etee4,1„/.. or,44.-/te ;eats- ai d ezt u,tch,. /45A 60,4,-,1 /,,,Tiaa€4 CAAth- ol i ein /rid/714,7ra% 6tda, 4,4,77,4 Jet' 4.A_ Klamath-Siskiyou Herpetofauna: Biogeographic Patterns and Conservation Strategies R. Bruce Bury Christopher A. Pearl Forest and Rangeland Ecosystem Science Center USGS Biological Resources Division 3200 Jefferson Way Corvallis, OR 97331 USA • Natural A teas Journal 19:341-350 (31 4A-87AetAtkoS , INTRODUCTION The western edge of the North American continent is characterized by extensive north—south corridors of uplifted terrain, including the Coast Ranges and Cascade Mountains in the Pacific Northwest (PNW). These ranges converge in the Klamath- Siskiyou (KS) region (in which we in- clude the Kalmiopsis, high Siskiyous, Trin- ity Alps, Marble Mountains, and portions of the Yolla Bolly range) of southwestern Oregon and northwestern California. Far- ther south, California's Central Valley again separates the ranges, and the Cascades give way to the Sierra Nevada Range. Only the highest elevations in the KS region (pre- dominately within the Trinity Alps and Marble Mountains) were glaciated during the Pleistocene (Davis 1988), allowing animal and plant species distributions to shift north and south during periods of climatic change. Several physical characteristics of the KS region contribute to a landscape of com- plicated habitat mosaics. Many rivers and streams, including the large Rogue and Klamath drainages, dissect rugged terrain, resulting in widely varied topographies (sea level to 2,200 m) and aspects. Exposed ultramafic substrates (e.g., serpentine and peridotite) and their unique floristic asso- ciates further contribute to a diverse land- scape mosaic. Spatially complex fire his- 3 LP/ 0 tories also typify these mountains (Agee 1991, Atzet and Martin 1991). Superim- posed over the entire region are strong differences in seasonal climates (extended cool, moist winter conditions and hot, semi- arid summers) and a west—east gradient in precipitation (from about 405 cm per year near the coast to about 80 cm in the eastern rain shadow; Froehlich et al. 1982). Such physical gradients produce an array of habitat types available for reptiles and amphibians within a relatively small geo- graphic area. Life history traits and habitat requirements are among the most important factors gov- erning the distribution of any taxonomic group. For example, the tailed frog is adapt- ed to cool, rocky streams in closed-canopy forests, whereas reptiles such as the south- ern alligator lizard require warmer condi- tions often associated with exposed, rocky slopes. (Scientific names for amphibians and reptiles are given in Appendices 1 and 2, respectively.) In the KS region, these vastly different habitats can occur in close proximity and may allow more species to occur than would be expected in a more uniform environment. The objectives of this paper are to describe the distribution patterns of the herpetofau- na within the KS region, summarize the ecological roles and habitat requirements of the native amphibians and reptiles of ABSTRACT: The Klamath-Siskiyou region of southwest Oregon and northwest Califor- nia (USA) has some of the most complex landscape mosaics and plant communities in western North America, reflecting its marked diversity of precipitation and topography. With 38 native species of amphibians and reptiles, the Klamath-Siskiyou region has the most species-rich herpetofauna of any similarly sized mountain range in the Pacific Northwest. Although it is a biodiversity "hot spot," there are only two endemic species, both salamanders, in the Klamath-Siskiyou region. High diversity is due to the overlap of two major biogeographic groups: the Arcto- (= northern) and Madro- (= southern) Tertiary herpetofaunas. Many of the amphibians in the Klamath-Siskiyou region are restricted to specialized habitats. Much of our knowledge about the biology of the regional fauna is based on studies elsewhere. Distributional surveys and ecological research are needed to address how the herpetofauna responds to timber harvest and other human activities that may reduce populations and increase fragmentation of suitable habitats. Conservation of the region's diverse herpetofauna should emphasize strategies directed at habitat specialists and species at the latitudinal limits of their ranges. Index terms: amphibians, biogeography, conservation, Klamath-Siskiyou region, reptiles Volume 19 (4), 1999 Natural Areas Journal 341
USGS Biological Resources Division3200 Jefferson Way
Corvallis, OR 97331 USA
•
Natural A teas Journal 19:341-350
(314A-87AetAtkoS ,
INTRODUCTION
The western edge of the North Americancontinent is characterized by extensivenorth—south corridors of uplifted terrain,including the Coast Ranges and CascadeMountains in the Pacific Northwest (PNW).These ranges converge in the Klamath-Siskiyou (KS) region (in which we in-clude the Kalmiopsis, high Siskiyous, Trin-ity Alps, Marble Mountains, and portionsof the Yolla Bolly range) of southwesternOregon and northwestern California. Far-ther south, California's Central Valley againseparates the ranges, and the Cascades giveway to the Sierra Nevada Range. Only thehighest elevations in the KS region (pre-dominately within the Trinity Alps andMarble Mountains) were glaciated duringthe Pleistocene (Davis 1988), allowinganimal and plant species distributions toshift north and south during periods ofclimatic change.
Several physical characteristics of the KSregion contribute to a landscape of com-plicated habitat mosaics. Many rivers andstreams, including the large Rogue andKlamath drainages, dissect rugged terrain,resulting in widely varied topographies (sealevel to 2,200 m) and aspects. Exposedultramafic substrates (e.g., serpentine andperidotite) and their unique floristic asso-ciates further contribute to a diverse land-scape mosaic. Spatially complex fire his-
3 LP/ 0
tories also typify these mountains (Agee1991, Atzet and Martin 1991). Superim-posed over the entire region are strongdifferences in seasonal climates (extendedcool, moist winter conditions and hot, semi-arid summers) and a west—east gradient inprecipitation (from about 405 cm per yearnear the coast to about 80 cm in the easternrain shadow; Froehlich et al. 1982). Suchphysical gradients produce an array ofhabitat types available for reptiles andamphibians within a relatively small geo-graphic area.
Life history traits and habitat requirementsare among the most important factors gov-erning the distribution of any taxonomicgroup. For example, the tailed frog is adapt-ed to cool, rocky streams in closed-canopyforests, whereas reptiles such as the south-ern alligator lizard require warmer condi-tions often associated with exposed, rockyslopes. (Scientific names for amphibiansand reptiles are given in Appendices 1 and2, respectively.) In the KS region, thesevastly different habitats can occur in closeproximity and may allow more species tooccur than would be expected in a moreuniform environment.
The objectives of this paper are to describethe distribution patterns of the herpetofau-na within the KS region, summarize theecological roles and habitat requirementsof the native amphibians and reptiles of
ABSTRACT: The Klamath-Siskiyou region of southwest Oregon and northwest Califor-nia (USA) has some of the most complex landscape mosaics and plant communities inwestern North America, reflecting its marked diversity of precipitation and topography.With 38 native species of amphibians and reptiles, the Klamath-Siskiyou region has themost species-rich herpetofauna of any similarly sized mountain range in the PacificNorthwest. Although it is a biodiversity "hot spot," there are only two endemic species,both salamanders, in the Klamath-Siskiyou region. High diversity is due to the overlapof two major biogeographic groups: the Arcto- (= northern) and Madro- (= southern)Tertiary herpetofaunas. Many of the amphibians in the Klamath-Siskiyou region arerestricted to specialized habitats. Much of our knowledge about the biology of theregional fauna is based on studies elsewhere. Distributional surveys and ecologicalresearch are needed to address how the herpetofauna responds to timber harvest and otherhuman activities that may reduce populations and increase fragmentation of suitablehabitats. Conservation of the region's diverse herpetofauna should emphasize strategiesdirected at habitat specialists and species at the latitudinal limits of their ranges.
Index terms: amphibians, biogeography, conservation, Klamath-Siskiyou region, reptiles
Volume 19 (4), 1999 Natural Areas Journal 341
entra oasRange
the KS region, and discuss considerationsfor conserving the diverse herpetofauna ofthe KS region.
DIVERSITY AND BIOGEOGRAPHY
We estimate that 48 species of amphibiansand reptiles occur in the temperate PNW,which we define as the region west of theCascade Mountains and from San Fran-cisco Bay northward to western Canada(Figure 1). This is a moderately rich her-petofauna with many regionally endemicamphibians (Nussbaum et al. 1983, Greenand Campbell 1984, Stebbins 1985). TheKS region supports at least 38 native spe-cies of amphibians and reptiles (Table 1,Appendix 1, Appendix 2), which repre-sents the most species-rich herpetofaunaamong similarly sized mountain provincesin the PNW. The KS region supports thehighest species richness of both sala-manders and snakes within the westernPNW (Table 1). Reptiles vary widely inspecies richness from only 4 species in theOlympic Mountains of northwestern Wash-ington to 19 species in the KS region. Ingeneral, herpetofaunal diversity increaseswith southern latitude, as do the numberand percentage of KS species shared be-tween regions (Table 2). We found that79% of the species in the KS region alsooccur in the Coast Range of northern Cal-ifornia, and 63% occur in both the centralCascade Mountains and Coast Ranges ofOregon. Only 37% of KS species are sharedwith the Olympic Mountains of Washing-ton, due in large part to the paucity ofreptile species in the Olympic region.
The KS region represents a "hot spot" ofspecies richness within the PNW for sev-eral reasons. Most important, the KS re-gion is an overlap zone for two majorbiogeographic elements: the Madro-(southern) and Arcto- (northern) Tertiarygeofaunas (Peabody and Savage 1958).The range limits of many amphibians oc-cur in the KS region, and the overlap ofnorthern (e.g., tailed frogs, red-backed, andDunn's salamanders) and southern taxa(e.g., California slender salamander andblack salamander) contribute to the rich-ness of the fauna (Figure 2). Further, themosaic and patchy nature of habitats in theKS region allows occurrence of a diverse Figure 1. Location of Klamath-Siskiyou region and comparable mountain regions in the Pacific
Northwest. The California Coast Range (not shown) is just south of the map limit.
342 Natural Areas Journal Volume 19 (4), 1999
Table 1. Herpetofaunal species richness in the Pacific Northwest (number of families in parenthe-ses). Abbreviations for mountain regions are: KS = Klamath-Siskiyou region, Oregon—Califor-nia; OLYM = Olympic Mountains, northwestern Washington; OR Coast = Central Coast Range,Oregon; OR Casc = Central Cascade Range, Oregon; CA Coast = Northern California CoastRange.
Table 2. Number of species shared (percent of KS species in parentheses) between Klamath-Siskiyou herpetofauna and four other mountain regions of the Pacific Northwest. Abbreviationsfollow those in Table 1.
Figure 2. Distribution of range limits for southern (= Madro-Tertiary) and northern (= Arcto-Tertiaryamphibian taxa within the Klamath-Siskiyou region (39°-43° north latitude). "Total species" includesonly southern and northern species and not widespread taxa (see Appendix 1).
herpetofauna. Reptiles tend to be mostcommon in hotter, drier conditions (inlandsites, lower elevations, and south-facingslopes), while amphibians generally areassociated with mesic or wet areas (coast-al or riparian sites, higher elevations, andnorth-facing slopes).
Amphibians
High endemism is a feature of amphibianbiogeography in western North America(Bury 1994). Most (64%) of the 22 am-phibian species inhabiting forest habitatsare regionally endemic to the PNW: 4 spe-cies of torrent salamanders, 2-3 species ofgiant salamanders, 6 species of lunglesssalamanders, 2 species of ranid frog, andthe tailed frog. One or more species fromeach of these groups occur in the KS re-gion (Appendix 1). The tailed frog is ofparticular interest because it represents agroup of primitive anurans that occur onlyin widely separated ranges: tailed frogsare endemic to the PNW (as far east aswestern Montana), but bell toads (genusLeiopelma), their closest relative, occuronly in New Zealand.
The high amphibian species richness in theKS region includes several different biogeo-graphic elements: northern taxa (Arcto-Ter-tiary), southern taxa (Madro-Tertiary), andwidespread taxa (Appendices 1 and 2). Fiveof the 19 total species (26%) are geograph-ically widespread within the region. Theseinclude species such as the roughskin newtand western toad. Three species (16%) havesouthern affinities, but most (58%) are asso-ciated with northem distributions. Exam-ples include the torrent salamanders andtailed frogs, both of which represent ancientArcto-Tertiary lineages (see Welsh 1990).Two closely related terrestrial species areendemic to the KS region: the Del Norte andSiskiyou Mountain salamanders. This num-ber of endemic species is low but similar toother areas of comparable size in the PacificNorthwest (e.g., the Olympic Peninsula).
The KS region also has one of the mostdiverse salamander assemblages in westernNorth America (Table 1, Figure 3). As manyas 10 salamander species from 5 familiescan be found in one river basin. Recently,we discovered that up to three species of
Volume 19 (4), 1999 Natural Areas Journal 343
Figure 3. Centers of species richness ("hot spots") for salamanders in the Klamath-Siskiyou region ofsouthwestern Oregon and northwestern California. Shading represents occurrence of different numbersof species: darkest shade = 9-10 species; moderate = 7-8; light = 5-6 species; no shading = < 5 speciespresent.
lungless salamanders (genus Plethodon) maybe encountered syntopically (overlapping indistribution) in small northern portions ofthe region (Figure 4). The KS region sharesmany characteristics with the southern Ap-palachian Mountains, which is the NorthAmerican center of lungless salamander di-versity (Highton 1995). Both of these botan-ically rich regions are ancient uplifts (>10million years old) with widely varied topog-raphy and metamorphic geology that werespared the direct effects of glaciation duringthe Pleistocene (Whittaker 1956, 1960; Ir-win 1981; Davis 1988).
Reptiles
Nineteen species of reptiles are known tooccur in the KS region, making it the mostspecies-rich assemblage of similarly sizedmountain regions of the PNW (Table 1).There is a large southern element (42%) inthe reptile fauna (Appendix 2). Many ofthe southern species are associated withMediterranean habitats (e.g., the commonand California mountain kingsnakes,striped whipsnakes). These are reptiles thatare broadly associated with the Madro-Tertiary geoflora of southern origin (Pea-body and Savage 1958). Two species
(10.5%) are representative of montane ornorthern latitudes: the northern alligatorlizard and northwestern garter snake. Oth-er reptiles, such as the gopher snake, west-ern fence lizard, and ring-neck snake, havewidespread distribution patterns. There areno endemic reptiles in the KS region.
This relatively high richness of reptiles isdue in part to the pronounced rain shadowand resulting hot, arid conditions in theeastern portions of the KS region. Manyreptile species (particularly snakes) areassociated with water (e.g., turtles, gartersnakes) or specific terrestrial habitats (e.g.,sharptail snakes in oak woodland). Xeric-associated plant assemblages such as oakwoodland, pine forest, chaparral, and grass-land occur in inland valleys and along theeastern flanks of the mountains. Histori-cally, these drier areas were characterizedby frequent fires that maintained their openstructure (Agee 1990) and created condi-tions favorable to reptile use.
ECOLOGY AND HABITATS
Amphibians
Amphibians of the KS region are diverse intheir trophic positions, life histories, andhabitat use. Eleven of the 19 KS amphibianspecies (58%) depend on aquatic habitatsfor egg deposition and development ofyoung. These species generally utilize bothaquatic and terrestrial habitats during differ-ent portions of their life cycles, necessitat-ing a dual set of suitable habitat conditionsfor their survival. All adult amphibians inthe KS region feed on invertebrates. Mostlarval anurans are grazers on phytoplanktonand diatoms, and can be important structur-ing agents in algal communities (Seale 1980,Kupferberg 1997). Larval salamanders aregape-limited predators, and may feed onzooplankton, mollusks, aquatic insects, andsmaller amphibian larvae. Neotenic and ter-restrial adults of the Pacific giant salamandermay also eat fish, other amphibians, and anoccasional small mammal. Due to their highlocal densities, aquatic salamander larvaecan directly influence zooplankton compo-sition (Dodson 1970) and larval anuran com-munities (Wilbur et al. 1983), as well asindirectly affect primary production (Morin1995).
344 Natural Areas Journal Volume 19 (4), 1999
Figure 4. Distributional limits of four species of lungless salamanders (genus Plethodon) in the Klamath-Siskiyou region. Abbreviations: PLEL = P. elongatus, Del Norte salamander; PLDU = P. dunni, Dunn'ssalamander; PLST = P. stormi, Siskiyou Mountain salamander; and PLVE = P. vehiculum, red-backedsalamander.
Amphibians are important components ofthe PNW fauna in terms of their numbersand biomass. Amphibians may be domi-nant predators on invertebrates of the for-est floor (Bury 1994) and may be the mostnumerous terrestrial vertebrates in manyforests. For example, Bury (1983) esti-mated that there were over 400 salamandersha-' in old-growth redwood forests, andRaphael (1984) reported densities of 10-180 salamanders ha-' in mixed conifer—hardwood forests in northern California.Burton and Likens (1975) estimated den-sities of approximately 3,000 salamandersha-' in eastern deciduous forests. Such ter-
restrial salamanders can exceed 5 individ-uals m-2 in local aggregations in favoredmicrohabitats (Jaeger 1979, Bury andRaphael 1983) and represent an importantenergetic pathway between microfauna andhigher predators in temperate forests (Hair-ston 1996).
Long-toed salamanders have been ob-served to exceed 5 larvae m - 2 in temporaryponds (C. Pearl, unpubl. data). Localizedaggregations of larval and juvenile west-ern toads can exceed 50 individuals m-2(Nussbaum et al. 1983; B. Bury and C.Pearl, unpubl. data). Stream amphibians
can also reach high densities in seeps,creeks, and headwater stream systems. Forexample, larvae of the Pacific giant sala-mander may occur at densities of 2-5 in-dividuals m-2 and are often the dominantpredators in headwater streams, where theymay exceed fish in numbers and biomass(Corn and Bury 1989).
The KS region derives much of its diversi-ty from an abundance of terrestrial lung-less salamanders (Family Plethodontidae).These salamanders complete reproductionand oviposition entirely on land. Sincerespiration occurs directly across dermalmembranes and the throat, lungless sala-manders generally require microenviron-ments with low temperature and highmoisture. Talus slopes offer the moisture,temperature, and summer refuge condi-tions required by many of these sala-manders. Some lungless salamanders, suchas the clouded salamander, are closely tiedto woody debris (Corn and Bury 1991).Ensatinas and red-backed salamandersprefer partially decomposed wood and arecommonly found under logs and loose barkin Douglas-fir stands in the Cascades(Aubry et al. 1988).
Reptiles
Reptiles are important predators in foodwebs and are prey to a variety of predatorsincluding larger reptiles, raptors, and manymammals. The regional reptile fauna iscomposed of species that are almost allobligate predators, and they demonstrate arange of prey specializations. Western pondturtles are mostly aquatic, and they con-sume invertebrates, tadpoles, vegetation,and sometimes carrion. All five species ofresident lizards feed primarily on inverte-brates. Snakes, the most species-rich groupof the reptiles, include small, fossorialspecies, such as the sharptail snake, thatfeed on slugs and other invertebrates; ac-tive, agile species like the western aquaticgarter snake and striped whipsnake thatfeed extensively on smaller amphibiansand reptiles, respectively; and large con-strictors, like the gopher snake, that preyprimarily on small mammals. The rubberboa is the sole representative of a family oftropical species, and the western rattle-snake is the only poisonous species in the
Volume 19 (4), 1999 Natural Areas Journal 345
region; both rely on small mammals forfood.
Some reptile species are active in coolweather (e.g., the sharptail snake, rubberboa, and northern alligator lizard may beactive in April), but most reptiles requirewarmer conditions for foraging and repro-duction (e.g., kingsnakes tend to be mostactive in May/June). The western pondturtle is generally aquatic but needs bask-ing sites in which to thermoregulate andincrease its body temperature for severalhours each day. There are few or no rep-tiles present in mature and old-growth for-ests of Douglas-fir or mixed conifers, pri-marily because of the cooler conditionsunder closed canopies.
Seven of the 19 KS reptile species bearlive young, including the northern alliga-tor lizard, 4 species of garter snakes, therubber boa, and the western rattlesnake,but most species (63%) are egg layers.Reptile diversity and abundance are great-est in habitats such as cliff faces, exposedtalus slopes, open woodlands or chaparralon south-facing slopes, and serpentinebarrens. Talus and rocky slopes are partic-ularly important as they may be used foregg-deposition and hibernation sites. Asmany as five to six reptile species havebeen documented sharing such overwin-tering habitats (Nussbaum et al. 1983,Leonard and Leonard 1998).
CONSERVATION
The herpetofauna of the KS region is di-verse taxonomically, as well as in terms oflife history and habitat requirements. Thus,conservation strategies must consider awide variety of taxa and associated habi-tats. Many species of PNW amphibiansare associated with late-successional orold-growth forests (Raphael 1984, Bury1994, Blaustein et al. 1995). These forestshave been highly modified by human ac-tivities, with timber harvest serving as theprimary agent of landscape alteration overthe last half-century (Jimerson and Hoover1991, Jules 1997). Many plethodontid spe-cies are sensitive to the effects of logging,and most are reduced in abundance inclearcut forests (Bury 1994). Coarse woodydebris is a major component of old-growth
forests, and amounts remaining on forestfloors have been greatly reduced by pastforestry practices (Maser and Trappe 1984,Harmon et al. 1956). Occurrence of sala-manders appears to be closely linked toabundance of coarse woody debris on theforest floor (Corn and Bury 1991), indi-cating a need to increase retention of largewood after logging.
Both the Siskiyou Mountain and Del Nortesalamanders appear to be adversely affect-ed by timber harvest on forested talusslopes that comprise their primary habitat(Welsh and Lind 1995). These salamandersare recognized as "Survey and ManageSpecies" under the Northwest Forest Plan(Thomas et al. 1995), and surveys for theirpresence are required prior to ground-dis-turbing activities. Also, current guidelinessuggest protection of historic salamandersites and retention of buffers around occu-pied sites. However, research is needed toevaluate the effectiveness of buffer widthsfor salamander conservation.
The KS region represents the southern ter-minus for a number of northern amphibianand reptile species, while a few southernspecies reach their northern limits in theSiskiyou Mountains of southwestern Ore-gon. These taxa are presumably at or neartheir thermal and/or moisture tolerances.Additional research is needed to betterunderstand their responses to changes intheir favored microhabitats, which may beassociated with short- or long-term cli-matic changes.
Aquatic taxa face a wide array of threatsand are sensitive to local environmentalchange (Richter et al. 1997). Due to limit-ed Pleistocene glaciation and the currentclimatic regime, large pond or lake habi-tats are relatively rare below the upper-most elevations in the KS region (mostlyin the Trinity and Marble Mountains ofnorthwestern California). However, the KSregion does contain a variety of smallerslump ponds, seeps, and vernal pools. Fouranuran species (Pacific chorus frog, west-ern toad, northern red-legged frog, andCascade frog) and three salamanders(roughskin newt, and long-toed and north-western salamanders) require pond habi-tats for successful reproduction.
The rarity and isolation of standing watershave important implications for recoloni-zation by amphibians and other aquatictaxa after local extinctions (see Sjogren1991). Amphibian dispersal and success-ful reestablishment in widely separatedpond and lake habitats, particularly acrossaltered terrestrial habitat, may be limitedin the KS region. Further, small isolatedpopulations are generally more vulnerableto stochastic environmental or demograph-ic stresses and are naturally more prone tolocal extinction (Lande 1988). The scarci-ty of standing water may partially explainthe apparent absence of the Cascade frogin the high Siskiyou Mountain range alongthe Oregon-California border. This spe-cies is present at many lentic sites in theOregon Cascades (Nussbaum et al. 1983)and occurs farther south in the formerlyglaciated areas of the Trinity Alps andMarble Mountains of northern California(Jennings and Hayes 1994).
Establishment of nonnative mammal spe-cies (e.g., feral pigs [Sus scrofa], oppo-sums [Didelphis marsupialis]) may havenegative impacts on some terrestrial her-petofauna, but, to our knowledge, therehave been no studies conducted in thewestern United States. Some experimentaland distributional evidence suggests thatintroduced species such as bullfrogs and asuite of stocked game fish may interactnegatively with native aquatic herpetofau-na (Hayes and Jennings 1986, Kieseckerand Blaustein 1998). Bullfrogs, native tothe eastern United States, have been wide-ly introduced in western North America,and even in other continents, as a foodsource. Bullfrogs may affect amphibiansand aquatic reptiles by predation, compe-tition, or acting as disease vectors (Buryand Whelan 1984, Kupferberg 1997,Kiesecker and Blaustein 1998). Bullfrogexpansion in the KS region has occurredprimarily in low-elevation stock ponds andalong large rivers, but mid-elevation aquatichabitats may also be colonized.
As many as 95% of lakes in the mountainareas of the western United States werehistorically fishless (Bahls 1992). Present-ly, most high-elevation lakes in the KSregion have been stocked with brook trout
346 Natural Areas Journal Volume 19 (4), 1999
(Salvelinus fontinalis) and rainbow trout(Oncorhynchus mykiss) for sport fishing(L. Webb, pers. corn.). Introduced troutare now implicated in local declines ofaquatic amphibians (Bradford 1989, Fellersand Drost 1993, Tyler et al. 1998). Intro-duced largemouth bass (Micropterussalmoides) are commonly stocked in low-elevation ponds and lakes, and are effec-tive predators on a variety of small verte-brates including amphibians (Moyle 1986).Our observations of largemouth bass inone lake at 1,200 m elevation (C. Pearl,pers. obs.) suggest that the moderate cli-mate within the KS region can allow thepersistence of this nonnative fish across awide elevational range.
The association of pond-breeding amphib-ians with older forests is unclear, but allpond-breeding species use terrestrial hab-itats to varying extent. Red-legged frogsand western toads may spend considerabletime in terrestrial areas, particularly inhumid weather. Similarly, roughskin newtsand mole salamanders (Ambystomatidae)migrate overland to breeding waters andreside on land for much of the year. Thereis little information on how habitat alter-ations may affect northwestern pond-breeding amphibians during the periodswhen they occur in terrestrial habitats, butresearch on mole salamanders in the east-ern United States indicates that forest hab-itat as much as 250 m away from breedingpools may be extensively used by adultsand probably is important in the persis-tence of the breeding population (Sem-litsch 1999).
Headwater streams also are critical habitatfor several species of amphibians endemicto the Pacific Northwest (e.g., tailed frog,torrent salamander). These taxa require cooltemperatures and persistent waters. Timberharvest affects aquatic amphibian popula-tions by increasing water temperature to le-thal levels and through siltation of stream-beds (Corn and Bury 1989). Recently, wefound that stream amphibian populations inthe Oregon Coast Range had not recovered35-50 years after clearcut harvesting (B.Bury and D. Major, unpubl. data). Althoughadults of the tailed frog are considered to beclosely associated with streams (Blausteinet al. 1995), we discovered that in forested
stands some adults may occur relatively longdistances (up to 500 m) from water duringthe wet season (Corn and Bury 1991). Also,we observed that recently transformed frogsappear to disperse into terrestrial habitatswith the onset of heavy fall rains (B. Buryand L. Jones, pers. obs.). Thus, retention ofriparian zones and adjacent forests in olderstages is likely critical to the survival ofstream amphibian species. Further researchon this topic is critically needed.
CONCLUSIONS ANDRECOMMENDATIONS
In the past, clearcut logging decreased localpopulations of the KS herpetofauna andfurther fragmented a montane landscapethat naturally contained patchy habitats.Today, many sensitive species are isolatedin disjunct habitat patches, and timberharvest at these sites may further depleteremaining populations. Although moreecologically based methods of timber har-vest are being implemented (e.g., standthinning, retention of riparian bufferzones), studies assessing the effect of thesenew practices on the herpetofauna are intheir infancy.
While studies of individual species are nec-essary, we recommend a habitat-based ap-proach for more effective conservation ofthe regional herpetofauna. Specializedhabitats are important to the survival ofmany species of the KS herpetofauna, par-ticularly endemic salamander species.Thus, we recommend that conservationefforts focus on three important habitats:talus slopes, headwater streams, and ponds.We can conserve a large portion of sensi-tive KS herpetofauna with continued re-search, improved management, and pro-tection of these spatially limited habitats.
The Klamath-Siskiyou region is one of theleast-explored mountain regions of west-ern North America. Distributions and lifehistories of the native herpetofauna arestill relatively poorly known, and much ofour knowledge of these species is derivedfrom other regions. We need greater knowl-edge of species' biology and habitat re-quirements for better management andprotection of the KS herpetofauna. Final-ly, an understanding of the complex bio-
geography of this diverse region will ben-efit from better coordination among re-search biologists in different disciplines.For example, many of the habitats neces-sary for survival of sensitive amphibianand reptile species are also needed by oth-er taxa (e.g., mollusks). Cooperative stud-ies involving resource managers, scien-tists, and conservationists have begun, andfurther collaborative efforts should be en-couraged.
ACKNOWLEDGMENTS
Fieldwork was supported in part by theU.S. Forest Service, Pacific NorthwestResearch Station, and the Bureau of LandManagement. We thank Don J. Major forfield assistance and final preparation ofthe figures, Michael J. Adams for com-ments on the manuscript, and the Depart-ment of Geography, University of Oregon,for use of maps.
Bruce Bury is a Research Zoologist in theUSGS Forest and Rangeland EcosystemScience Center in Corvallis, Oregon. He isa 27-year veteran of government service.His research interests feature ecologicaland conservation issues of amphibians andreptiles.
Christopher Pearl is a temporary WildlifeBiologist with the USGS Forest and Range-land Ecosystem Science Center. His re-search interests include biology of ranidfrogs in the Pacific Northwest, wetlandecology, and paleoecology of the Wil-lamette Valley.
LITERATURE CITED
Agee, J.K. 1990. The historical role of fire inPacific Northwest forests. Pp. 25-37 in J.D.Walstad, S.R. Radosevich, and D.V. Sand-berg, eds., Natural and Prescribed Fire inPacific Northwest Forests. Oregon StateUniversity Press, Corvallis.
Agee, J.K. 1991. Fire history along an eleva-tional gradient in the Siskiyou Mountains,Oregon. Northwest Science 65:188-199.
Atzet, T. and R.E. Martin. 1991. Natural dis-turbance regimes in the Klamath Province.Pp. 40-48 in R.R. Harris and D.E. Erman,tech. coords., and H.M. Kerner, ed., Pro-
Volume 19 (4), 1999 Natural Areas Journal 347
ceedings of Symposium on Biodiversity ofNorthwestern California. Report No. 29,University of California, Wildland Re-sources Center, Berkeley.
Aubry, K.B., L.L.C. Jones, and P.A. Hall. 1988.Use of woody debris by plethodontid sala-manders in Douglas-fir forests in Washing-ton. Pp. 32-37 in R.C. Szaro, K.E. Sever-son, and D.R. Patton, eds., Management ofamphibians, reptiles, and small mammals inNorth America: Proceedings of the sympo-sium. General Technical Report RM-166,U.S. Department of Agriculture, ForestService, Tempe, Ariz.
Bahls, P. 1992. The status of fish populationsand management of high mountain lakes inthe western United States. Northwest Sci-ence 66:183-193.
Blaustein, A.R., J.J. Beatty, D.H. Olson, andR.M. Storm. 1995. The biology of amphib-ians and reptiles in old-growth forests in thePacific Northwest. General Technical Re-port PNW-337, U.S. Department of Agri-culture, Forest Service, Portland, Ore.98 pp.
Bradford, D.F. 1989. Allotopic distribution ofnative frogs and introduced fishes in highSierra Nevada lakes of California: implica-tion of the negative effect of fish introduc-tions. Copeia 1989:775-778.
Burton, T.M. and G.E. Likens. 1975. Sala-mander populations and biomass in theHubbard Brook Experimental Forest, NewHampshire. Copeia 1975:541-546.
Bury, R.B. 1983. Differences in amphibianpopulations in logged and old growth red-wood forest. Northwest Science 57:167-178.
Bury, R.B. 1994. Vertebrates in the PacificNorthwest: species richness, endemism anddependency on old-growth forests. Pp. 392-404 in S. K. Majumdar, F.J. Brenner, J.E.Lovich, J.F. Schalles, and E.W. Miller, eds.,Biological Diversity: Problems and Chal-lenges. The Pennsylvania Academy of Sci-ences, Easton.
Bury, R.B. and M.G. Raphael. 1983. Inventorymethods for amphibians and reptiles. Pp.416-419 in J.F. Bell and T. Atterbury, eds.,Renewable Resource Inventories for Mon-itoring Changes and Trends. SAF 83-14,Oregon State University, Corvallis. 737 pp.
Bury, R.B. and J.A. Whelan. 1984. Ecologyand management of the bullfrog. ResourcePublication 155, U.S. Department of Interi-or, Fish and Wildlife Service, Washington,D.C. 23 pp.
Com, P.S. and R.B. Bury. 1989. Logging inwestern Oregon: responses of headwaterhabitats and stream amphibians. Forest Ecol-ogy and Management 29:1-19.
Com, P.S. and R.B. Bury. 1991. Terrestrial
amphibian communities in the Oregon CoastRange. Pp. 304-317 in L.F. Ruggierio, K.B.Aubry, A.B. Carey, and M.H. Huff, tech.coords., Wildlife and vegetation of unman-aged douglas-fir forests. General TechnicalReport PNW-285, U.S. Department of Ag-riculture, Forest Service, Portland, Ore.
Davis, P.T. 1988. Holocene glacier fluctua-tions in the American cordillera. Quaterna-ry Science Reviews 7:129-157.
Dodson, S.I. 1970. The effect of predation onthe community structure of fresh-waterzooplankton. Ph.D. diss., University ofWashington, Seattle.
Fellers, G.M. and C.A. Drost. 1993. Disap-pearance of the Cascades frog Rana cas-cadae at the southern end of its range, Cal-ifornia, USA. Biological Conservation65:177-181.
Froehlich, H.A., D.H. McNabb, and F. Gawe-da. 1982. Average annual precipitation1960-1980 in Southwest Oregon. Cooper-ative Extension Service, Oregon State Uni-versity, Corvallis. 4 pp.
Green, D.M. and R.W. Campbell. 1984. TheAmphibians of British Columbia. HandbookNo. 45, British Columbia Provincial Muse-um, Victoria, Canada. 101 pp.
Hairston, N.G., Sr. 1996. Predation and com-petition in salamander communities. Pp.161-190 in M.L. Cody and J.A. Smallwood,eds., Long-term Studies of Vertebrate Com-munities. Academic Press, San Diego,Calif.
Hayes, M.P. and M.R. Jennings. 1986. Declineof ranid frog species in western NorthAmerica: are bullfrogs (Rana catesbeiana)responsible? Journal of Herpetology 20:490-509.
Highton, R. 1995. Speciation in eastern NorthAmerican salamanders of the genus Pleth-odon. Annual Review of Ecology and Sys-tematics 26:579-600.
Irwin, W.P. 1981. Tectonic accretion of theKlamath Mountains. Pp. 29-49 in W.G.Ernst, ed., The Geotectonic Developmentof California. Prentice-Hall, EnglewoodCliffs, N.J.
Jaeger, R.G. 1979. Seasonal spatial distribu-tions of the terrestrial salamander Pleth-odon cinereus. Herpetologica 35:90-93.
Jennings, M.R. and M.P. Hayes. 1994. Am-phibian and reptile species of special con-
cern in California, Final Report. CaliforniaDepartment of Fish and Game, Inland Fish-eries Division. 255 pp.
Jinierson, T.M. and L.D. I loover. 1991. Old-growth forest fragmentation: changes inamount, patch size, and edge as a result oflogging. Pp. 168-174 in R.R. Harris andD.E. Erman, tech. coords., and H.M. Kern-er, ed., Proceedings of Symposium on Biodi-versity of Northwestern California. ReportNo. 29, University of California, WildlandResources Center, Berkeley.
Jules, E.S. 1997. Rapid changes in plant distri-bution and abundance in the Sucker Creekwatershed: a case study of Trillium ovatum.Pp. 71-79 in J. Beigel, E.S. Jules, and B.Snitkin, eds., Proceedings of the First Con-ference of Siskiyou Ecology. The Siskiyouregional Education Project, Cave Junction,Ore.
Kiesecker, J. and A.R. Blaustein. 1998. Effectsof introduced bullfrogs and smallmouth basson microhabitat use, growth, and survivalof native red-legged frogs (Rana aurora).Conservation Biology 12:776-787.
Kupferberg, S.J. 1997. Bullfrog (Rana cates-beiana) invasion of a California river: therole of larval competition. Ecology 78:1736-1751.
Lande, R. 1988. Genetics and demography inbiological conservation. Science 241:1455-1460.
Leonard, W.P. and M.A. Leonard. 1998. Oc-currence of the sharptail snake (Contiatenuis) at Trout Lake, Klickitat County,Washington. Northwestern Naturalist 79:75-76.
Maser, C. and J.M. Trappe. 1984. The seen andunseen world of the fallen tree. GeneralTechnical Report PNW-164, U.S. Depart-ment of Agriculture, Forest Service, Port-land, Ore. 56 pp.
Morin, P.J. 1995. Functional redundancy, non-additive interactions, and supply-side dy-namics in experimental pond communities.Ecology 76:133-149
Moyle, P.B. 1986. Fish introductions into NorthAmerica: patterns and ecological impact.Pp. 27-43 in H.A. Mooney and J.A. Drake,eds., Ecology of Biological Invasions ofNorth America and Hawaii. Springer-Ver-lag, New York.
Nussbaum, R.A., E.D. Brodie Jr. and R.M.Storm. 1983. Amphibians and Reptiles ofthe Pacific Northwest. University of IdahoPress, Moscow. 332 pp.
Peabody, F.E. and J.M. Savage. 1958. Evolu-tion of a Coast Range corridor in Californiaand its effect on the origin and dispersal ofliving amphibians and reptiles. Pp. 159-186in C.L. Hubbs, ed., Zoogeography. Publi-
cation 51, American Association for theAdvancement of Science, Washington, D.C.
Raphael, M.G. 1984. Wildlife populations inrelation to stand age and area in Douglas-firforests of northwestern California. Pp.259-274 in W.R. Meenhan, T.R. Merrell Jr.and T.A. Hanley, tech. eds., Fish and Wild-life Relationships in Old-growth Forests.American Institute of Fisheries ResearchBiologists, Juneau, Alaska.
Richter, B.D., D.P. Braun, M.A. Mendelson,and L.L. Master. 1997. Threats to imperiledfreshwater fauna. Conservation Biology11:1081-1093.
Seale, D.B. 1980. Influence of amphibian lar-vae on primary production, nutrient flux,and competition in a pond system. Ecology61:1531-1550.
Semlitsch, R. 1999. Biological delineation ofterrestrial buffer zones for pond-breedingsalamanders. Conservation Biology12:1113-1119.
Sjogren, P. 1991. Extinction and isolation gra-dients in metapopulations: the case of thepool frog (Rana lessonae). Biological Jour-nal of the Linnean Society 42:135-147.
Stebbins, R.C. 1985. A Field Guide to WesternAmphibians and Reptiles. 2nd Ed. Hough-ton Mifflin, Boston, Mass.
Thomas, J.W., M.G. Raphael, R.G. Anthony,and others. 1993. Viability assessment andmanagement considerations for species as-sociated with late successional and old-growth forests of the Pacific Northwest.Report by Scientific Analysis Team, U.S.Department of Agriculture, Forest Service,Portland, Ore. 530 pp.
Tyler, T., W.J. Liss, L.M. Ganio, G.L. Larson,R. Hoffman, E. Deimling, and G. Lom-nicky. 1998. Interaction between introducedtrout and larval salamanders (Ambystomamacrodactylum) in high-elevation lakes.Conservation Biology 12:94-105.
Welsh, H.H. Jr. and A.J. Lind. 1995. Habitatcorrelates of the Del Norte salamander,Plethodon elongatus (Caudata: Plethodon-tidae), in northwestern California. Journalof Herpetology 29:198-210.
Whittaker, R.H. 1956. Vegetation of the GreatSmoky Mountains. Ecological Monographs26:1-80.
Whittaker, R.H. 1960. Vegetation of the Siskiy-ou Mountains, Oregon and California. Eco-logical Monographs 30:279-338.
Wilbur, H.M., P.J. Morin, and R.H. Harris.1983. Salamander predation and the struc-ture of experimental communities: anuranresponses. Ecology 64:1423-1429.
Appendix 1. Species of native amphibians found in the Klamath-Siskiyou region insouthwestern Oregon and northern California. This list was compiled from Nuss-baum et al. (1983), Stebbins (1985), and the Integrated Taxonomic InformationSystem (http://www.itis.usda.gov/itis/).
Appendix 2. Species of natives reptiles found in the Klamath-Siskiyou region insouthwestern Oregon and northern California. This list was compiled from Nuss-baum et al. (1983), Stebbins (1985), and the Integrated Taxonomic InformationSystem (http://www.itis.usda.gov/itis/).
