birds, the percentage of D example, for plants, reptiles and mammals there loss changes little when the percentage of habitat are only 2–5 occupied habitats with species richness loss is between 5% and 10%. For all occupied more than 5; while for birds it is even less. For all habitats of each species group, the percentage of species under consideration in this area, there are D loss can reach 50% when the percentage of only around 10 (or less) occupied habitats with habitat loss is over 30%. Spatial location of habitats species richness greater than 5. For most individual is important for the integrity of the entire habitat habitats, species richness is limited, and more than network. From a regional perspective, even a slight half of the total number of patches comprises habi- loss of some habitat patches at critical locations tats with only one and two species (Figure 3). In may affect the sensitivity and the tolerance of the order to maintain higher biodiversity at regional overall preserve network. Maintaining preservation level, more habitats and a preservation network are in strategic areas will be critical for the whole net- needed, especially for isolated small patches. work. Keeping the total habitat area above at least 70% of its original covered area may be important for sustaining the integrity of the regional habitat Tolerance of reserve network for habitat network and for biodiversity conservation. These loss results demonstrate that choosing the proper loca- The reserve network of each species group has low tions and size of preserved areas is crucial to create tolerance for habitat loss. Further, with any habitat a functional reserve system. loss, the total distance between occupied habitats increases (Figure 4). For different species groups, DISCUSSION the change of total distance following habitat loss varies. Rare mammals have the lowest tolerance of In the study area, most location records contain few habitat loss in comparison with other groups. A rare species and most of the covered species are nonlinear relationship exists between habitat loss habitat-specific, so that most of the habitat patches (x) and percentage of D loss (y), y = 18.698Ln(x) are occupied by only 1–2 species. In order to pre- + 2.382 (r = 0.9732, p < 0.01). Plant species have serve habitat for a large number of these species, Figure 3 The percentage of points occupied with species richness of less than 3 in all locations of MSHCP-covered species of plants, reptiles, mammals, birds and all species 92 International Journal of Biodiversity Science and Management Tolerance analysis of habitat loss Chen et al. Figure 4 The relationship between habitat loss and the total distance between occupied points for MSHCP-covered species of plants, reptiles, mammals and birds ultimately be a critical limiting factor, and patterns are not well understood for the covered species, but dispersal ability appears to be relatively high. For example, Dipodomys stephensi (Merriam), a small mammal, disperses over 400 m, likely farther along dirt roads (Price et al. 1994). Larger animals, such as cougars, bobcats and coyotes, move over long dis- tances, especially dispersing juveniles (Beier 1995; Riley et al. 2003). Both plants (Ferguson 1999) and Figure 5 The conceptual model with ecological safety in tolerance of habitat loss rodents (Metcalf and Nunney 2001) show a low degree of population genetic separation, indicat- large core areas need to be protected and a reserve ing connected populations. However, even with network must be integrated between these core high dispersal ability, species cannot persist outside areas. The reasons that most species have restricted of suitable patches. distributions are due to complex geological and Currently, habitat networks are not well orga- climate constraints, and intrinsically high levels of nized. Some resource-rich habitats are not shared local disturbances (e.g. fire, drought, landslide, among species because of numerous issues ranging and flooding). Thus, environmental resources used from vegetation structure, to predators, to pollina- by species can be divided into discrete ‘mosaic tors. In order to preserve these species, the current elements’ that represent the different needs of reserves have to function as a single network and each species (Hutchinson and MacArthur 1959). In mechanisms have to be available for migration arid and semi-arid areas, such as western Riverside under conditions of global climatic change, local County, physical conditions are highly variable in environmental pollution and habitat fragmenta- space and time. Dispersal distances may or may not tion. Many of these species exhibit metapopulation International Journal of Biodiversity Science and Management 93 Tolerance analysis of habitat loss Chen et al. dynamics and these species will be lost when there the isolated habitats need special consideration are no linkages between them (e.g. Armsworth for protection. 2002). Finally, dispersal networks are dynamic, so Our methodology may be used to provide a that constructing adequate buffer areas around measure of ecological safety (Chen 2002) for a current locations will be necessary. habitat network system at regional level. Theoreti- The overall spatial structure of the MSHCP has cal research shows the existence of a common the potential to be highly connected, based on the destruction threshold for all species; when habitat structure of the proposed area to be incorporated loss reaches certain value, a sudden collapse in (Allen et al. 2003). The intent of the plan is to create biodiversity may occur (Loehle and Li 1996; Solé linkages between core areas such that genetic et al. 2004). For example, Gibbs (1998) studied the constraints are not created, and species can move in distribution of five species of woodland amphibians response to environmental change (Allen et al. along a forest fragmentation gradient. His results 2003). Of the 76,000 ha available for the analysis, indicated that some species (such as red-spotted only 64,000 ha will actually be integrated into the newts and wood frogs) were absent from different reserve network. The exact structure will be based portions of the gradient when forest cover was on a combination of available lands, habitat condi- reduced below a threshold of between 30% and tions and species distributions that remain to be 50%. Our results, based on the historical dataset determined. An additional concern is that many of of all species, provide a benchmark for habitat the ‘linkages’ are ‘constrained’. The definition of conservation and reserve construction. This ‘constrained’ used by the authors of the MSHCP approach can also be used to assess the ecological is ‘a constricted connection expected to provide for safety of a regional reserve network under the movement of identified Planning Species between pressure of land-use change. Core Areas, where options for assembly of the Western Riverside County poses an important connection are limited due to existing patterns of test of the MSHCP concept. Public and quasi-public use’ (http://rcip.org/mshcpdocs/vol11/mshcpv lands recognized as protected lands, constitute less ol11toc.htm). Thus, care is needed in the selection than 30% of the land area and would be badly frag- of lands incorporated into the MSHCP to ensure mented upon unchecked suburban development. that ‘constrained’ is not too constrained, and link- The additional 64,000 ha of private land, if prop- ages remain a priority for land acquisition in the erly chosen, may provide a connected network for reserve network. the species of concern. The current plan delineates The relationship between habitat loss and the 76,000 ha within which 64,000 ha will be purchased loss of integrity (e.g. total distance) of the network dependent on several criteria, including acceptable suggests that the responses of tolerance in habitat habitat, willing sellers and conservation value, loss for different species groups vary by group. But including connectivity. It remains important that for each species group, the habitat network will ecological factors, such as providing linkages lose about half of integrity (i.e. total distance D) between core reserves as well as economic areas, be when habitat loss is over 30%. There is a linear included in lands chosen for the MSHCP. response for habitat loss of plants, but no linear In conclusion, across western Riverside County, responses for reptiles, birds and mammals were as well as in many regions across the USA, most found. The occupancy network of mammals is species of rare, concerned, threatened and endan- more sensitive to habitat loss than for other species gered status have small habitats, and few patches groups. This outcome may suggest that an optimal with high species richness occur. The tolerance of habitat network would provide for little change of habitat loss in a network differs for plants, reptiles, integrity (such as total distance) and only a small mammals and birds. Habitats of mammals appear degree of habitat loss. The conceptual model to be very sensitive to the loss of a small patch should be S-shaped (Figure 5). This means that quantity. Thus, an optimal conceptual model of a the reserve network would be vulnerable under reserve network with high tolerance of habitat even a small amount of loss of habitats. With loss is proposed. Protected areas integrating the increasing loss of habitats, the integrity (i.e. total individual core preserves should be maintained or distance) of the network will decrease greatly. In constructed in order to achieve an ecologically order to create a safety net within the MSHCP areas, sound habitat network. 94 International Journal of Biodiversity Science and Management Tolerance analysis of habitat loss Chen et al. True for their help in organizing the record infor- ACKNOWLEDGEMENTS mation and Vanessa Isabel Rivera Del Río for Figure This research was partially supported by the UC 1. We also acknowledge numerous herbaria and Agricultural Experiment Station, Center for Con- museums for their response, with special thanks to servation Biology of UCR, County of Riverside, the the UCR herbarium, Rancho Santa Ana herbarium, California Department of Fish and Game and the Dickey Collection (University of California Los School of Agricultural and Environmental Sciences Angeles), Yale Peabody Museum, Los Angeles at Alabama A & M University. Special thanks to County Museum, the Museum of Vertebrate Zool- Tracy Tennant, Robert Cox, Adam Malisch, Monica ogy (University of HCalifornia Berkeley), CalFlora Swart, Edith Allen, Anthony Celis, Kathleen (University of California Berkeley), Riverside Fleming, Andrew Sanders, Steven Boyd and Chris Municipal Museum, and San Jose State University. REFERENCES Akcakaya HR. Conservation and management for Forman RTT and Godron M. Landscape Ecology. New multiple species: integrating field research and York: Wiley; 1986 modeling into management decisions. Environ- Gering JC, Crist TO and Veech JA. Additive partition- mental Management 2000;26:75–83. ing of species diversity across multiple spatial Albert R, Jeong H and Barabási A-L. Diameter of the scales: implications for regional conservation of world-wide web. Nature 1999;401:130–1 biodiversity. Conservation Biology 2003;17:488–99 Allen MF, Scott TA, Allen EB, Boyd S, Minnich R, Gibbs JP. Distribution of woodland amphibians along Malcolm J, Nunney L, Redak R and Reznick D. a forest fragmentation gradient. Landscape Ecology Scientific Review Panel Review of: Final Draft Western 1998;13:263–8 Riverside County MSHCP Document. http://reposito- Hutchinson GE and MacArthur RH. A theoretical eco- ries.cdlib.org/ccb/; 2003 logical model of size distributions among species. Armsworth PR. Recruitment limitation, population American Naturalist 1959;93:117–25 regulation, and larval connectivity in reef fish Jump AS and Peñuelas J. Running to stand still: adap- metapopulations. Ecology 2002;83:1092–104 tation and response of plants to rapid climate Beier P. Dispersal of juvenile cougars in fragmented change. Ecology Letters 2005;8:1010–20 habitat. Journal of Wildlife Management 1995;59: Kelley PM and Adger WN. Theory and practice in 228–37 assessing vulnerability to climate change and Bergin TM, Best LB, Freemark KE and Koehler KJ. facilitating adaptation. Climatic Change 2000;47: Effects of landscape structure on nest predation in 325–52 roadsides of a Midwestern agroecosystem: a Loehle C and Li B-L. Habitat destruction and the multiscale analysis. Landscape Ecology 2000;15: extinction debt revisted. Ecological Applications 131–43 1996;6:784–9 Chen X. Study on the ecological safety of the forest Metcalf AE, Nunney L and Hymann BC. Geographic in Northeast China under climate change. Interna- patterns of genetic differentiation within the tional Journal of Sustainable Development and World restricted range of the endangered Stephens’ Ecology 2002;9:49–58 kangaroo rat Dipodomys stephensi. Evolution Estrada A, Mendoza A, Castellanos L, Pacheco R, Belle 2001;55:1233–44 SV, García Y and Muñoz D. Population of the Black Moritz MA, Morais ME, Summerell LA, Carlson JM Howler monkey (Alouatta pigra) in a fragmented and Doyle J. Wildfires, complexity, and highly landscape in Palenque, Chiapas, Mexico. American optimized tolerance. Proceedings of National Journal of Primatology 2002;58:45–55 Academy of Sciences of USA 2005;13:17912–17 Fenn ME, Baron JS, Allen EB, Rueth HM, Nydick KR, Mouillot F, Rambal S and Joffre R. Simulation climate Geiser L, Bowman WD, Sickman JO, Meixner T change impacts on fire frequency and vegetation and Johnson DW. Ecological effects of nitrogen dynamics in a Mediterranean-type ecosystem. deposition in western United States. BioScience Global Change Biology 2002;8:423–37 2003;53:404–20 Myers N, Mittermeier RA, Mittermeier CG, da Fonseca Ferguson NJ. Demographic and genetic variation in GAB and Kent J. Biodiversity hotspots for conserva- Dodecahema leptoceras (Gray) Rev. and tion priorities. Nature 2000;403:853–8 Hardham (endangered species). University of Plotnick RE and Gardner RH. A general model for California, Riverside: PhD Thesis; 1999 simulating the effects of landscape heterogeneity International Journal of Biodiversity Science and Management 95 Tolerance analysis of habitat loss Chen et al. and disturbance on community pattern. Ecological Stein BA, Breden T and Warner R. Significance of Modelling 2002;147:171–97 federal lands for endangered species. In LaRoe ET Price MV, Kelly PA and Goldinggay RL. Distances (ed.), Our living resources: a report to the nation on the moved by staphens kangaroo-rat (Dipodomys distribution, abundance, and health of U.S. plants, stephensi Merriam) and implications for conserva- animals, and ecosystems. Washington DC: U.S. tion. Journal of Mammalogy 1994;75:929–39 National Biological Service; 1995:398–401 Riley, SPD, Sauvajot RM, Fuller TK, York EC, Kamradt Stein BA, Kuter LS and Adam JS. Previous Heritage: DA, Bromley C and Wayne RK. Effects of urbaniza- The Status of Biodiversity in the United States. Oxford tion and habitat fragmentation on bobcats and University Press; 2000 coyotes in southern California. Conservation Biology Stylinski CD and Allen EB. Lack of native species 2003;17:566–76 recovery following disturbance in southern Roland J. Large-scale forest fragmentation increase California shrublands. Journal of Applied Ecology the duration of tent caterpillar outbreak. Oecologia 1999;36:544–54 1993;93:25–30 US General Accounting Office. Endangered Species Act: Scott TA and Sullivan JE. The selection and design of information on species protection on nonfederal lands. multiple-species habitat preserves. Environmental Washington DC: US General Accounting Office; Management 2000;26:37–53 1994 Smith SD, Huxman TE, Zitzer SF, Charlet TN, Wiegand T, Revilla E and Moloney KA. Effects of Housman DC, Coleman JS, Fenstermaker LK, habitat loss and fragmentation on population Seemann JR and Nowak RS. Elevated CO increase dynamics. Conservation Biology 2005;19:108–21 productivity and invasive species success in an arid With KA and King AW. Extinction thresholds for ecosystem. Nature 2000;408:79–82 species in fractal landscapes. Conservation Biology Solé RV, Alonson D and Saldana J. Habitat frag- 1999;13:314–26 mentation and biodiversity collapse in neutral communities. Ecological Complexity 2004;1:65–75 96 International Journal of Biodiversity Science and Management

Journal

International Journal of Biodiversity Science & Management – Taylor & Francis

Published: Jun 1, 2006

Keywords: ECOLOGICAL SAFETY; MULTIPLE SPECIES HABITAT CONSERVATION PLANS; HABITAT LOSS; RESERVE NETWORK SYSTEM; TOLERANCE ANALYSIS

There are no references for this article.