Abstract

The contemporary genetic structure of animal populations is sculpted by past events, including demographic bottlenecks and expansions and movement of animals by humans. In an analysis of microsatellite DNA of black bears (Ursus americanus; n = 540) across California, we discovered distinct population structure and genetic evidence of 2 historic colonization events. First, genotypes of bears sampled in southern California are most related to those from the Yosemite National Park region and not with spatially intervening populations. Historical records recount the translocation of 28 black bears from the Yosemite National Park area of the central Sierra Nevada to the San Bernardino Mountains in southern California in the 1930s. Second, before colonization of California by Europeans, the Central Coast region was inhabited by the now extinct California grizzly bear (Ursus arctos californiensis), but not black bears. Following an apparent competitive release and range expansion during the past century, black bears now inhabit the Central Coast region of California. Black bears in California's Central Coast display lower genetic diversity (founder effect) and a genetic signature most closely allied with black bears from the southern Sierra Nevada. In both these cases, molecular genetic techniques allowed historical reconstruction of anthropogenic events leading to changes in animal distributions.

Mammals that are highly vagile are expected to display high gene flow and little population structure across the landscape (Wayne and Koepfli 1996). However, genetic differentiation can result from variation in gene flow associated not only with geographic barriers (Ernest et al. 2003; Keyghobadi et al. 1999; Sacks et al. 2004), but also habitat associations (Sacks et al. 2005), and animal aggregations and movements due to social dynamics (Tiedemann et al. 2000). Increasingly important to assess in highly impacted regions such as California are influences of anthropogenic factors on population genetic diversity and structure (Riley et al. 2006). Elucidating factors influencing gene flow in wildlife populations at the landscape level is important for an understanding of demography and for effective management (Storfer et al. 2007).

Distinct geographic boundaries within California have resulted in diverse ecotones and bioregions (Lapointe and Rissler 2005), including the vast Sacramento–San Joaquin Central Valley, and the Sierra Nevada, a range of mountains that runs north–south across much of the state. Studies in a number of taxonomic groups (birds, plants, insects, mammals, reptiles, and amphibians) with continuous distributions show population genetic substructure, as dissected by the Transverse Ranges, the Coast Ranges, and the Sierra Nevada (Calsbeek et al. 2003; Ernest et al. 2003; Feldman and Spicer 2006; Hull et al. 2008; Lapointe and Rissler 2005; Rissler et al. 2006).

The range of black bear (Ursus americanus) in California encompasses a variety of habitats distributed across multiple distinct bioregions. An estimated 30,000 black bears inhabit around 135,000 km2 of montane hardwood, chaparral, and mixed conifer forests in California (California Department of Fish and Game 1998). Black bears occupy the northern Coast Range, Cascades, Sierra Nevada, Transverse Ranges, southern California Ranges, and the Central Coast. The only habitats that exclude viable black bear populations are the highly agricultural Central Valley and the Mojave Desert. Black bears in western North America have average home ranges of 119 km2 for males and 49 km2 for females (Amstrup and Beecham 1976), with males dispersing and females typically remaining in natal habitat (Elowe and Dodge 1989; Schwartz and Franzman 1992). The range of black bears expanded into the Central Coast and southern California regions, areas previously inhabited by the competitively dominant California grizzly bear (Ursus arctos californicusGrinnell et al. 1937; Storer and Tevis 1996).

To date there has been no comprehensive analysis of the population genetics of black bears in California. Research examining the genetic structure of black bear populations elsewhere in North America has generally revealed isolation by distance with additional structure in response to habitat fragmentation (Dixon et al. 2007; Onorato et al. 2004; Triant et al. 2004), metapopulation dynamics (Onorato et al. 2007), and geographic barriers (Peacock et al. 2007).

Our goal was to use knowledge of the geography of California and the distribution of black bear habitat to evaluate population structure and patterns of genetic diversity using data from analyses of microsatellite DNA. This study will help delineate distinct populations of black bears in California based on genetic similarity, while also determining the source pool(s) for newly colonized regions and translocated individuals. We hypothesized that nonsuitable habitat for black bears (Central Valley, San Francisco/Bay–Delta, and the Los Angeles basin) would hinder gene flow; if descendents from the historical translocation were present and reproducing, the San Bernardino and San Gabriel mountains would display genetic similarity to the contemporary Sierra Nevada population; and genetic signatures of the Central Coast bears would indicate likely source population(s).

Materials and Methods

Sampling and DNA extraction.—Black bear samples (n = 540; females = 202, males = 328, unknown sex = 10) from across the range of the species in California were collected by collaborators, hunters, and researchers in 1990, 1995, 2000, and 2002–2004 (Fig. 1). Animal care and use guidelines were in compliance with the American Society of Mammalogists (Gannon et al. 2007). Location data reported by hunters or researchers were identified in the program TOPO! (National Geographic maps, National Geographic Society, Hanover, Pennsylvania), and latitudes and longitudes were recorded. Sample locations were plotted with ArcGIS 3.1 (Environmental Systems Research Institute, Redlands, California). Known relatives (sibling cubs or cubs associated with a female bear) were excluded from the analysis. DNA was extracted from dried gum tissue and muscle of bears using a Qiagen DNeasy 96 well extraction kit (Qiagen Inc., Valencia, California). Samples were stored at −80°C until extracted, and DNA was stored at −20°C.

Fig. 1

Locations for 540 tissue samples from black bears (Ursus americanus) collected between 1990 and 2004 throughout the geographic range of black bears in California.

Fig. 1

Locations for 540 tissue samples from black bears (Ursus americanus) collected between 1990 and 2004 throughout the geographic range of black bears in California.

Micro satellite typing.—The following 13 microsatellites from previously published studies of American black bears, Asiatic black bears (Ursus thibetanus), and brown bears (Ursus arctos) were optimized in multiplex panels: G1D, G1A, G10B, and G10L (Paetkau and Strobeck 1994); G10P, G10X, and G10C (Paetkau et al. 1995); UarMU59 and UarMU50 (Taberlet et al. 1997); G10J, G10, and G10H (Paetkau et al. 1998); and MSUT-8 (Kitahara et al. 2000). Two DNA polymerase chain reaction multiplexes of 6 loci each, and a reaction containing a single marker (MSUT-8) were utilized. All reactions were performed in ABI 3700 thermal cyclers (Applied Biosystems, Foster City, California) under the following thermal profile: 95°C for 2 min, followed by 33 cycles of 92°C for 30 s, 56°C for 30 s, 72°C for 1 min, with a final elongation step of 72°C for 30 min. The thermal profile for MSUT-8 was as follows: 95°C for 2 min, followed by 33 cycles of 92°C for 30 s, 53°C for 30 s, 72°C for 1 min, with a final elongation step of 72°C for 30 min. Forward primers were fluorescently labeled (6FAM, NED, VIC, and PET; Applied Biosystems). The reverse primer for MSUT-8 (AB040113) had a pig-tail (gttctt) added to the 5′ end to reduce stutter (Brownstein et al. 1996). The total polymerase chain reaction volume was 12 μl, including; 4.76 μl H2O, 0.2 mM each of deoxynucleoside triphosphates (Fisher Scientific, Pittsburgh, Pennsylvania), 1.5 mM MgCl2, 0.02 U Platinum Taq polymerase (Invitrogen, Carlsbad, California), IX Platinum Taq PCR Buffer (Invitrogen), primer concentrations varied from 0.25 μl to 0.50 μl per reaction, and 1 μl DNA. Polymerase chain reaction products were visualized on an ABI 3730 automated sequencer (Applied Biosystems) with an internal size standard, GeneScan 500 LIZ (Applied Biosystems). Alleles were scored using the program STRand version 2.3.69 (Toonen and Hughes 2001).

Analysis of microsatellite data.—Data input files were created with Microsatellite Toolkit 3.1 (Park 2001) and CONVERT 1.31 (Glaubitz 2004). Microsatellite Toolkit was used to calculate expected heterozygosity (HE) and observed heterozygosity (HO), allele frequencies, and diversity statistics. GENEPOP version 3.4 (Raymond and Rousset 1995) was used to test for departures from genotypic linkage equilibrium and used the probability test for Hardy–Weinberg equilibrium on the entire data set. Default settings were kept for both Hardy–Weinberg equilibrium (option 1, suboption 3) and linkage equilibrium (option 2, suboption 1). Tests for linkage equilibrium and Hardy–Weinberg equilibrium were adjusted for multiple comparisons using the Bonferroni correction (Rice 1989).

Genetic structure.—Clusters of genetically similar black bears in the state were defined using the software STRUCTURE version 2.1 (Pritchard et al. 2000), a model-based clustering method that groups individuals into clusters based on genotype without consideration of sampling geography. STRUCTURE was used to describe population groups and reveal patterns of genetic connectivity across the state. Nine independent runs of K = 1–20, where K denotes number of clusters, were performed at 500,000 Markov chain Monte Carlo repetitions and a burn-in period of 250,000. The admixture model of the program was used because of the expectation of high gene flow. The final K was chosen that best fit the data (denoted as X), by plotting the negative In Pr(XIK) versus K (K = 1–20) and then further assessing the K that did not sacrifice substantial explanatory power (Pritchard and Wen 2002; Waples and Gaggiotti 2006). The ln Pr(X|K) versus K provides an estimate of the actual number of clusters, but it should only be considered as an ad hoc guide to the correct value of K (Pritchard et al. 2000). Population clusters for the determined value of K were then overlain on a geographic information system map of California to aid in this process. Populations were designated based on geographic features in the region. Once Bayesian genetic clusters were determined, each group was tested for Hardy–Weinberg equilibrium and linkage equilibrium in a similar manner as the global data set, with a Bonferroni correction to account for locus by population interactions.

Microsatellite Toolkit was used to calculate HE and HO, allele frequencies, and diversity statistics of each genetic cluster. Allelic richness (Ar) was determined via FSTAT version 2.9.3 (Goudet 2001). Ar detects the allele count based on a corrected sample size (Petit et al. 1998). The degree of genetic differentiation between each population was assessed through pairwise estimates of FST in Arlequin 3.11 (Excoffier et al. 2005). Some microsatellite-based studies have used RST instead of FST; however, evidence suggests that RST may not be as robust in comparison to FST (Balloux and Goudet 2002; Gaggiotti et al. 1999).

To further assess genetic structure among black bears in the Sierra Nevada, we used analysis of isolation by distance between individuals and analysis of spatial autocorrelation. Tests of isolation by distance between populations were performed using partial correlations of pairwise distance matrices of the putative populations, using a Mantel test within the ISOLDE suboption in GENEPOP. Genetic distance was expressed as FST/(1 − FST), and geographic distance was expressed in kilometers. Geographic distance was determined by calculating the shortest (straight-line) distance through black bear habitat between the centers of each population. The genetic distance matrix was created in GENEPOP using FST. Isolation by distance between individuals (total, and in the Sierra Nevada as a separate analysis) was tested by a Mantel test in the program Alleles In Space (Miller 2005). Spatial autocorrelation analysis also was performed in Alleles In Space to investigate the correlation between genetic and geographic distance (by comparing an individual in one geographic class to all other distance classes) and to determine the inflection point between lower and higher than average correlation (Miller et al. 2006). The frequency of individuals with identical genotypes was calculated and compared to the other distance classes. Analysis was performed with 5,000 replicates over 10 distance classes (i.e., 0–82 km, 82–165 km, etc., to 826 km) to identify distance classes where genetic distances were significantly different from the null hypothesis (i.e., no correlation).

We used assignment tests to determine the likelihood an individual belonged to the source populations identified by the STRUCTURE analysis for southern California (n = 17) and Central Coast individuals (n = 15). Assignment tests were performed via GENECLASS2 (Piry et al. 2004), using the partial Bayesian approach of Rannala and Mountain (1997).

Results

Characteristics of microsatellite loci.—All 13 loci were polymorphic, with an average of 8.3 alleles/locus (range: 3–16 alleles/locus). HO (0.53) was within the range reported for populations of black bears from other regions of North America (0.36–0.81—Clarke et al. 2001; Paetkau et al. 1998; Paetkau and Strobeck 1994), and did not significantly differ from HE (0.58).

Hardy–Weinberg and linkage disequilibrium.—When considering all individuals as 1 population, 8 of 13 markers departed from expectations of Hardy–Weinberg equilibrium (Bonferroni-corrected level of significance of P = 0.001). One locus, G10C, was removed from the data set because it departed from expectations of Hardy–Weinberg equilibrium and linkage equilibrium in all populations and comparisons.

Linkage equilibrium tests revealed disequilibrium at 22 of 66 comparisons when all individuals were considered as 1 population, with a Bonferroni correction of P = 0.0008, thus suggesting the presence of geographical population structure.

Analysis of population structure.—Bayesian genetic clustering analysis (Pr(X|K)) was examined for K = 1–20 and revealed a range of potential populations. Only K = 1–7 are shown here, because after K = 7 the values plateau (Fig. 2). The value K = 4 had greatest explanatory power based on likelihood of K reaching maximal value at K = 4. This provides evidence that black bears in California are separated into 4 population genetic groups (Fig. 3A) partitioned according to geography: North Coast/Klamath (NC/K), Cascade/northern Sierra Nevada (C/NS), central Sierra Nevada/southern California (CS/SC), and southern Sierra Nevada/Central Coast (SS/CC). Additionally, at K = 4, the population clusters displayed metrics closer to Hardy–Wein-berg and linkage equilibria than at other values of K. Very similar geographic substructure was evident when analyzed on a subset of 323 individuals that had a probability of assignment at ≥70%. However, K = 3 also provided a plausible population subdivision when visualized on the geographic information system map.

Fig. 2

Results from STRUCTURE depicting In Pr(X|K) versus K 1–7. L(K) refers to the likelihood of K, X represents the data, K is the number of genetic clusters, and Pr is the probability.

Fig. 2

Results from STRUCTURE depicting In Pr(X|K) versus K 1–7. L(K) refers to the likelihood of K, X represents the data, K is the number of genetic clusters, and Pr is the probability.

Fig. 3

Distribution of 4 populations of black bears (Ursus americanus) detected by the program STRUCTURE. The populations are subdivided into the North Coast/Klamath (black circles), Cascades/northern Sierra Nevada (clear squares), central Sierra Nevada/southern California (gray triangles), and southern Sierra Nevada/Central Coast (white circles).

Fig. 3

Distribution of 4 populations of black bears (Ursus americanus) detected by the program STRUCTURE. The populations are subdivided into the North Coast/Klamath (black circles), Cascades/northern Sierra Nevada (clear squares), central Sierra Nevada/southern California (gray triangles), and southern Sierra Nevada/Central Coast (white circles).

Microsatellite genetic diversity within clusters.—Heterozy-gosities for each genetic grouping (Table 1) revealed no significant differences between HE and HO (P < 0.05). The North Coast/Klamath has the highest heterozygosity (HE = 0.63), compared to the southern Sierra Nevada/Central Coast with the lowest (HE = 0.46). Genetic diversity was similar to the values above, with southern California (HE = 0.48) and Central Coast (HE = 0.41) not showing significant difference between HE and HO.

Table 1

Sample size, expected heterozygosity (HE) and observed heterozygosity (HO), mean number of alleles per population, and allelic richness (Ar) for black bear (Ursus americanus) samples from the North Coast/Klamath, Cascade/northern Sierra Nevada, central Sierra Nevada/southern California, and southern Sierra Nevada/Central Coast. Populations in boldface type are included in the general population, but considered separately due to their disjunct location. Ar presented was based on an estimated population of n = 12, based on the lowest sample size for which there was a complete genotype data set. Regions denoted with an asterisk (*) display significantly different Ar in comparison to North Coast/Klamath.

Population Sample size HE HO Average alleles Ar 
North Coast/Klamath 167 0.63 0.61 9.2 5.4 
Cascade/northern Sierra Nevada 93 0.54 0.52 5.5 3.9 
Central Sierra Nevada/southern California 161 0.53 0.51 4.9 3.6* 
Southern Sierra Nevada/Central Coast 119 0.46 0.44 4.5 3.4* 
Southern California 17 0.48 0.52 3.6 3.4* 
Central Coast 17 0.41 0.39 3.1 2.9* 
Population Sample size HE HO Average alleles Ar 
North Coast/Klamath 167 0.63 0.61 9.2 5.4 
Cascade/northern Sierra Nevada 93 0.54 0.52 5.5 3.9 
Central Sierra Nevada/southern California 161 0.53 0.51 4.9 3.6* 
Southern Sierra Nevada/Central Coast 119 0.46 0.44 4.5 3.4* 
Southern California 17 0.48 0.52 3.6 3.4* 
Central Coast 17 0.41 0.39 3.1 2.9* 

The average number of alleles per population ranged from 4.5 (southern Sierra Nevada/Central Coast) to 9.2 (North Coast/Klamath). Samples from southern California and the Central Coast were analyzed separately to determine if these geographically disjunct populations had lower genetic diversity than their probable source populations (Table 2).

Table 2

Pairwise FST comparisons between each subpopulation (K = 4). NC/K = North Coast/Klamath, C/NS = Cascade/northern Sierra Nevada, CS/SC = central Sierra Nevada/southern California, and SS/CC = southern Sierra Nevada/Central Coast.

 CS/SC SS/CC C/NS 
SS/CC 0.06* — — 
C/NS 0.03* 0.10* — 
NC/K 0.07* 0.06* 0.06* 
 CS/SC SS/CC C/NS 
SS/CC 0.06* — — 
C/NS 0.03* 0.10* — 
NC/K 0.07* 0.06* 0.06* 
*

denotes significant value.

Population sizes among the 4 major groups and geographically disjunct groups (Central Coast and southern California) ranged from n = 17 to 167; however, because of missing alleles at 1 locus, Ar was estimated to a standard population size of n = 12. Ar ranged from 5.4 (North Coast/Klamath) to 2.9 (southern California subset). The North Coast/Klamath displayed significantly higher Ar than the southern Sierra Nevada/Central Coast (P = 0.02), central Sierra Nevada/southern California (P = 0.05), Central Coast (P = 0.02), and southern California (P = 0.02).

We then analyzed the probability of population structure with K = 2, 3, 4, and 5. Tests revealed that K of 2, 3, and 4 departed from Hardy–Weinberg equilibrium for 3, 4, and 1 locus, respectively. Tests showed that at a K of 2, 3, and 4 there were 2, 0, and 2 pairs, respectively, departing from linkage equilibrium. These results lend support to either K = 3 or K = 4, because these population groupings have the fewest loci out of linkage equilibrium and Hardy–Weinberg equilibrium.

Significant FST values between all possible pairs of the 4 genetic clusters ranged from 0.03 (between the Cascade/northern Sierra Nevada and central Sierra Nevada/southern California) to 0.10 (southern Sierra Nevada and the Cascade/northern Sierra Nevada), suggesting moderate levels (between 0.05 and 0.15—Wright 1951) of gene flow between all populations (Table 2). Tests of isolation by distance between all putative clusters revealed a positive, although not significant correlation (R2= 0.13, P = 0.09). We calculated isolation by distance for individuals from the Sierra Nevada to investigate whether individuals within this ecoregion exhibited an underlying isolation by distance pattern and to help discern between K = 3 or K = 4. A slight positive correlation (R2= 0.19) was found to be significant (P = 0.0009), suggesting a slight isolation by distance effect. Analysis of spatial autocorrelation found that genetic structure occurs in the distance classes between 300 and 700 km (P > 0.003; Fig. 4).

Fig. 4

Results of spatial autocorrelation analysis of individuals in the Sierra Nevada based on multilocus genotypes. Distances classes (10) are given on the x-axis between 0 and 826 km, and pairwise genetic distance (Ay) are given on the y-axis. The horizontal line indicates average genetic distance for the observed data set. Values above the horizontal line suggest correlation between distance class and genetic distance.

Fig. 4

Results of spatial autocorrelation analysis of individuals in the Sierra Nevada based on multilocus genotypes. Distances classes (10) are given on the x-axis between 0 and 826 km, and pairwise genetic distance (Ay) are given on the y-axis. The horizontal line indicates average genetic distance for the observed data set. Values above the horizontal line suggest correlation between distance class and genetic distance.

Individuals from the Central Coast and southern Sierra Nevada were genetically assigned to reference populations (Table 3) in all 34 cases except for 5 individuals. Four individuals assigned to a population with probabilities <60%: 3 of these individuals assigned to their population of sampling origin and 1 individual (CC2) assigned to a population that was different from its sampling origin. The 5th individual was sampled in southern California but had a high assignment to the North Coast/Klamath cluster.

Table 3

Assignment test results of individual black bears (Ursus americanus) from southern California (SC) and the Central Coast (CC) that were assigned to southern Sierra Nevada/Central Coast (SS/CC), central Sierra Nevada/southern California (CS/SC), or North Coast/Klamath (NC/K). Bayesian likelihood of assignment is presented. Four individuals (a) were assigned to a population with probabilities <60%: 3 of these individuals were assigned to their population of sampling origin and 1 individual (CC2) was assigned to a population that was different from its sampling origin. One individual (b) from southern California was assigned with high probability to the North Coast/Klamath population.

Sampling location Assigned population % assignment 
CC1 SS/CC 74.8 
CC2 SS/CCa 54.3 
CC3 SS/CC 99.0 
CC4 SS/CC 98.2 
CC5 SS/CC 99.3 
CC SS/CC 97.9 
CC7 SS/CC 99.4 
CC8 SS/CC 98.0 
CC9 SS/CC 99.2 
CC10 SS/CC 99.6 
CC11 SS/CC 99.9 
CC12 SS/CC 96.9 
CC13 SS/CC 90.7 
CC14 SS/CCa 48.6 
CC15 SS/CC 86.7 
CC16 SS/CC 79.6 
CC17 SS/CC 70.7 
SC1 CS/SC 90.1 
SC2 CS/SC 97.4 
SC3 CS/SC 87.9 
SC4 CS/SC 90.8 
SC5 CS/SC 92.7 
SC6 CS/SC 99.9 
SC7 CS/SC 98.5 
SC8 CS/SC 99.7 
SC9 CS/SC 96.7 
SC10 CS/SCa 59.9 
SC11 CS/SC 99.9 
SC12 CS/SC 94.0 
SC13 CS/SC 82.3 
SC14 NC/Kb 95.6 
SC15 CS/SC 99.7 
SC16 CS/SCa 53.8 
SC17 CS/SC 82.8 
Sampling location Assigned population % assignment 
CC1 SS/CC 74.8 
CC2 SS/CCa 54.3 
CC3 SS/CC 99.0 
CC4 SS/CC 98.2 
CC5 SS/CC 99.3 
CC SS/CC 97.9 
CC7 SS/CC 99.4 
CC8 SS/CC 98.0 
CC9 SS/CC 99.2 
CC10 SS/CC 99.6 
CC11 SS/CC 99.9 
CC12 SS/CC 96.9 
CC13 SS/CC 90.7 
CC14 SS/CCa 48.6 
CC15 SS/CC 86.7 
CC16 SS/CC 79.6 
CC17 SS/CC 70.7 
SC1 CS/SC 90.1 
SC2 CS/SC 97.4 
SC3 CS/SC 87.9 
SC4 CS/SC 90.8 
SC5 CS/SC 92.7 
SC6 CS/SC 99.9 
SC7 CS/SC 98.5 
SC8 CS/SC 99.7 
SC9 CS/SC 96.7 
SC10 CS/SCa 59.9 
SC11 CS/SC 99.9 
SC12 CS/SC 94.0 
SC13 CS/SC 82.3 
SC14 NC/Kb 95.6 
SC15 CS/SC 99.7 
SC16 CS/SCa 53.8 
SC17 CS/SC 82.8 

Discussion

Nuclear microsatellite analysis revealed a pattern of population genetic differentiation in black bears in California delimited both across and within biogeographic provinces. Black bears within the Sierra Nevada, a 650-km-long narrow strip of mountains extending north–south in California, are reproductively differentiated into at least 2 and possibly 3 genetic clusters. This finding illustrates a pattern of reproductive isolation, despite the capacity of this mobile species to travel large distances. Most remarkable is the genetic signature of a 75-year-old southern California translocation of black bears originating from Yosemite National Park in the central Sierra Nevada. This genetic analysis represents the 1st evidence of successful introduction and expansion of a translocated bear population in California. Genetic diversity (average heterozygosity, allelic diversity, and polymorphism of microsatellite markers) of our study groups was comparable to those of other populations of black bears in North America (Clarke et al. 2001; Paetkau et al. 1998; Paetkau and Strobeck 1994).

Population differentiation.—Examination of our data indicates either 3 or 4 genetic clusters (“populations”) that appear geographically relevant. Population subdivision was supported by the STRUCTURE analysis and FST, suggesting that each group is its own interbreeding population. For K = 4 the population clusters are North Coast/Klamath, Cascade/northern Sierra Nevada, central Sierra Nevada/southern California, and southern Sierra Nevada/Central Coast (Fig. 3). At K = 3, population subdivision follows the same pattern; however, the Sierra Nevada was partitioned into 2 subgroups instead of 3. The subdivision between central Sierra Nevada and southern Sierra Nevada remained intact, suggesting it is biologically relevant. However, for K = 3 the break between Cascades/northern Sierra Nevada and central Sierra Nevada was lost (not shown). Interestingly, at both K = 3 and K = 4 we discovered a genetic signature of the translocation of black bears from the central Sierra Nevada to southern California in the 1930s, and support for a southern Sierra Nevada origin for black bears in the Central Coast region.

We found evidence that spatial scale influenced the relative importance of isolation by distance to genetic subdivision. Isolation by distance has been reported in bears and other large carnivores occupying wide geographic regions (Ernest et al. 2003; Onorato et al. 2004; Paetkau et al. 1997). Other factors possibly interacting with isolation by distance in the Sierra Nevada to cause the observed genetic structure may include habitat imprinting and natal habitat bias (Sacks et al. 2004).

Areas of inhospitable habitat for black bears were correlated with reduced gene flow. Central Valley lowlands, agriculture, and urban development separate North Coast/Klamath from Sierra Nevada bear habitats as reflected by FST measures. This effect may be caused by distance alone or in combination with the (inhospitable) habitat type. On an east to west axis, the Central Valley very likely acts as a barrier to the movement of black bears, based on very few sightings of bears in this area. Examination of our data suggests that the Central Coast was colonized by the southern Sierra Nevada population, by way of the Transverse Ranges. Other studies (Calsbeek et al. 2003; Hull et al. 2008) have shown that a common pattern of genetic distinction in other animal groups occurs in relation to the geographic features of California's mountains: north–south differentiation defined by the Transverse Ranges, and east–west differentiation delimited by the Sierra Nevada and Coast Ranges. Black bears in California deviate from this pattern, at least in part due to recent population expansion from the Sierra Nevada. Studies of mitochondrial DNA also have uncovered genetic discontinuities along the Sierra Nevada in other species (reptiles [Feldman and Spicer 2006], amphibians [Vredenburg et al. 2007], and spiders [Starrett and Hedin 2007]). However, our microsatellite data likely reflect more recent population differentiation than mitochondrial differences and might be associated with habitat and climatic gradients as cryptic and semipermeable barriers.

Genetic evidence of translocation.—The genetic similarity we identified for black bears in the Sierra Nevada and southern California areas is presumably due to the translocation of bears from Yosemite National Park to several sites in the San Bernardino mountains in the 1930s (Burghduff 1935). Twenty-eight black bears were removed from the valley floor in Yosemite National Park and released near Crystal Lake (n = 11) in Los Angeles County, and Bear Lake (n = 6) and Santa Ana Canyon (n = 10) in San Bernardino County (Burghduff 1935). Records suggest that the translocation took place as an early attempt to develop ecotourism in the region (Burghduff 1935). Translocations of black bears for conservation and population augmentation were reportedly successful in Arkansas (Smith and Clark 1994) and Louisiana (Csiki et al. 2003); however, those populations remain fragmented and of conservation concern. Translocated populations typically show reduced levels of genetic diversity (Maudet et al. 2002; Vernesi et al. 2003), and can quickly differentiate from their source population due to founder effects and genetic drift (Larson et al. 2002; Polziehn et al. 2000).

Examination of our data shows that the southern California translocation resulted in a reproductively successful and persistent population, and the density of black bears in this area is 0.1 individuals/km2 (Novick 1979; Stubblefield 1992; Moss 1972). Heterozygosities and Ar remain roughly equal in both source and translocated populations. Although 75 years (roughly 12 generations based on average generation time of 6 years—Onorato et al. 2004) is relatively short compared to evolutionary timescales, decreased heterozygosity and allelic diversity have been shown in some studies in as little as 50 years (Mock et al. 2004) and 100 years (Sigg 2006).

Southern Sierra Nevada/Central Coast.—The ranges of black bears and grizzly bears overlapped historically in northern California, and in the Sierra Nevada (Storer and Tevis 1996). Grizzly bears occupied coastal areas of California (northern, central, and southern), the Central Valley, and Sierra Nevada foothills, and once numbered around 10,000 individuals (Storer and Tevis 1996). Grizzly bears likely excluded black bears where the former species was abundant (i.e., coast), and once the grizzly was extirpated (1922), black bears may have been competitively released and expanded their range; such evidence has been reported for black and grizzly bears elsewhere in North America (Veitch and Harrington 1996). Examination of our data (Fig. 3) indicates that black bears in the Central Coast region are an extension of populations from the southern Sierra Nevada and Tehachapi Mountains, rather than from southern California.

Natural colonization events may cause similar population-level effects as translocations, including genetic bottlenecks and low genetic diversity (Broders et al. 1999; Fabbri et al. 2007). Heterozygosities for black bears in the Central Coast region were lower than for the source population in the southern Sierra Nevada; however, there was no significant difference between these 2 regions in Ar. Because examination of our data indicated that this area was likely colonized from the southern Sierra Nevada, the functional integrity of the migration corridor through the Transverse and Tehachapi Ranges will control natural gene flow.

North Coast/Klamath.—This region has the highest density of black bears in California (0.4–1.0 individuals/km2Kellyhouse 1977; Piekielek and Burton 1973), and bears in this area are genetically distinct from those in southerly regions. Black bears from this area displayed the highest level of genetic diversity, perhaps a reflection of connectivity with large populations of black bears in the Pacific Northwest. There was a noticeable overlap with the Cascade/northern Sierra Nevada grouping in a habitat transition zone between mesic coniferous forests and the more arid regions of northeastern California (Modoc Plateau). The maintenance of such a genetic break in this region may be a consequence of natal habitat-biased dispersal, as reported for coyotes (Canis latrans) in California (Sacks et al. 2004).

Cascade/northern Sierra Nevada.—High levels of genetic diversity and gene flow likely exist among Cascade/northern Sierra Nevada, North Coast/Klamath, and central Sierra Nevada populations, thus revealing a transition zone between these biogeographically disparate regions.

Management implications and conclusions.—Black bears in California show population substructure by geographic barriers and isolation by distance. Populations of black bears in this state are not all spatially discrete subunits, but likely experience migrations among contiguous habitats, and as shown in this study, the genetic signature of human-mediated translocation. Although black bears in the southern California and Central Coast regions are geographically isolated from larger source populations, these populations maintain levels of genetic diversity on par with other populations of bears in North America (Clarke et al. 2001; Paetkau et al. 1998; Paetkau and Strobeck 1994). These areas will be especially important to monitor in the future. Habitats that function as migration corridors for black bears, such as the corridor through the Transverse and Tehachapi ranges, will need to be protected for the southern California and Central Coast populations to remain genetically viable. Future research on the genetics of black bears in California should be focused on identifying the levels of ongoing or potential gene flow between subpopulations in the Transverse Ranges and the Central Coast, and between the Central Coast and southern California areas.

Acknowledgments

This research was supported by the California Department of Fish and Game and the University of California Davis Veterinary Genetics Laboratory, Davis, California. Thanks to all of those who contributed samples for this analysis, including B. Amado, D. Bowman, R. Brown, D. Buchanan, T. Burton, J. Bushly, F. Cox, R. Elliot, D. Graber, M. Harvey, M. Kie, W. Klein, K. McCurdy, H. Pierce, T. Seher, T. Stone, S. Thompson, J. P. Walker, H. Werner, R. Mazur, National Park Service, California Department of Fish and Game, and Hoopa Valley Indian Tribe. We appreciate the technical assistance of J. Well, J. Kurushima, K. Records, C. Williams, A. Irish, N. Fauzi, and T. Kun. Many thanks to M. Stephens, A. Drauch, E. Heske, R. Sweitzer, and anonymous reviewers who improved our paper by comments and insights.

Literature Cited

Amstrup
S. C.
Beecham
J.
.
1976
.
Activity patterns of radio-collared black bears in Idaho
.
Journal of Wildlife Management
 
40
:
340
348
.
Balloux
F.
Goudet
J.
.
2002
.
Statistical properties of population differentiation estimators under stepwise mutation in a finite island model
.
Molecular Ecology
 
11
:
771
783
.
Broders
H. G.
Mahoney
S. P.
Montevecchi
W. A.
Davidson
W. S.
.
1999
.
Population genetic structure and the effect of founder events on the genetic variability of moose, Alces alces, in Canada
.
Molecular Ecology
 
8
:
1309
1315
.
Brownstein
M. J.
Carpten
J. D.
Smith
J. R.
.
1996
.
Modulation of non-templated nucleotide addition by taq DNA polymerase: primer modifications that facilitate genotyping
.
BioTechniques
 
20
:
1004
1010
.
Burghduff
A.
1935
.
Black bears released in southern California
.
California Fish and Game
 
21
:
83
84
.
California Department of Fish and Game
.
1998
.
Black bear management plan. California Department of Fish and Game, Sacramento
 . http://www.dfg.ca.gov/wildlife/hunting/bear/.
Accessed 12 November 2008
.
Calsbeek
R.
Thompson
J. N.
Richardson
J. E.
.
2003
.
Patterns of molecular evolution and diversification in a biodiversity hotspot: the California Floristic Province
.
Molecular Ecology
 
12
:
1021
1029
.
Clarke
C. M.
Immell
D.
Wasser
S. K.
.
2001
.
Technical considerations for hair genotyping models in black bears
.
Western Workshop for Black Bear Research and Management
 
7
:
24
29
.
Csiki
I.
et al
.
2003
.
Genetic variation in black bears in Arkansas and Louisiana using microsatellite DNA markers
.
Journal of Mammalogy
 
84
:
691
701
.
Dixon
J. D.
Oli
M. K.
Wooten
M. C.
Eason
T. H.
McCown
J. W.
Cunningham
M. W.
.
2007
.
Genetic consequences of habitat fragmentation and loss: the case of the Florida black bear (Ursus americanus floridanus)
.
Conservation Genetics
 
8
:
455
464
.
Elowe
K. D.
Dodge
W. E.
.
1989
.
Factors affecting black bear reproductive success and cub survival
.
Journal of Wildlife Management
 
53
:
962
968
.
Ernest
H. B.
Boyce
W. M.
Bleich
V. C.
May
B. P.
Stiver
S. J.
Torres
S. G.
.
2003
.
Genetic structure of mountain lion (Puma concolor) populations in California
.
Conservation Genetics
 
4
:
353
366
.
Excoffier
L.
Laval
G.
Schneider
S.
.
2005
.
Arlequin version 3.0: an integrated software package for population genetics data analysis
.
Evolutionary Bioinformatics Online
 
1
:
47
50
.
Fabbri
E.
et al
.
2007
.
From the Apennines to the Alps: colonization genetics of the naturally expanding Italian wolf (Canis lupus) population
.
Molecular Ecology
 
16
:
1661
1671
.
Feldman
C. R.
Spicer
G. S.
.
2006
.
Comparative phylogeography of woodland reptiles in California: repeated patterns of cladogenesis and population expansion
.
Molecular Ecology
 
15
:
2201
2222
.
Gaggiotti
O. E.
Lange
O.
Rassmann
K.
Gliddon
C.
.
1999
.
A comparison of two indirect methods for estimating average levels of gene flow using microsatellite data
.
Molecular Ecology
 
8
:
1513
1520
.
Gannon
W. L.
Sikes
R. S.
the Animal Care and Use Committee of the American Society of Mammalogists
.
2007
.
Guidelines of the American Society of Mammalogists for the use of wild mammals in research
.
Journal of Mammalogy
 
88
:
809
823
.
Glaubitz
J. C.
2004
.
CONVERT: a user-friendly program to reformat diploid genotypic data for commonly used population genetic software packages
.
Molecular Ecology Notes
 
4
:
309
310
.
Goudet
J.
2001
.
FSTAT, a program to estimate and test gene diversities and fixation indices (version 2.9.3)
 . from http://www.unil.ch/izea/softwares/fstat.html.
Accessed 1 May 2008
.
Grinnell
J.
Dixon
J. S.
Linsdale
J. M.
.
1937
.
Fur-bearing mammals of California: their natural history, systematic status, and relations to man
 .
University of California Press
,
Berkeley
.
Hull
J. M.
Hull
A. C.
Sacks
B. N.
Smith
J. P.
Ernest
H. B.
.
2008
.
Landscape characteristics influence morphological and genetic differentiation in a widespread raptor (Buteo jamaicensis)
.
Molecular Ecology
 
17
:
810
824
.
Kellyhouse
D. G.
1977
.
Habitat utilization of black bears in northern California
. Pp.
26
31
in
Bears—their biology and management
  (
Herrero
S.
, ed.).
International Union for Conservation of Nature Publications New Series 23
.
Keyghobadi
N.
Roland
J.
Strobeck
C.
.
1999
.
Influence of landscape on the population genetic structure of the alpine butterfly Parnassius smintheus (Papilionidae)
.
Molecular Ecology
 
8
:
1481
1495
.
Kitahara
E.
Isagi
Y.
Ishibashi
Y.
Saitoh
AND T.
.
2000
.
Polymorphic microsatellite DNA markers in the Asiatic black bear Ursus thibetanus
.
Molecular Ecology
 
9
:
1661
1662
.
Lapointe
F.-J.
Rissler
L. J.
.
2005
.
Congruence, consensus, and the comparative phylogeography of codistributed species in California
.
American Naturalist
 
166
:
290
299
.
Larson
S.
Jameson
R.
Bodkin
J.
Staedler
M.
Bentzen
P.
.
2002
.
Microsatellite DNA and mitochondrial DNA variation in remnant and translocated sea otter (Enhydra lutris) populations
.
Journal of Mammalogy
 
83
:
893
906
.
Maudet
C
et al
.
2002
.
Microsatellite DNA and recent statistical methods in wildlife conservation management: applications in alpine ibex Capra ibex (ibex)
.
Molecular Ecology
 
11
:
421
436
.
Miller
M. P.
2005
.
Alleles In Space (AIS): computer software for the joint analysis of interindividual spatial and genetic information
.
Journal of Heredity
 
96
:
722
724
.
Miller
M. P.
Haig
S. M.
Wagner
R. S.
.
2006
.
Phylogeography and spatial genetic structure of the southern torrent salamander: implications for conservation and management
.
Journal of Heredity
 
97
:
561
570
.
Mock
K. E.
Latch
E. K.
Rhodes
O. E.
.
2004
.
Assessing losses of genetic diversity due to translocation: long-term case histories in Merriam's turkey (Meleagris gallopavo merriami)
.
Conservation Genetics
 
5
:
631
645
.
Moss
H.
1972
.
A study of the black bear in the San Gabriel Mountains
.
M.S. thesis
 ,
California State Polytechnic University
.
Novick
H.
1979
.
Home range and habitat preferences of black bears (Ursus americanus) in the San Bernardino Mountains of southern California
.
M.S. thesis
 ,
California State Polytechnic University
,
Pomona
.
Onorato
D. P.
Hellgren
E. C.
Van den Bussche
R. A.
Doan-Crider
D. L.
.
2004
.
Phylogeographic patterns within a metapopulation of black bears (Ursus americanus) in the American southwest
.
Journal of Mammalogy
 
85
:
140
147
.
Onorato
D. P.
Hellgren
E. C.
Van den bussche
R. A.
Doan-Crider
D. L.
Skiles
J. R.
Jr.
2007
.
Genetic structure of American black bears in the desert southwest of North America: conservation implications for recolonization
.
Conservation Genetics
 
8
:
565
576
.
Paetkau
D.
Calvert
W.
Sterling
I.
Strobeck
C.
.
1995
.
Microsatellite analysis of population structure in Canadian polar bears
.
Molecular Ecology
 
4
:
347
354
.
Paetkau
D.
Shields
G. F.
Strobeck
C.
.
1998
.
Gene flow between insular, coastal and interior populations of brown bears in Alaska
.
Molecular Ecology
 
7
:
1282
1292
.
Paetkau
D.
Strobeck
C.
.
1994
.
Microsatellite analysis of genetic variation in black bear populations
.
Molecular Ecology
 
3
:
489
495
.
Paetkau
D.
Waits
L. P.
Clarkson
P. L.
Craighead
L.
Strobeck
C.
.
1997
.
An empirical evaluation of genetic distance statistics using microsatellite data from bear (Ursidae) populations
.
Genetics
 
147
:
1947
1953
.
Park
S. D. E.
2001
.
Trypanotolerance in West African cattle and the population genetic effects of selection
.
Ph.D. dissertation
 ,
University of Dublin
,
Dublin, Ireland
.
Peacock
E.
Peacock
M. M.
Titus
K.
.
2007
.
Black bears in southeast Alaska: the fate of two ancient lineages in the face of contemporary movement
.
Journal of Zoology (London)
 
271
:
445
454
.
Petit
R. J.
El Mousadik
A.
Pons
O.
.
1998
.
Identifying populations for conservation on the basis of genetic markers
.
Conservation Biology
 
12
:
844
855
.
Piekielek
W.
Burton
T.
.
1973
.
A black bear population study in northern California
.
California Fish and Game
 
61
:
4
25
.
Piry
S.
Alapetite
A.
Cornuet
J. M.
Paetkau
D.
Baudouin
L.
Estoup
A.
.
2004
.
GENECLASS2: a software for genetic assignment and first-generation migrant detection
.
Journal of Heredity
 
95
:
536
539
.
Polziehn
R. O.
Hamr
J.
Mallory
F. F.
Strobeck
C.
.
2000
.
Micro satellite analysis of North American wapiti (Cervus elaphus) populations
.
Molecular Ecology
 
9
:
1561
1576
.
Pritchard
J. K.
Stephens
M.
Donnelly
P.
.
2000
.
Inference of population structure using multilocus genotype data
.
Genetics
 
155
:
945
959
.
Pritchard
J. K.
Wen
W.
.
2002
.
Documentation for STRUCTURE software version 2
 . http://pritch.bsd.uchicago.edu.
Accessed 1 April 2007
.
Rannala
B.
Mountain
J.
.
1997
.
Detecting immigration by using multilocus genotypes
.
Proceedings of the National Academy of Sciences
 
94
:
9197
9201
.
Raymond
M.
Rousset
F.
.
1995
.
GENEPOP ver. 3.2 a: a population genetics software for exact tests and ecumenicism
.
Journal of Heredity
 
86
:
248
249
.
Rice
W. R.
1989
.
Analyzing tables of statistical tests
.
Evolution
 
43
:
223
225
.
Riley
S. P. D.
et al
.
2006
.
A southern California freeway is a physical and social barrier to gene flow in carnivores
.
Molecular Ecology
 
15
:
1733
1741
.
Rissler
L. J.
Hijmans
R. J.
Graham
C. H.
Moritz
C.
Wake
D. B.
.
2006
.
Phylogeographic lineages and species comparisons in conservation analyses: a case study of California herpetofauna
.
American Naturalist
 
167
:
655
666
.
Sacks
B. N.
Brown
S. K.
Ernest
H. B.
.
2004
.
Population structure of California coyotes corresponds to habitat-specific breaks and illuminates species history
.
Molecular Ecology
 
13
:
1265
1275
.
Sacks
B. N.
Mitchell
B. R.
Williams
C. L.
Ernest
H. B.
.
2005
.
Coyote movements and social structure along a cryptic population genetic subdivision
.
Molecular Ecology
 
14
:
1241
1249
.
Schwartz
C. C.
Franzmann
A. W.
.
1992
.
Dispersal and survival of subadult black bears from the Kenai Peninsula, Alaska
.
Journal of Wildlife Management
 
56
:
426
431
.
Sigg
D. P.
2006
.
Reduced genetic diversity and significant genetic differentiation after translocation: comparison of the remnant and translocated populations of bridled nailtail wallabies (Onychogalea fraenata)
.
Conservation Genetics
 
7
:
577
589
.
Smith
K. G.
Clark
J. D.
.
1994
.
Black bears in Arkansas: characteristics of a successful translocation
.
Journal of Mammalogy
 
75
:
309
320
.
Starrett
J.
Hedin
M.
.
2007
.
Multilocus genealogies reveal multiple cryptic species and biogeographical complexity in the California turret spider Antrodiaetus riversi (Mygalomorphae, Antrodiaetidae)
.
Molecular Ecology
 
16
:
583
604
.
Storer
T. L.
Tevis
J.
.
1996
.
California grizzly
 .
University of California Press
,
Los Angeles
.
Storfer
A.
et al
.
2007
.
Putting the ‘landscape’ in landscape genetics
.
Heredity
 
98
:
128
142
.
Stubblefield
C.
1992
.
Characteristics of black bear ecology in the San Gabriel Mountains of southern California
.
M.S. thesis
 ,
California State Polytechnic University
,
Pomona
.
Taberlet
P.
et al
.
1997
.
Non-invasive genetic tracking of the endangered Pyrenean brown bear population
.
Molecular Ecology
 
6
:
869
876
.
Tiedemann
R.
Hardy
O.
Vekemans
X.
Milinkovitch
M. C.
.
2000
.
Higher impact of female than male migration on population structure in large mammals
.
Molecular Ecology
 
9
:
1159
1163
.
Toonen
R. J.
Hughes
S.
.
2001
.
Increased throughput for fragment analysis on an ABI PRISM 377 automated sequencer using a membrane comb and STRand software
.
Biotechniques
 
31
:
1320
1324
.
Triant
D. A.
Pace
R. M.
Stine
M.
.
2004
.
Abundance, genetic diversity and conservation of Louisiana black bears (Ursus americanus luteolus) as detected through noninvasive sampling
.
Conservation Genetics
 
5
:
647
659
.
Veitch
A. M.
Harrington
F. H.
.
1996
.
Brown bears, black bears, and humans in northern Labrador: an historical perspective and outlook to the future
.
Journal of Wildlife Research
 
1
:
244
249
.
Vernesi
C.
et al
.
2003
.
The genetic impact of demographic decline and reintroduction in the wild boar (Sus scrofd): a microsatellite analysis
.
Molecular Ecology
 
12
:
585
595
.
Vredenburg
V. T.
Bingham
R.
Knapp
R.
Morgan
J. A. T.
Moritz
C.
Wake
D.
.
2007
.
Concordant molecular and pheno-typic data delineate new taxonomy and conservation priorities for the endangered mountain yellow-legged frog
.
Journal of Zoology (London)
 
271
:
361
374
.
Waples
R. S.
Gaggiotti
O.
.
2006
.
What is a population? An empirical evaluation of some genetic methods for identifying the number of gene pools and their degree of connectivity
.
Molecular Ecology
 
15
:
1419
1439
.
Wayne
R. K.
Koepfli
K.
.
1996
.
Demographic and historical effects on genetic variation in carnivores
. Pp.
453
484
in
Carnivore behavior, ecology, and evolution
  (
Gittleman
J. L.
, ed.). Vol.
2
.
Cornell University Press
,
Ithaca, New York
.
Wright
S.
1951
.
The genetical structure of populations
.
Annals of Eugenics
 
15
:
323
354
.

Author notes

Associate Editor was Rick A. Sweitzer.