Single Species as Indicators of Species Richness and ... - csusm

Single Species as Indicators of Species Richness and Composition in California Coastal Sage Scrub Birds and Small Mammals MARY K. CHASE*, WILLIAM B. KRISTAN III, ANTHONY J. LYNAM†, MARY V. PRICE, AND JOHN T. ROTENBERRY Department of Biology and Center for Conservation Biology, University of California, Riverside, CA 92521, U.S.A.

Abstract: Individual species may be useful as indicators of biodiversity if an association exists between the presence of a species and another component of biodiversity. We evaluated 40 species of birds and small mammals, including 11 species of conservation concern, as potential indicators of species richness and species composition in southern California coastal sage scrub habitats. This habitat, which is the target of largescale conservation planning, has been greatly reduced by human development and supports many plants and animals of conservation concern. We asked whether there is an association between the presence of a potential indicator species and the species richness and composition of the bird or small-mammal community in which it is found. We used point counts and live-trapping to quantify the distribution of birds and small mammals, respectively, at 155 points in 16 sites located in three counties. Of the few species we found associated with species richness, some were associated with higher species richness and others with lower richness, and species of conservation concern were not more frequently associated with species richness than were common species. Ordination analysis revealed a geographic gradient in coastal sage scrub bird and smallmammal species composition across southern California, and 18 of the species we evaluated were associated with the composition of the bird and small-mammal community in which they were found. Our results suggest that efforts to conserve bird and small-mammal biodiversity in coastal sage scrub should not focus exclusively on rare species or on locations with the highest species richness, but instead should focus on a diverse suite of species that are representative of the range of variation in communities found in coastal sage scrub habitats. Especies Individuales como Indicadores de la Riqueza y la Composición de Especies de Aves y Mamíferos Pequeños en un Matorral Costero de Salvia en California Resumen: Especies individuales pueden ser utilizadas como indicadores de la biodiversidad si existe una asociación entre la presencia de una especie y otro componente de la biodiversidad. Evaluamos 40 especies de aves y mamíferos pequeños, incluyendo 11 especies de interés para la conservación como indicadores potenciales de la riqueza y la composición de especies en hábitats costeros de matorral de salvia del sur de California. Ese hábitat, objeto de la planificación para la conservación a gran escala, ha sido ampliamente reducido debido al desarrollo humano y sirve de apoyo para muchas plantas y animales de interés para la conservación. Evaluamos si existe una asociación entre la presencia de una especie como potencial indicador y la riqueza y composición de especies de la comunidad de aves y mamíferos en las cuales se encuentra. Utilizamos puntos de conteo y trampeo de organismos vivos de aves y mamíferos pequeños respectivamente para cuantificar la distribución en 155 puntos en 16 sitios localizados en tres condados. De las pocas especies que encontramos asociadas con la riqueza de especies, algunas estuvieron asociadas con una riqueza de especies alta y otras con una riqueza de especies baja, y las especies de interés para la conservación no estu-

*email [email protected] † Current address: Wildlife Conservation Society, Box 170, Laksi, Bangkok 10210, Thailand Paper submitted June 19, 1998; revised manuscript accepted July 21, 1999.

474 Conservation Biology, Pages 474–487 Volume 14, No. 2, April 2000

Chase et al.

Coastal Sage Scrub Indicator Species

475

vieron frecuentemente más asociadas con la riqueza de especies que las especies comunes. Un análisis de ordenamiento revela un gradiente geográfico en la composición de aves y mamíferos pequeños en las comunidades costeras de matorral de salvia a lo largo del sur de California, y 18 de las especies evaluadas estuvieron asociadas con la composición de las comunidades de aves y mamíferos pequeños en las cuales fueron encontradas. Nuestros resultados sugieren que los esfuerzos para la conservación de la biodiversidad de aves y mamíferos pequeños en matorral de salvia costero no se deberían enfocar exclusivamente en las especies raras ni en las localidades con la mayor riqueza de especies, sino que deberán enfocarse en un grupo diverso de especies que son representativas del rango de variación en las comunidades que se encuentran en los hábitats costeros de matorral de salvia.

Introduction The establishment of protected natural areas is often necessary for the in situ conservation of biodiversity in threatened habitats (Soulé 1991). Because the amount of habitat that can be protected is usually limited, designers of reserve systems must prioritize the conservation value of potential reserves (Margules & Usher 1981; Usher 1986; Pressey et al. 1993; Williams et al. 1996). Biodiversity is a commonly used criterion for selecting reserves, and potential reserves can be evaluated for multiple levels of biological diversity, including genetic diversity, species diversity, and community diversity (Noss 1990). This assessment would be simplified if biologists could identify relatively easily measured indicators of biodiversity. A commonly used approach is the indicator species, defined by Landres et al. (1988) as “an organism whose characteristics, such as presence or absence, population density, dispersion, reproductive success, are used as an index of attributes too difficult, inconvenient, or expensive to measure.” Many recent studies of biodiversity indicator taxa have focused on whether patterns of species richness or rarity are correlated among taxa or whether areas with large numbers of rare species are also rich in species (e.g., Ryti 1992; Kerr 1997; Prendergast & Eversham 1997; Lawton et al. 1998; Pearson & Carroll 1998; Tardif & DesGranges 1998; Troumbis & Dimitrakopoulos 1998; also reviewed in McGeoch 1998). Less often addressed is whether the presence of an individual species in a potential reserve can serve as an indicator of species richness or species composition (Launer & Murphy 1994; Nilsson et al. 1995; Niemi et al. 1997). Although criticism of the indicator species concept has led to growing interest in habitat-based, multi-species conservation planning (Beatley 1994; Martin 1994), single species remain important foci of conservation efforts because they are easier to identify and study than other levels of biodiversity and are more likely to be protected by law (Noss 1990; Martin 1994). For an individual species to be used as an indicator, its distribution must provide relevant information, for example, about the distributions of other species. We define biodiversity indicator species as

those whose presence is correlated with high species richness or with the presence of a threatened biological community or subassociation. Coastal sage scrub habitat in Southern California provides an excellent testing ground for the use of indicator species in conservation planning. Coastal sage scrub, which has been reduced to 10–30% of its former extent by conversion for human use, supports approximately 100 animal and plant species considered rare, sensitive, threatened, or endangered by California or U.S. federal wildlife agencies (Atwood 1993; McCaull 1994). California’s Natural Community Conservation Planning (NCCP) program aims to design a reserve system to protect biodiversity in coastal sage scrub habitat while allowing economic development in areas of lower biological significance (Atwood 1993; McCaull 1994). The NCCP program has evolved largely in response to the legal protection given to one coastal sage scrub species, the California Gnatcatcher (Polioptila californica). Planning decisions have emphasized the conservation of this and other target species (Atwood 1993; State of California 1993a; Atwood & Noss 1994). Therefore, it is important to understand the relationship between the distribution of individual species and the overall patterns of biodiversity within coastal sage scrub habitat. We used data from surveys of passerine birds and small mammals to evaluate the relationship between potential indicator species and several attributes of the coastal sage scrub community in which they were found. Specifically, we asked two questions: (1) Is there an association between the presence of a potential indicator species and the species richness of the bird or smallmammal community in which it was found? This might be true if, for example, a single rare species tends to occur in species-rich sites. (2) Is there an association between the presence of a potential indicator species and the species composition of the bird and small-mammal community in which it was found? If so, and if species composition varies within coastal sage scrub habitat, then indicator species could be used to help ensure that this variation would be represented in a reserve system. In addition, we asked if patterns of species richness and composition in birds and small mammals are correlated.

Conservation Biology Volume 14, No. 2, April 2000

476


Methods Focal Species We chose two groups of taxa as potential indicators in our analyses: (1) species or subspecies of conservation concern (i.e., those identified as sensitive, threatened, or endangered, or suggested as targets of conservation planning) and (2) species occurring commonly in our samples (Table 1). The first group includes 11 taxa that have been listed as species or subspecies of special concern by the California Department of Fish and Game (State of California 1992, 1993b) and three taxa that have

Chase et al.

been suggested as foci of conservation efforts in Riverside County because they occur at low densities (Anonymous 1994). The second group includes species detected at ⬎10% and ⬍90% of our sampling points, because species found in fewer samples are less likely to be characteristic of coastal sage scrub habitat and because species found in almost all samples cannot represent within-habitat variation in species richness or composition. Study Area Coastal sage scrub is a partially drought-deciduous shrubland found in southwestern California and northwestern

Table 1. Species chosen for evaluation as potential indicators of species richness and species composition in coastal sage scrub habitat, and their frequency of occurrence among sampling points (n ⫽ 155), among sites (n ⫽ 16), and among sites with ⱖ10 points (n ⫽ 9). Number of detections Species

Species code

points

sites

California Quail (Callipepla californica) Mourning Dove (Zenaida macroura) Greater Roadrunner (Geococcyx californianus)a Anna’s Hummingbird (Calypte anna) Costa’s Hummingbird (Calypte costae) Horned Lark (Eremophila alpestris)b Western Scrub-Jay (Aphelocoma californica) Bushtit (Psaltriparus minimus) House Wren (Troglodytes aedon) Bewick’s Wren (Thryomanes bewickii) Canyon Wren (Catherpes mexicanus)a Rock Wren (Salpinctes obsoletus)a Cactus Wren (Campylorhynchus brunneicapillus)b Wrentit (Chamaea fasciata) California Gnatcatcher (Polioptila californica)b Northern Mockingbird (Mimus polyglottos) California Thrasher (Toxostoma redivivum) Loggerhead Shrike (Lanius ludovicianus)b Orange-crowned Warbler (Vermivora celata) Common Yellowthroat (Geothlypis trichas) Western Meadowlark (Sturnella neglecta) White-crowned Sparrow (Zonotrichia leucophrys) Sage Sparrow (Amphispiza belli)b Black-chinned Sparrow (Spizella atrogularis) Rufous-crowned Sparrow (Aimophila ruficeps)b Song Sparrow (Melospiza melodia) Black-headed Grosbeak (Pheucticus melanocephalus) House Finch (Carpodacus mexicanus) Lesser Goldfinch (Carduelis psaltria) Pacific kangaroo rat (Dipodomys agilis) Stephens’ kangaroo rat (Dipodomys stephensi)b Dulzura pocket mouse (Chaetodipus californicus femoralis)b San Diego pocket mouse (Chaetodipus fallax fallax)b Los Angeles pocket mouse (Perognathus longimembris brevinasus)b Dusky-footed woodrat (Neotoma fuscipes) San Diego woodrat (Neotoma lepida intermedia)b California mouse (Peromyscus californicus) Cactus mouse (Peromyscus eremicus) Deer mouse (Peromyscus maniculatus) Western harvest mouse (Reithrodontomys megalotis)

CAQU MODO GRRO ANHU COHU HOLA SCJA BUSH HOWR BEWR CANW ROWR CACW WREN CAGN NOMO CATH LOSH OCWA COYE WEME WCSP SAGS BCSP RCSP SOSP BHGR HOFI LEGO DIAG DIST CHCA CHFA PELO NEFU NELE PECA PEER PEMA REME

80 85 8 51 93 3 43 91 19 115 14 13 40 108 19 63 76 1 24 23 45 46 37 64 89 41 18 59 49 43 1 4 58 1 39 58 54 99 46 17

16 15 4 13 16 1 12 15 7 16 3 4 6 14 7 12 15 1 9 8 8 12 8 10 15 14 5 14 14 8 1 2 10 1 9 12 11 15 8 10

a b

Species suggested as foci of conservation planning (Anonymous 1994). Species or subspecies of special conservation concern (State of California 1992, 1993b).


sites with ⱖ10 points 9 9 3 8 9 1 7 8 6 9 1 2 5 8 6 9 9 1 7 6 6 8 4 6 9 8 3 9 7 4 1 1 5 1 5 7 6 9 4 8

Chase et al.

Baja California that is dominated by shrubs 0.5–2.0 m in height (Westman 1981). Plant species composition varies within this broadly defined habitat, and there are several distinct types of coastal sage scrub (Westman 1983; White & Padley 1997). Dominant shrubs include California sagebrush (Artemisia californica), black sage (Salvia mellifera), white sage (Salvia apiana), California encelia (Encelia californica), brittlebush (Encelia farinosa), and California buckwheat (Eriogonum fasciculatum; Westman 1981, 1983). Our study sites were located in Orange, Riverside, and San Diego Counties (Fig. 1). We selected sites and sampling points based on the presence of the above shrub species and the ability to secure permission to access the site. Sampling In designing our sampling, we were faced with the classic dilemma of the tradeoff between extensive versus intensive sampling. Following recommendations from Ralph et al. (1995), and considering our need to survey a broad region, we traded exhaustive sampling at each point for an increase in the number of points and sites sampled. Therefore, our sampling was not meant to provide an

Figure 1. Location of study sites in California. Site names (and number of sampling points) are as follows: 1, University of California, Riverside (5); 2, Sycamore Canyon Park (5); 3, Lake Perris State Recreational Area (20); 4, Motte Rimrock Reserve (10); 5, Kabian Park (10); 6, Santa Margarita Ecological Reserve (5); 7, Pamo Valley (4); 8, Black Canyon (5); 9, Wild Animal Park (8); 10, Sweetwater River National Wildlife Refuge (11); 11, Point Loma (10); 12, Torrey Pines State Park (8); 13, Rancho Mission Viejo (10); 14, Starr Ranch (18); 15, Sycamore Hills (10); 16, Limestone Canyon (16).


477

exhaustive count of the number of species present but rather to provide a relative index of species richness and composition. We established 155 sampling points at 16 sites (Fig. 1). Within each site, we placed points systematically in coastal sage scrub habitat with the following constraints: points were at least 50 m from roads, trails, and ecotones with other habitats, and points were at least 200 m apart to avoid sampling the same individual birds at more than one point. A maximum of 20 points was established per site. In sites that were too small to contain 20 points, we established as many points as possible. We sampled birds by conducting two, 5-minute unlimited-radius counts at each point (Ralph et al. 1995), one early in the breeding season (19 March–1 May 1996) and one later in the season (3–25 May 1996). All birds detected from the point center were included, except for those located in adjacent nonscrub habitat types or detected only flying overhead. Second visits to each site were made in the same order as first visits to ensure that each site was sampled in both early and late spring. To avoid observer bias, point counts were conducted by four experienced observers, and each point was sampled by a different observer on the first and second visit. To avoid potential bias due to time of day and weather conditions, we conducted point counts between sunrise and 5 hours after sunrise on mornings with no rain or strong wind, and we reversed the order in which points were sampled within each site between the first and second visits to the site. Bird species not well sampled by point counts include flocking species and nocturnal species (Ralph et al. 1993). Small mammals were sampled over 3 consecutive days of trapping at each point. We used 16 Sherman live-traps spaced 8 m apart in a 4 ⫻ 4 grid. Mammal trapping grids were centered within 15 m of the point from which birds were sampled. Points were sampled between 7 May 1996 and 27 June 1996. Three-day trapping periods were chosen because longer-term trapping at a subset of points showed that 90% of all species detected at each trapping point with a 7-day trapping period were detected by the third day (M.V.P., unpublished data). Because small-mammal activity can be affected by moonlight (Price et al. 1984), trapping was not done for 2 days before and after a full moon. Traps were baited with a mixture of rolled oats, peanut butter, and corn syrup. Mammals were identified to species using keys derived from Ingles (1965) and Jameson and Peeters (1988). Sherman traps of the size used in this study (8 ⫻ 9 ⫻ 23 cm) are effective for detecting most groups of nonvolant small mammals (⬍500 g; mainly rodents) that potentially occur in coastal sage scrub habitats, including murids, heteromyids, cricetines, and microtines (M’Closkey 1972; Meserve 1976; Price & Waser 1984; Salvioni & Lidicker 1995). Our sampling method, however, was not geared toward several species of small mammals that po-


478


tentially occur in our study sites: gray shrew (Notiosorex crawfordi), ornate shrew (Sorex ornatus), broad-footed mole (Scapanus latimanus), valley pocket gopher (Thomomys bottae), and California ground squirrel (Spermophilus beecheyi ). Analysis Many studies of biodiversity focus on the species richness of entire study sites, which is known to be influenced by site characteristics such as area and habitat heterogeneity (e.g., Soulé et al. 1988). To focus on patterns of alpha diversity in coastal sage scrub rather than patterns associated with variation in area among sites, we instead analyzed species richness and composition at individual sampling points. We evaluated (1) whether the presence of a potential indicator at certain points was associated with the species richness or composition at those points, and (2) whether the presence of a potential indicator at certain sites was associated with the richness or composition averaged over all points within those sites. Species Richness at Point and Site Scales To determine if the presence of a focal species was an indicator of species richness, we first compared species richness among individual sampling points (n ⫽ 155). We estimated species richness as the number of species detected per point over the two sampling intervals, excluding the focal species that was the subject of the test. Bird and mammal species richness were analyzed separately. Because our analyses were repeated for each focal species, we used the standard Bonferroni correction for multiple tests to keep the table-wide Type I error rate at 0.05. For all point-scale analyses, we tested 37 focal species (␣ ⫽ 0.0014). Randomization tests (Sokal & Rohlf 1995) were used to compare mean species richness at points where a focal species was detected to the expected species richness at a random sample of points. We used this as an alternative to a parametric test because the number of points at which a focal species is found can be much smaller (or much larger) than the number of points where it is not found (e.g., California Gnatcatchers were detected at 19 out of 155 points). In this procedure, the mean species richness (excluding the focal species) was calculated from a random sample of points drawn from the 155 observed richness values, where the size of the random sample equaled the number of points at which the focal species was detected. This procedure was iterated 10,000 times to create an expected distribution of sample means. The observed mean species richness at points where the focal species was detected was compared to this distribution, and its statistical significance (departure from the expected mean) was determined.


Chase et al.

We also compared species richness at sites where a focal species was detected to sites where it was not detected. The mean number of species detected per point differs significantly among sites (Chase et al. 1998). We wished to determine whether this variation among sites in the number of species detected per point was correlated with the distribution of any of our focal species. To maintain relatively uniform sample sizes across sites and to avoid sites with a small number of points, we restricted this analysis to sites with 10 or more points (n ⫽ 9). We did not test focal species that were detected at only one site or that were absent from only one site (Table 1). We tested 19 focal species (␣ ⫽ 0.0026). To compare mean species richness between sites with and without each of the focal species, we used a one-way analysis of variance (ANOVA). The ANOVAs were performed with STATA statistical software (Stata Corporation 1997). Species Composition Species composition was analyzed by detrended correspondence analysis (DCA) based on species presence or absence at each point and was conducted with PC-ORD statistical software (McCune & Mefford 1995). The DCA is an ordination technique that quantifies the relationship among a set of points based on the similarity of their species composition, and the relationship among species based on the similarity of their distribution among points (Gauch 1982). Points and species are ordered on axes so that points with similar species composition will have similar axis scores, and species with similar patterns of distribution will also have similar scores. Thus, the scores of points on ordination axes can be used as an index of the species composition at those points, and the mean scores of points within a site can be used as an index of the species composition at a site. A further advantage is that DCA axes are linearized such that beta diversity (the compositional difference between two points) is constant. Thus, a unit difference between two points at one end of a gradient represents the same compositional change as a unit difference between two points at the other end. Patterns of species composition in birds and mammals were similar when analyzed separately, so we present results from a single DCA analysis of both bird and mammal species. We compared species composition at points where focal species were detected to the expected species composition at a random sample of points using a randomization test of point scores on the first DCA axis. With the Bonferroni correction, the alpha level for each test was 0.0014. To compare species composition between sites with and without each of the focal species, we conducted a one-way ANOVA on the mean of the DCA axis1 point scores for each site. We did not test focal species detected at only one site or absent from only one site. The alpha level for each test was 0.0026. We examined

Chase et al.

the relationship between species composition and geographical location of our sampling points with a multiple linear regression of the DCA axis-1 point scores, using as independent variables the Universal Transverse Mercator (UTM) eastings and northings and the interaction of eastings and northings. For illustrative purposes, geographic contours of DCA axis-1 scores were estimated by an inverse distance weighting method ( Jandel Scientific 1994). In this method, values at all sampled points are used in interpolating contour intervals but are weighted inversely to their distance to each map point.


479

runner (mean ⫽ 3.6 mammal species at three sites where they were found; mean ⫽ 2.4 at sites where they were not found; F ⫽ 4.28, p ⫽ 0.08), San Diego pocket mice (mean ⫽ 10.9 bird species at five sites where they were found; mean ⫽ 13.1 at sites where they were not found; F ⫽ 7.54, p ⫽ 0.03), dusky-footed woodrat (mean ⫽ 1.9 mammal species at five sites where they were found; mean ⫽ 3.3 at sites where they were not found; F ⫽ 5.65, p ⫽ 0.05), California mouse (mean ⫽ 1.9 mammal species at six sites where they were found; mean ⫽ 3.6 at sites where they were not found; F ⫽ 9.14, p ⫽ 0.02). Species Composition

Results Species Richness On average, 12 bird species (range, 5–19) and 3 mammal species (range, 0–6) were detected at each sampling point. Bird and mammal species richness were not correlated among points (r ⫽ 0.05, p ⫽ 0.64). Two species of conservation concern, the Loggerhead Shrike and Stephens’ kangaroo rat, were detected at only one sampling point and thus could not be used in the point-scale analyses as focal species. Of the other potential indicator species, only Sage Sparrows were associated with species richness at points where they were detected (n ⫽ 37 points). At these points, the number of co-occurring bird species was 10.4 on average, compared to 11.7 species at randomly sampled points ( p ⬍ 0.0014). The mean mammal species richness at points where Sage Sparrows were detected was 3.6, compared with 2.8 at randomly sampled points ( p ⬍ 0.0014). The mean number of bird species detected per point also varied among sites, ranging from 9.0 at Point Loma to 13.4 at Limestone Canyon. The mean number of mammal species detected per point ranged from 1.8 at Point Loma and Rancho Mission Viejo to 4.8 at Motte Rimrock. Bird and mammal species richness were not correlated among sites (r ⫽ 0.08, p ⫽ 0.84). Nine potential indicator species were detected at all nine sites and therefore could not be analyzed as indicators at the site scale; also not analyzed at the site scale were 12 species found at only one or at all but one site (Table 1). Of the remaining 19 species, only Common Yellowthroats were associated with higher bird species richness at sites where they were found (mean ⫽ 12.6 at three sites where Common Yellowthroats were found; mean ⫽ 9.9 at sites where they were not found; F ⫽ 38.77, p ⫽ 0.0004). The Western Scrub-Jay was associated with lower mammal species richness (mean ⫽ 2.3 at seven sites where Western Scrub-Jays were found; mean ⫽ 4.3 at sites where they were not found; F ⫽ 26.38, p ⫽ 0.0013). Although the results were not statistically significant, several other species tended to be associated with species richness at sites where they were found: Greater Road-

The first two DCA axes explained 44.9% of the covariation in sample and species dispersion (axis-1 eigenvalue ⫽ 0.313; axis-2 eigenvalue ⫽ 0.136). The distribution of points (classed by their approximate location) with respect to DCA axis 1 suggests that this axis might represent geographic variation in species composition (Fig. 2). This is confirmed by a statistically significant multiple regression of DCA axis-1 point scores on UTM east and north coordinates and their interaction (F 3,151 ⫽ 84.18, p ⬍ 0.001). This regression accounted for 62.6% of the total variation in DCA axis-1 scores. Each term in the regression equation was significant (all p ⬍ 0.01). The geographical nature of this variation was revealed when we plotted contours of DCA axis-1 scores on a regional map (Fig. 3). Of the 37 species we evaluated, seven birds and three mammals were significantly associated with points characterized by an “inland” species association, whereas six birds and two mammals were found at points with a more “coastal” species association (Table 2; Fig. 4). The remaining 19 species occurred at points with a species composition that was not significantly different from

Figure 2. Detrended correspondence analysis (DCA) ordination of 155 Californian coastal sage scrub sampling points based on presence or absence of 59 bird and 13 mammal species. Axes are DCA scores ⫻ 100.


480


Chase et al.

cluded California Towhee, Spotted Towhee, Costa’s Hummingbird, Bewick’s Wren, and California Quail. Although the species composition of sites in San Diego county was generally intermediate to that of coastal (Orange county) and inland (Riverside County) sites, we did not find a group of species that was associated primarily with the center of the gradient (Table 3). Instead, San Diego sites (sites 7–12 in Table 3) tended to contain both species characteristic of inland sites, such as Sage Sparrow and San Diego pocket mouse, and species characteristic of coastal sites, such as Wrentit and California mouse.

Discussion Species Richness

Figure 3. Contours of detrended correspondence analysis (DCA) scores of 155 Californian coastal sage scrub sampling points. Points overlap on map due to their proximity to one another within sites. Numbers are DCA scores ⫻ 100.

that of randomly sampled points. Of the 14 species of conservation concern, 9 were clearly associated with inland points, 2 were associated with coastal points, and 3 were intermediate (Fig. 5). Those species that occurred at only one point (LOSH, DIST, PELO) were “peripheral” in their distribution with respect to the ordination, suggesting that their apparent rarity was due to our sampling in a habitat that was marginal for them. Although Canyon Wrens and Horned Larks were indicators of species composition at a point scale, they could not be analyzed at the site scale because they were each found at only one site. Of the 19 species tested at the site scale, one bird and two mammals were significantly associated with sites with a more coastal species composition, whereas two birds and two mammals were associated with sites with a more inland species composition (Table 2; Fig 4). The gradient in species composition among sites can be described in more detail when the species-by-site matrix is ordered by DCA axis-1 scores, thus revealing where each species occurs along the gradient (Table 3). For some species the pattern was strong: three species associated with the inland community (Sage Sparrow, Pacific kangaroo rat, and deer mouse) were detected only at sites lacking the two species associated most strongly with coastal sites (Cactus Wren and dusky-footed woodrat), and vice versa. In addition, we identified a core group of species found at all the sites we sampled (Table 3), which in-


Three of the 37 species we tested were significantly associated with species richness, and four additional species showed marginally nonsignificant associations. In three cases the presence of an individual species was associated with higher species richness, whereas in four cases it was associated with lower species richness. Species of conservation concern were not more frequently associated with species richness than were common species. Therefore, species of conservation concern cannot be assumed to be indicators of “hotspots” of bird and small-mammal richness in coastal sage scrub. Similarly, two recent studies show little geographical correspondence between locations with high numbers of rare taxa and locations with high species richness (birds, Williams et al. 1996; birds, liverworts, and aquatic plants, Prendergast et al. 1993). In contrast, Debinski and Brussard (1994) found an overlap between sites that support high species diversity and sites that support rare species of birds and butterflies. The usefulness of rare species as indicators of species richness is likely to vary among geographic locations and spatial scales (McGeoch 1998). In several cases the presence of a bird species was associated with small-mammal richness or vice versa, suggesting that single species could potentially serve as cross-taxa indicators. But the overall lack of correlation between bird and mammal species richness in our study suggests that there is no substitute for targeting multiple taxa in conservation planning. Similarly, several other studies show few consistent relationships between species richness in different taxa (butterflies and plants, Kremen 1992; six vertebrate taxa, Lombard 1995; birds, butterflies, and dragonflies and damselflies, Prendergast & Eversham 1997; birds, butterflies, and six invertebrate taxa, Lawton et al. 1998). One study, which took place on a global scale, shows a correlation among species richness in tiger beetles, birds, and butterflies (Pearson & Cassola 1992). These results also suggest that the useful-

Chase et al.

Table 2.


481

Results of analyses testing single species as potential indicators of bird and small mammal species composition.a Point scale

Speciesb Black-chinned Sparrow Cactus Wren f Canyon Wren f Greater Roadrunner f Horned Lark f House Wren Orange-crowned Warbler Rufous-crowned Sparrow f Rock Wren f Sage Sparrow f Western Scrub-Jay Western Meadowlark Wrentit Pacific kangaroo rat San Diego pocket mouse f Dusky-footed woodrat California mouse Deer mouse

Site scale

mean DCA axis-1 score of points where detected c

pd

84.6 65.6 199.0 179.9 249.7 46.8 65.5 131.4 186.9 179.5 56.3 172.4 73.7 173.3 161.2 54.7 57.0 173.4

⬍0.0002 ⬍0.0002 ⬍0.0002 ⬍0.0008 ⬍0.0002 ⬍0.0002 ⬍0.0003 ⬍0.0006 ⬍0.0002 ⬍0.0001 ⬍0.0001 ⬍0.0001 ⬍0.0001 ⬍0.0001 ⬍0.0001 ⬍0.0001 ⬍0.0001 ⬍0.0001

mean DCA axis-1 score for sites where detected

mean DCA axis-1 score for sites where not detected

pe

108.5 58.0

100.0 166.0

⬎0.8 ⬍0.002*

187.2

65.4

⬍0.0001*

78.8 99.0

160.5 130.3

⬎0.06 ⬎0.5

121.4 166.0 85.1 119.9

101.6 58.0 179.3 78.1

⬎0.07 ⬍0.002* ⬎0.05 ⬎0.3

166.0 148.8 58.0 65.4 166.0

58.0 52.5 166.0 187.2 58.0

⬍0.002* ⬎0.01 ⬍0.002* ⬍0.0001* ⬍0.002*

a Points where potential indicator species were detected were compared to random points (n ⫽ 155 points), and points within sites where potential indicator species were detected were compared to points within sites where they were not detected (n ⫽ 9 sites). b Only species with significant associations at the point or site scale are included. See Table 1 for scientific names. c Mean DCA axis-1 score of all points, 108.8. d Values from randomization test; all are statistically significant. e Values from analysis of variance; statistically significant values are marked with an asterisk. f Species of conservation concern.

ness of biodiversity indicators varies among spatial scales (Flather et al. 1997; Reid 1998). Species Composition Ordination analysis (DCA) revealed a geographic gradient bird and small-mammal species composition in coastal sage scrub across southern California (Figs. 2 & 3). A similar coastal-inland gradient has been found in the coastal sage scrub plant community at our sampling points ( J.T.R., unpublished data). Also, classification analyses of coastal sage scrub vegetation types (Westman 1983) distinguish between a coastal species association (Diegan) and a higher-elevation inland species association (Riversidian). Underlying the vegetation gradient is a strong climatic gradient, with the inland area experiencing higher evapotranspiration during the summer than the coastal area (Westman 1981). If, as this suggests, parallel geographic variation exists in the species composition of bird, small-mammal, and plant communities in coastal sage scrub, then individual species or groups of species in one taxon might be useful as indicators of species assemblages in other taxa. Eighteen of the species we evaluated were associated with the composition of the bird and small-mammal community in which they were found (Table 2). In particular, the Cactus Wren, dusky-footed woodrat, and Cal-

ifornia mouse were strongly associated with a more coastal-type bird and small-mammal community at both the point and site scales. Of these three, the Cactus Wren has two advantages as a biodiversity indicator species: it is more easily detected than the small-mammal species, and it is a species of special conservation concern. The Greater Roadrunner, Sage Sparrow, Pacific kangaroo rat, and deer mouse were the species most strongly associated with a more inland-type bird and small-mammal community at both spatial scales. The Sage Sparrow, a species of conservation concern, is also associated with low bird species richness and high mammal species richness. Several species significantly associated with species composition at the point scale were not significant at the site scale (Table 2). For some species, such as the Western Scrub-Jay and the San Diego pocket mouse, this may be due to the reduced power of the site-scale tests. Other species, however, such as the Black-chinned Sparrow and the Rufous-crowned Sparrow, appear to have been associated with a distinct species assemblage at the point scale but not at the larger spatial scale of sites. These species were widely distributed among sites, but within sites they occurred primarily in a subset of points with a distinct species assemblage. For example, although the Rufous-crowned Sparrow was found at 15 out of 16 sites, it was strongly associated with points


482


Figure 4. Detrended correspondence analysis (DCA) ordination of bird and small-mammal species with statistically significant indicator species plotted. The ordination axes are the same as in Fig. 2. Species label is centered over species score. Asterisks indicate species of conservation concern. Species in italics were significant at the site scale as well as the point scale; other species significant only at point scale. See Table 1 and Appendix 1 for species codes.

where the species composition was more typical of inland sites. This suggests that the variation we found in bird and mammal species composition may be associated with variation in habitat characteristics within sites, as well as with larger-scale geographic variation. This is consistent with what is known about variation in the coastal sage scrub plant community. For example, DeSimone and Burk (1992) found small-scale variation in plant species associations within the Starr Ranch site that paralleled regional-scale variation. The two mammals associated with the coastal species assemblage, the dusky-footed woodrat and California mouse, were also associated with low small-mammal species richness, and two of the species associated with the inland species assemblage, the Sage Sparrow and San Diego pocket mouse, were associated with low bird-species richness. This suggests that a reserve system encompassing the range of variation in coastal sage scrub bird and small-mammal communities must include sites with both high and low species richness. Other Uses of Indicators Single species may also be used as indicators of environmental change—for example, to detect environmental contamination, to evaluate the effects of changes in habitats over time, or to monitor population trends of other species (Landres et al. 1988; Temple & Wiens 1989; Noss 1990), and for this purpose the evaluation criteria may differ (e.g., Kremen 1992; Debinski & Brussard


Chase et al.

Figure 5. Detrended correspondence analysis (DCA) ordination of bird and small-mammal species with species of conservation concern plotted. The ordination axes are the same as in Fig. 2. Species label is centered over species score (except for NELE). Species in italics were detected at only a single point. See Table 1 and Appendix 1 for species codes.

1994; Lawton et al. 1998; McGeoch 1998; Rodríguez et al. 1998). Several of the species associated with species composition in our study may also be useful as indicators of environmental change due to their sensitivity to habitat fragmentation (Cactus Wren, Soulé et al. 1988; Sage Sparrows, Black-chinned Sparrows, Rufous-crowned Sparrows, Bolger et al. 1997). An additional group of species that may warrant special conservation effort are umbrella species—“species with large area requirements, which if given sufficient protected habitat area, will bring many other species under their protection” (Noss 1990)—although large area requirements per se do not ensure that co-occurring species will benefit from the protection of a single species (Landres et al. 1988). Of the species we evaluated, Greater Roadrunners are potential umbrella species within the inland-type species assemblage because of their large area requirements (Hughes 1996). The California Gnatcatcher, which has been a major focus of conservation efforts in this region, was not an indicator either of community composition or of species richness within coastal sage scrub, although it may still be an indicator of environmental change because of its sensitivity to habitat fragmentation (Soulé et al. 1988) or an umbrella species because of its large territory requirement (Fleury et al. 1998; Preston et al. 1998). Caveats Any sample that is not exhaustive will not detect all species present and thus will always underestimate true species richness at a point or other sampling unit. In our

Chase et al.

Table 3.


Occurrences of species among study sites.* Site

Species WAVI WIWR NUWO LASP PSFL HOOR OATI SAPH HOWR ACWO PHAI NEFU SCJA GRSP PECA RCKI CHCA BLPH CACW WIWA OCWA BHGR LAZB WREN NOFL ATFL BCSP CATH MODO SPTO AMGO COHU ANHU COBU NELE REME CAKI COYE BEWR LEGO CALT PEER NOMO CAQU BGGN EUST SOSP HOFI CAGN RCSP WCSP NOOR WEKI CHFA BHCO DIAG PEMA GRRO SAGS BRBL WEME

483

13

14

15

16

12

6

8

11

10

7

9

4

3

2

5

1

1 1 1 1 1

1

1 1 1 1 1 1 1

1 1 1 1 1 1 1 1 1 1

1 1

1

1 1 1 1 1 1 1

1

1

1 1

1

1 1 1 1

1 1 1

1 1 1 1 1 1

1 1 1 1 1 1

1 1 1 1

1 1 1 1 1

1

1

1 1 1

1 1 1

1 1 1 1 1

1 1 1 1 1

1 1 1 1 1

1 1 1 1 1

1 1

1 1 1 1 1 1 1 1

1 1 1 1 1 1 1

1 1 1 1 1 1 1 1

1 1

1

1 1 1

1 1 1 1

1 1 1 1 1

1 1 1 1 1

1 1

1 1

1

1

1

1 1 1

1 1 1

1

1 1

1

1 1

1

1 1 1 1 1 1 1

1 1 1 1 1

1 1

1

1 1 1 1

1 1 1 1

1 1 1 1

1

1 1

1 1

1

1

1 1 1 1

1

1 1 1 1 1 1 1

1 1 1 1 1 1 1

1 1 1 1 1 1 1 1

1 1 1 1

1 1 1 1

1 1 1 1

1 1 1

1 1 1 1 1 1 1

1 1 1 1

1 1 1 1

1

1

1 1 1

1 1

1

1

1

1 1 1 1 1 1 1 1 1

1 1 1 1 1 1 1 1 1

1 1

1

1

1 1 1

1

1 1

1 1 1 1

1 1 1 1

1 1 1 1

1 1 1

1 1 1

1 1 1 1

1

1 1 1 1 1

1 1 1 1 1

1 1

1 1 1 1 1 1 1 1 1 1 1

1 1 1 1 1

1 1 1 1 1 1 1 1 1 1

1

1 1 1

1 1 1 1 1 1 1

1 1 1 1 1 1

1 1

1 1 1 1 1

1 1 1 1 1

1 1 1 1 1 1 1

1 1

1 1 1 1 1

1

1

1

1

1

1 1

1

1 1

1

1

1 1 1 1

1 1 1 1

1 1 1 1

1

1

1 1 1 1

1 1 1 1 1 1 1

1 1 1 1 1 1 1

1 1 1 1

1

1

1

1

1 1

1 1

1 1

1

1 1 1 1 1 1

1

1 1 1 1 1 1 1 1 1 1 1 1 1

1 1

1

1 1

1 1 1 1 1 1

Species score ⫺262 ⫺262 ⫺214 ⫺196 ⫺190 ⫺134 ⫺132 ⫺125 ⫺108 ⫺103 ⫺87 ⫺85 ⫺83 ⫺79 ⫺78 ⫺59 ⫺59 ⫺54 ⫺51 ⫺48 ⫺43 ⫺42 ⫺24 ⫺20 ⫺16 ⫺10 19 62 79 82 87 92 93 93 95 97 107 107 123 124 125 131 133 144 152 157 159 179 208 215 273 277 291 292 297 303 303 310 311 312 312 continued


484

Table 3.


Chase et al.

(continued) Site

Species ROWR RWBL CANW DEJU PELO MUMU MICA SAVS DIST HOLA LOSH Site Scores

13

14

15

16

12

6

8

11

10

7

9

4

3

2

5

1

1 1 1

1

1

1

1 1

1 1 1

1 1 1 40

52

58

60

67

72

75

80

102

109

111

176

182

198

1 1 203

Species score 318 320 330 330 331 332 333 339 349 382 438

205

*Species are ordered by their DCA axis-1 scores (⫻100), and sites are ordered by the mean axis-1 sample scores of their points (⫻100). Site numbers as in Fig. 1; species codes as in Table 1 and Appendix 1.

analyses, however, we are concerned with comparative rather than absolute species richness. For our analyses to be biased with respect to richness, the species accumulation rates—the number of species added at a point as survey effort increases—would have to differ systematically among points, such that additional sampling would change the rankings of points based on the number of species detected. This is unlikely in our surveys because our sampling points were intentionally located in similar coastal sage scrub vegetation throughout the survey region. Another drawback of nonexhaustive sampling is that potential indicator species that are inconspicuous may have been missed at individual points. Again, this would introduce a bias in our analyses only if the presence of a potential indicator was less likely to be detected at points with a nonrandom species richness or composition. Even if there is no systematic bias, however, random between-point error in estimating richness will remain in any nonexhaustive sampling effort. This source of variation reduced the power of our statistical tests and hence reduced our ability to identify species that are weak indicators (in statistical parlance, species associated with a low effect size). It did not reduce our confidence in those species that the statistical tests had already identified. Increasing the sampling effort at a point, by reducing error variance, would likely yield additional statistically significant indicator species. One might argue, however, that because these additional species are associated with small effect sizes, their usefulness in a planning context is diminished compared to that of the others already identified. The issue of “undersampling” occurs with respect to community composition as well. Again, however, unless there is some systematic bias across sites (e.g., if lessconspicuous species, which are undersampled in less than exhaustive surveys, occur nonrandomly with respect to species composition), our results should be robust. We


also have an independent assessment that indicates that our estimate of community composition at a point is robust. In the year preceding the collection of the data reported here (1995), we initiated a pilot effort in which we sampled birds and small mammals using the same techniques at 78 points distributed among eight sites. Of those 78 points, 65 were in common with the 155 reported here. We performed a DCA of those 78 points and then correlated the 1995 DCA axis-1 scores of the 65 points in common with their scores on the 1996 DCA axis 1. If point sampling inadequately captures composition—if one were simply collecting a small, random subsample of the species occurring at each point—then one would expect little association between the two scores. Instead, the correlation we found was highly statistically significant (r ⫽ 0.88, df ⫽ 63, p ⬍ 0.0001), implying a high similarity of species composition between years. Conclusion The presence of some individual species may serve as indicators of the overall species composition of birds and small mammals in a coastal sage scrub habitat patch but may say less about its species richness. Therefore, to allow the most effective use of individual species as targets of conservation planning in coastal sage scrub, the list of species should at least include those from both ends of the geographic spectrum of species composition. Conservation efforts are particularly needed in inland coastal sage scrub where the habitat is threatened by air pollution and grazing, as well as by the more widespread threats of urbanization, increasing fire frequency, and invasion by exotic plants (Westman 1987; O’Leary 1990; Minnich & Dezzani 1998). Ideally, future evaluations of indicator species in coastal sage scrub would include additional taxa, such as the bird and mammal species not sampled in our study and herptiles, invertebrates, and plants. Even if certain individual species are

Chase et al.

found to be reliable indicators of biodiversity, however, more study is needed to show whether the protection of an indicator will necessarily result in the protection of other species sharing the same habitat. Efforts to conserve the coastal sage scrub community should not focus exclusively on rare species or on locations with the highest species richness but instead should focus on a diverse suite of species representative of the range of variation found in coastal sage scrub habitats.

Acknowledgments This study was funded by the California Department of Fish and Game and the U.S. Geological Survey, Biological Resources Division. We thank W. Tippitts, P. Stine, and S. Viers for administrative assistance. We are grateful for the excellent field work of P. Addison, D. Kristan, M. Misenhelter, M. A. Patten, L. Pagni, J. Ruvinsky, and P. Wellburn. We thank M. Bryant for writing the program for the randomization test. Comments on an earlier version of the manuscript were provided by K. Ellison, M. Patten, M. Misenhelter, S. Henderson, K. Preston, S. Hejl, and two anonymous reviewers. We thank the following individuals and organizations for providing permission for us to work on lands they manage and/or for providing logistical help: B. Carlson, P. DeSimone, G. Hund, D. Kamata, D. Lydy, M. Asam, L. Munoz, J. Opdycke, T. Smith, M. Sanderson, S. Shapiro, S. Weber, M. Wells, A. Yuen, the U.S. Forest Service, the Irvine Company, and the Rancho Mission Viejo Company.

Literature Cited Anonymous. 1994. Multi-species habitat conservation plan, California Gnatcatcher-Sage Scrub Conservation Plan information collection and evaluation. Western Riverside Habitat Consortium, Riverside, California. Atwood, J. L. 1993. California Gnatcatchers and coastal sage scrub: the biological basis for endangered species listing. Pages 149–166 in J. E. Keeley, editor. Interface between ecology and land development in California. Southern California Academy of Sciences, Los Angeles. Atwood, J. L., and R. F. Noss. 1994. Gnatcatchers and development: a “train wreck” avoided? Illahee 2:123–130. Beatley, T. 1994. Habitat conservation planning: endangered species and urban growth. University of Texas Press, Austin. Bolger, D. T., T. A. Scott, and J. T. Rotenberry. 1997. Breeding bird abundance in an urbanizing landscape in coastal southern California. Conservation Biology 11:406–421. Chase, M. K., J. T. Rotenberry, and M. D. Misenhelter. 1998. Is the California Gnatcatcher an indicator of bird-species richness in coastal sage scrub? Western Birds 29:468–474. Debinski, D. M., and P. F. Brussard. 1994. Using biodiversity data to assess species-habitat relationships in Glacier National Park, Montana. Ecological Applications 4:833–843. DeSimone, S. A., and J. H. Burk. 1992. Local variation in floristics and distributional factors in Californian coastal sage scrub. Madroño 39:170–188.


485

Flather, C. H., K. R. Wilson, D. J. Dean, and W. C. McComb. 1997. Identifying gaps in conservation networks: of indicators and uncertainty in geographic-based analyses. Ecological Applications 7:531–542. Fleury, S. A., P. J. Mock, and J. F. O’Leary. 1998. Is the California Gnatcatcher a good umbrella species? Western Birds 29:453–467. Gauch, H. G. 1982. Multivariate analysis in community ecology. Cambridge University Press, Cambridge, United Kingdom. Hughes, J. M. 1996. Greater Roadrunner (Geococcyx californianus). No. 244 in A. Poole and F. Gill, editors. The birds of North America. The Academy of Natural Sciences, Philadelphia, Pennsylvania, and the American Ornithologists’ Union, Washington, D.C. Ingles, L. G. 1965. Mammals of the Pacific states. Stanford University Press, Stanford, California. Jameson, E. W. J., and H. J. Peeters. 1988. California mammals. University of California Press, Berkeley. Jandel Scientific. 1994. SigmaPlot. Version 2.0. User’s manual. Jandel Corp., San Rafael, California. Kerr, J. T. 1997. Species richness, endemism, and the choice of areas for conservation. Conservation Biology 11:1094–1100. Kremen, C. 1992. Assessing the indicator properties of species assemblages for natural areas monitoring. Ecological Applications 2: 203–217. Landres, P. B., J. Verner, and J. W. Thomas. 1988. Ecological uses of vertebrate indicator species: a critique. Conservation Biology 2: 316–328. Launer, A. E., and D. D. Murphy. 1994. Umbrella species and the conservation of habitat fragments: a case of a threatened butterfly and a vanishing grassland ecosystem. Biological Conservation 69:145–153. Lawton, J. H., D. E. Bignell, B. Bolton, G. F. Bloemers, P. Eggleton, P. M. Hammond, M. Hodda, R. D. Holt, T. B. Larsen, N. A. Mawdsley, N. E. Stork, D. S. Srivastava, and A. D. Watt. 1998. Biodiversity inventories, indicator taxa and the effects of habitat modification in tropical forest. Nature 391:72–76. Lombard, A. T. 1995. The problems with multi-species conservation: do hotspots, ideal reserves and existing reserves coincide? South African Journal of Zoology 30:145–163. Margules, C., and M. B. Usher. 1981. Criteria used in assessing wildlife conservation potential: a review. Biological Conservation 21:79–109. Martin, C. M. 1994. Recovering endangered species and restoring ecosystems: conservation planning for the twenty-first century in the United States. Ibis 137:S198–S203. McCaull, J. 1994. The Natural Community Conservation Planning Program and the coastal sage scrub ecosystem of southern California. Pages 281–292 in R. E. Grumbine, editor. Environmental policy and biodiversity. Island Press, Washington, D.C. McCune, B., and M. J. Mefford. 1995. PC-ORD. Multivariate analysis of ecological data. Version 2.0. MjM Software Design, Gleneden Beach, Oregon. McGeoch, M. A. 1998. The selection, testing and application of terrestrial insects as bioindicators. Biological Reviews 73:181–201. M’Closkey, R. T. 1972. Temporal changes in populations and species diversity in a California rodent community. Journal of Mammalogy 53:657–676. Meserve, P. 1976. Habitat and resource utilization by rodents of a California coastal sage scrub community. Journal of Animal Ecology 45: 647–666. Minnich, R. A., and R. J. Dezzani. 1998. Historical decline of coastal sage scrub in the Riverside-Perris plain, California. Western Birds 29:366–391. Niemi, G. J., J. M. Hanowski, A. R. Lima, T. Nicholls, and N. Weiland. 1997. A critical analysis on the use of indicator species in management. Journal of Wildlife Management 61:1240–1251. Nilsson, S. G., U. Arup, R. Baranowski, and S. Ekman. 1995. Treedependent lichens and beetles as indicators in conservation forests. Conservation Biology 9:1208–1215. Noss, R. F. 1990. Indicators for monitoring biodiversity: a hierarchical approach. Conservation Biology 4:355–364.


486


O’Leary, J. F. 1990. Californian coastal sage scrub: general characteristics and considerations for biological conservation. Pages 24–41 in A. A. Schoenherr, editor. Endangered plant communities of Southern California. Proceedings of the 15th annual symposium. Special publication 3. Southern California Botanists, Claremont, California. Pearson, D. L., and S. S. Carroll. 1998. Global patterns of species richness: spatial models for conservation planning using bioindicator and precipitation data. Conservation Biology 12:809–821. Pearson, D. L., and F. Cassola. 1992. World-wide species richness patterns of tiger beetles (Coleoptera: Cicindelidae): indicator taxon for biodiversity and conservation studies. Conservation Biology 6: 376–391. Prendergast, J. R., and B. C. Eversham. 1997. Species richness covariance in higher taxa: empirical tests of the biodiversity indicator concept. Ecography 20:210–216. Prendergast, J. R., R. M. Quinn, J. H. Lawton, B. C. Eversham, and D. W. Gibbons. 1993. Rare species, the coincidence of diversity hotspots and conservation strategies. Nature 365:335–337. Pressey, R. L., C. J. Humphries, C. R. Margules, R. I. Vane-Wright, and P. H. Williams. 1993. Beyond opportunism: key principles for systematic reserve selection. Trends in Ecology and Evolution 8:124–128. Preston, K. L., P. J. Mock, M. A. Grishaver, E. A. Bailey, and D. F. King. 1998. California Gnatcatcher territorial behavior. Western Birds 29: 242–257. Price, M. V., and N. M. Waser. 1984. On the relative abundance of species: post-fire changes in a coastal sage scrub community. Ecology 65:1161–1169. Price, M. V., N. M. Waser, and T. A. Bass. 1984. Effects of moonlight on microhabitat use by desert rodents. Journal of Mammalogy 65:353–356. Ralph, C. J., G. R. Geupel, P. Pyle, T. E. Martin, and D. F. DeSante. 1993. Handbook of field methods for monitoring landbirds. General technical report PSW-GTR-144. U.S. Forest Service, Pacific Research Station, Albany, California. Ralph, C. J., J. R. Sauer, and S. Droege, technical editors. 1995. Monitoring bird populations by point counts. General technical report PSW-GTR-149. U.S. Forest Service, Pacific Research Station, Albany, California. Reid, W. V. 1998. Biodiversity hotspots. Trends in Ecology and Evolution 13:275–280. Rodríguez, J. P., D. L. Pearson, and R. Barrera. 1998. A test for the adequacy of bioindicator taxa: are tiger beetles (Coleoptera: Cicindelidae) appropriate indicators for monitoring the degradation of tropical forests in Venezuela? Biological Conservation 83:69–76. Ryti, R. T. 1992. Effect of the focal taxon on the selection of nature reserves. Ecological Applications 2:404–410.


Chase et al.

Salvioni, M., and W. Z. J. Lidicker. 1995. Social organization and space use in California voles: seasonal, sexual, age-specific strategies. Oecologia 101:426–438. Sokal, R. R., and F. J. Rohlf. 1995. Biometry. W.H. Freeman, New York. Soulé, M. E. 1991. Conservation: tactics for a constant crisis. Science 253:744–749. Soulé, M. E., D. T. Bolger, A. C. Alberts, J. Wright, M. Sorice, and S. Hill. 1988. Reconstructed dynamics of rapid extinctions of chaparral requiring birds in urban habitat islands. Conservation Biology 2:75–92. State of California. 1992. Bird species of special concern. The Resources Agency, California Department of Fish and Game, Nongame Bird and Mammal Section, Sacramento. State of California. 1993a. Southern California coastal sage scrub natural community conservation plan, scientific review panel conservation guidelines and documentation. The Resources Agency, California Department of Fish and Game, Sacramento. State of California. 1993b. Mammal species of special concern. The Resources Agency, California Department of Fish and Game, Nongame Bird and Mammal Section, Sacramento. Tardif, B., and J-L. DesGranges. 1998. Correspondence between bird and plant hotspots of the St. Lawrence River and the influence of scale on their location. Biological Conservation 84:53–63. Temple, S. A., and J. A. Wiens. 1989. Bird populations and environmental changes: can birds be bio-indicators? American Birds 43: 260–270. Troumbis, A. Y., and P. G. Dimitrakopoulos. 1998. Geographic coincidence of diversity threatspots for three taxa and conservation planning in Greece. Biological Conservation 84:1–6. Usher, M. B., editor. 1986. Wildlife conservation evaluation. Chapman and Hall, London. Westman, W. E. 1981. Factors influencing the distribution of species of Californian coastal sage scrub. Ecology 62:439–455. Westman, W. E. 1983. Xeric Mediterranean-type shrubland associations of Alta and Baja California and the community/continuum debate. Vegetatio 52:3–19. Westman, W. E. 1987. Implications of ecological theory for rare plant conservation in coastal sage scrub. Pages 133–140 in T. S. Elias, editor. Conservation and management of rare and endangered plants. California Native Plant Society, Sacramento. White, S. D., and W. D. Padley. 1997. Coastal sage scrub series of Western Riverside county, California. Madroño 44:95–105. Williams, P., D. Gibbons, C. Margules, A. Rebelo, C. Humphries, and R. Pressey. 1996. A comparison of richness hotspots, rarity hotspots, and complementary areas for conserving diversity of British Birds. Conservation Biology 10:155–174.

Chase et al.


487

Appendix 1 Additional species detected at 155 sampling points and not evaluated as indicators, ordered from most to least frequently detected. Common name Birds California Towhee Spotted Towhee Northern Flicker Acorn Woodpecker Grasshopper Sparrow Lazuli Bunting Phainopepla Wilson’s Warbler Blue-gray Gnatcatcher Oak Titmouse Ash-throated Flycatcher European Starling Say’s Phoebe Red-winged Blackbird Western Kingbird Brewer’s Blackbird Bullock’s Oriole Nuttall’s Woodpecker Brown-headed Cowbird American Goldfinch Black Phoebe Cassin’s Kingbird Dark-eyed Junco Hooded Oriole Lark Sparrow Pacific-slope Flycatcher Ruby-crowned Kinglet Savannah Sparrow Warbling Vireo Winter Wren Mammals California vole House mouse

Scientific name

Number of points

Species code

Pipilo crissalis Pipilo maculatus Colaptes auratus Melanerpes formicivorus Ammodramus savannarum Passerina amoena Phainopepla nitens Wilsonia pusilla Polioptila caerulea Parus inornatus Myiarchus cinerascens Sturnus vulgaris Sayornis saya Agelaius phoeniceus Tyrannus verticalis Euphagus cyanocephalus Icterus bullockii Picoides nuttallii Molothrus ater Carduelis tristis Sayornis nigricans Tyrannus vociferans Junco hyemalis Icterus cucullatus Chondestes grammacus Empidonax dificilis Regulus calendula Passerculus sandwichensis Vireo altiloquus Troglodytes troglodytes

145 140 16 15 15 13 11 10 9 8 6 6 6 4 4 3 3 3 2 1 1 1 1 1 1 1 1 1 1 1

CALT SPTO NOFL ACWO GRSP LAZB PHAI WIWA BGGN OATI ATFL EUST SAPH RWBL WEKI BRBL BUOR NUWO BHCO AMGO BLPH CAKI DEJU HOOR LASP PSFL RCKI SAVS WAVI WIWR

3 1

MICA MUMU

Microtus californicus Mus musculus
