WESTERN BIRDS

Volume 23, Number 1, 1992

A MAXIMUM ESTIMATE OF THE CALIFORNIA GNATCATCHER'S POPULATION SIZE IN THE UNITED STATES

JONATHAN L. ATWOOD, Manomet Bird Observatory,P.O. Box 1770, Manomet, Massachusetts 02345

The CaliforniaGnatcatcher, Polioptila californica, was recentlyrecog- nizedas a speciesdistinct from the widespreadBlack-tailed Gnatcatcher, P. melanura, of the southwesterndesert regionsof the United States and Mexico (Atwood 1988, American Ornithologists'Union 1989). Although CaliforniaGnatcatchers are distributedthroughout much of BajaCalifornia, the northernmostsubspecies, P. c. californica, now occursonly in remnant fragmentsof coastalsage scrub habitat from LosAngeles County, Califor- nia, southto El Rosario,Baja California(Atwood 1991). Atwood (1980) speculatedthat the numberof CaliforniaGnatcatchers remainingin the UnitedStates was "no more than 1,000 to 1,500 pairs," from estimatesof 30 pairs in VenturaCounty, 130 pairs in Los Angeles County,50 pairsin San BernardinoCounty, 325 pairsin OrangeCounty, 400 pairsin RiversideCounty, and 400 pairsin San Diego County.These valueswere derivedfrom reportsof variousobservers, limited field work in differentportions of the species'range, and visualestimates of habitat availabilityin differentareas. Despite the preliminarynature of theseresults, the patternof continuinghabitat loss evident at that time indicated"imme- diate concern for the survival"of P. c. californica in the United States (Atwood 1980). Extensive destruction of suitable California Gnatcatcher habitat has con- tinuedsince 1980 (Rea and Weaver 1990). Of 56 southernCalifornia sites where Atwood (1980) recordedcoastal sage scrub vegetation and Califor- nia Gnatcatchers,18 (32%) had been completelydestroyed by urban developmentby 1990, and 15 (27%) had been partially degradedby current constructionprojects (J. Atwood and K. Shields,unpubl. data). Similarhabitat loss, primarily associated with agriculturaldevelopment, has alsoincreased in northwesternBaja California;Best (1983) foundthat the San Quintin KangarooRat (Dipodomysgravipes), which historicallyoc- curredalong Baja California'sPacific coast from San Telmo southto El Western Birds 23:1-9, 1992 1 CALIFORNIA GNATCATCHER'S POPULATION SIZE

Rosario,"has been virtually eliminated from mostof its nativerange during the past decade[owing] to extensivecultivation." Because of continuing habitatloss throughout the gnatcatcher'sUnited Statesand Mexicanrange, federaland Californiaendangered species status was recently proposed for the CaliforniaGnatcatcher's northern subspecies. As an effortto providea baselinefor futurestudies, in thispaper I estimatethe maximumpopulation of P.c. californica that mightcurrently exist in the United States.

METHODS

Determining numbersof a small passerinewhose populationdensity variesthroughout a relativelyextensive geographic range is difficult.Be- causeCalifornia Gnatcatchers are confinedto coastalsage scrub vegetation (Atwood 1980), populationestimates could, in theory, be derived from measurementsof the availableacreage of this habitat coupledwith the species'average territory size. However, many areasof vegetationthat are correctlyclassified as coastalsage scrub are not occupiedby gnatcatchers, suggestingthat population estimatesbased on vegetationmapping will invariablybe inflated.Direct censusing of gnatcatchersthroughout southern Californiahas also proven impossible, because public access to many large tracts of private land has been denied. Althoughsome censusdata have beencollected by the southernCalifornia development industry, little of this informationhas been released to the public,and independentverification of theseresults has beenprevented. In this analysis,I use a more generalapproach to estimatecrudely the maximum number of California Gnatcatchersremaining in the United States. Atwood and Bolsinger(in press)showed that in Los Angeles, Orange, and San Diego counties,93% of recent CaliforniaGnatcatcher observationswere made at elevationsunder 250 m; in RiversideCounty, most recent records (75%) were made at elevations under 500 m. Here I use these elevationsto approximatethe maximum range of California Gnatcatchersin each of thesecounties. Using the geographicinformation systemCAMRIS (ComputerAided Mappingand ResourceInventory Sys- tem; R. Glenn Ford, EcologicalConsulting, Inc., Portland,Oregon) and elevation contours digitized from U.S. Geological Survey (USGS) 1:100,000 scaletopographic maps, I calculatedthe extent of land lying below 250 m elevationin Los Angeles,Orange, and westernSan Diego counties,and below500 m in westernRiverside County. Next, againusing thesetopographic maps, I examinedall 1-minuteblocks of latitude-longi- tude below these elevations,and entered into CAMRIS those blocksthat were predominatelydesignated as "BuiltUp" or as "Orchard,Vineyard." Finally, I calculatedthe remainingamount of undevelopedlowland that might support California Gnatcatchersin each county. Although P.c. californica historicallyoccurred in Ventura and San Bernardinocounties, Atwood(1980) foundthat the specieshas been largelyor entirelyextir- pated from theseareas; consequently, these counties were excludedfrom analysis.Similarly, I excludedfrom area calculationsthose lowland portions of Los AngelesCounty located near the SantaMonica Mountains, where CaliforniaGnatcatchers have never been reliably reported (Atwood 1980). CALIFORNIA GNATCATCHER'S POPULATION SIZE

USGS maps (1:100,000 scale)used in this analysisincluded (followed by publicationdate and, in brackets,dates of aerial photographyused in the most recent revisions):Borrego Valley, 1982 [1976-77]; El Cajon, 1979 [1975]; Long Beach, 1981 [1979]; Los Angeles,1979 [1976]; Oceanside, 1982 [1972-74, 1979]; Palm Springs,1984 [1975, 1979]; San Bernar- dino, 1982 [1975, 1978]; San Diego, 1979 [1972, 1975-77]; SantaAna, 1983 [1978-79]; Santa Barbara, 1982 [1977-78]. Focusedsurveys conducted during 1990 by the U.S. Fish and Wildlife Service(L. Salata, unpubl. data) and U.S. Navy (T. Burr, unpubl. data) locatedapproximately 175-200 pairsof CaliforniaGnatcatchers on Camp PendletonMarine Corps Base,San Diego County,and 25-50 pairson the adjacentFallbrook Naval Weapons Station. On the basisof topographic map designationsof "BuiltUp" and "Orchard"areas, approximately 405 km2 of undevelopedland lie below250 m elevationon thesetwo military bases.These acreagemeasurements and populationestimates suggest that CaliforniaGnatcatchers occur in undevelopedlowlands of coastalsouthern Californiaat densitiesranging from 0.49 to 0.62 pairsper squarekilome- ter. Note that these values do not refer to population densitieswithin particularareas of coastalsage scrub that are occupiedby gnatcatchers,or to densitieswithin pure standsof occupiedand unoccupiedcoastal sage scrub.Rather, the "habitat"definition used here is simply"low-elevation, undeveloped"land. I usedthese densitiesin conjunctionwith the measure- mentsof undevelopedacreage in each countyto estimatethe population sizeof CaliforniaGnatcatchers remaining in the United Statesduring the late 1970s. Assumptionsinherent in this analysissuggest that current (1991) populationlevels cannot significantly exceed these values. Unpublisheddocuments cited in the text have been depositedin the JosselynVan Tyne MemorialLibrary of the WilsonOrnithological Society, Universityof Michigan,Museum of Zoology,Ann Arbor, Michigan48109.

RESULTS AND DISCUSSION

Figure1 showsthe distributionof undevelopedlowland within the United Statesrange of the CaliforniaGnatcatcher during the mid- to late 1970s, when the USGS maps on which my analysisis based were compiled. Multiplyingthe acreageof thisarea by the densitiesderived from surveyson Camp Pendletonand Fallbrookyields populationestimates of 1811 to 2291 pairs(Table 1). These resultsare similarto Atwood's(1980) popula- tion estimate(1000-1500 pairs),which was basedon differenttypes of informationthat were obtainedshortly after the time period considered here. This analysisassumes that no CaliforniaGnatcatchers occur above the elevationallimits used to define the species'potential distribution. In fact, small numbers of P c. californica do occur above 250 m elevation in Orange and San Diego counties,and above 500 m elevationin Riverside County(Atwood and Bolsinger,in press).Use of elevationas an indexof gnatcatcherdistribution thus tends to underestimatethe total number of birdsremaining in the United States.However, this inaccuracyis more than offsetby four other assumptionsthat inflate estimatesof the population 3 CALIFORNIA GNATCATCHER'S POPULATION SIZE size, and suggestthat fewer than approximately2000 pairs currently remain north of Mexico. First, the amount of land currentlydeveloped in southernCalifornia is certainlygreater than indicatedon maps preparedduring the mid- to late 1970s. For example,the RiversideCounty PlanningDepartment (1990) noted that "from 1980 throughJanuary 1, 1985, the county'spopulation grewfrom 663,166 to 800,949 people,an increaseof 137,783 or 20.8%. From January1, 1985, to January1, 1990, the populationgrew... to 1,110,021 people,an increaseof... 38.6%." Comparisonsof 1980 and

LOSANGELES CO.-- RIVERSIDECO.

ORANGE

SANDIEGO Go.

SOUTHERN CALIFORNIA

mid-late 1970's

0 10 20 30 40 50 ' ..... Ki'lomet;r9 ' ' I UNDEVELOPEDLOWLAND

117 30

Figure1. Extentof undevelopedlowland in coastalsouthern California during the mid-to late 1970s. Basedon (a)designations of "BuiltUp" and "Orchard,Vineyard" areasshown on USGS 1:100,000 scaletopographic maps and (b)the 250-m (Los Angeles,Orange, San Diego counties)or 500-m (RiversideCounty) elevation con- tours. CALIFORNIA GNATCATCHER'S POPULATION SIZE

1990 U.S. CensusBureau data indicate that fourof the ten fastestgrowing major cities in the United States (Escondido,Oceanside, Rancho Cucamonga,and Chula Vista) are locatedin southernCalifornia; more than 90,000 people movedto the city of Moreno Valley between 1980 and 1990, representingan increaseof 322%. All of thesecities are locatedin areas from which there are historic or recent records of California Gnat- catchers(Atwood 1980, 1990); however,none of this humanpopulation growth,or its associateddestruction of naturalhabitats, is reflectedin the calculationsof undevelopedacreage that were usedhere as the basisfor gnatcatcherpopulation estimates. Also, the USGS topographicmaps on whichthis analysis was based failed to identifyagricultural lands other than orchardsor vineyards;thus, many areasidentified here as "undeveloped" may, in fact, havebeen planted in a varietyof floweror vegetablecrops. Second,gnatcatcher surveys based on unbandedindividuals, such as the censusesconducted on CampPendleton and Fallbrook, frequently overesti- mate the numberof birdsactually present (e.g., Lettieri-Mclntyreand Associates1987, ERC Environmentaland EnergyServices Co. 1990). In general,this tendencymay reflectthe surprisinglyextensive movements thatsome pairs of gnatcatchersundertake; the averagehome range size of color-bandedbirds in SanDiego County has been reported as greater than 18 acres,and the breeding season home range of onepair was documented at 45 acres(ERC Environmentaland EnergyServices Co. 1990, Mock et al. 1990, P. Mockpers. comm.). Consequently, the densities implied by the CampPendleton and Fallbrook surveys may be too high,thereby inflating the final populationestimate.

Table 1 MaximumPopulation Estimates of CaliforniaGnat- catchers in the United States Undeveloped PopulationEstimates b County Lowland(km2) ø 0.49 pairs/km2 0.62 pairs/km•

Los Angeles 48 24 30 Orange 458 224 284 Riverside 1478 724 916 San Diego 1712 839 1061 Total 3696 1811 2291

aBasedon delineationsof "BuiltUp" and"Orchard, Vineyard" areas as shown on 1:100,000 scaleUSGS topographicmaps prepared in the mid-to late 1970s. In LosAngeles, Orange, and San Diegocounties, the 250-rn elevationcontour was used to define the maximum extent of suitable California Gnatcatcher habitat;in RiversideCounty, the 500-rn elevationcontour was used(Atwood and Bolsingerin press). bDensitiesbased on surveysof CampPendleton Marine Corps Base (L. Salata, U.S. Fish and WildlifeService, unpubl. data) and FallbrookNaval Weapons Station(T. Burr, U.S. Navy, unpubl.data). CALIFORNIA GNATCATCHER'S POPULATION SlZE

Third, of 405 km2 of undevelopedland located below 250 m elevationon Camp Pendletonand Fallbrookmilitary bases, general vegetation maps preparedby Oberbauer(1977) showedthat 346 km2 (85%) were covered by "coastalsage scrub" or "grassland"vegetation types (Figure 2); Atwood (1980) found that Oberbauer's definitions of both of these habitats fre- quentlyincluded areas occupied by CaliforniaGnatcatchers. In contrast,of

ß'';•1111111•11;i •'-" l(il'omet'ers ' I SUITABLEHABITAT '•'::•UNSUITABI.E HABITAT • DEVELOPEDAREAS ELEq•ATIONß250 m

ß . ::: ::::: :.:::: :.::::::: :: ::::: :.:: ::::: ::: :: ::: :::

......

...... '.'.'.'.-.....'...-...... ,...,......

','.'.'....,.,-.'.'.-.-.,...... -.'.....

SAN DIEGO ......

:::::COASTAL,:.:.:.:.:.:.:-:.:.:.:.:. :i:'$AN DIEGO CO.!:!:!

......

Figure2. Relativeabundance of coastalsage scruband grasslandvegetation in lowland(<250 m elevation),undeveloped portions of CampPendleton Marine Corps Base/FallbrookNaval Weapons Station and the remainderof westernSan Diego County. Boundariesof Camp PendletonMarine Corps Base and FallbrookNaval WeaponsStation shown in northwestcorner of county.Based on (a) vegetation mappingby Oberbauer(1977) and (b) designationsof "Built Up" and "Orchard, Vineyard"areas shown on USGS 1:100,000 scaletopographic maps. CALIFORNIA GNATCATCHER'S POPULATION SIZE

1307 km2 of undevelopedland in lowlandportions of westernSan Diego Countyaway from thesemilitary bases, only 304 km2 (23%) supported habitatsidentified by Oberbauer(1977) as coastalsage scrub or grassland (Figure2). Theseresults indicate that, in general,the undevelopedlowlands of Camp Pendletonand Fallbrookprobably include more extensivegnat- catcherhabitat than does most of the restof undevelopedcoastal San Diego County. Therefore,population densities derived from gnatcatchersurveys on Camp Pendletonand Fallbrookprobably exceed, on average, the densitiescharacteristic of other lowland,undeveloped regions of coastal southernCalifornia. Application of the Camp Pendleton/Fallbrookdensi- tiesto lesssuitable portions of the species'range would predictably overes- timatethe true populationsize of CaliforniaGnatcatchers. Fourth,this analysisassumes that the densityof CaliforniaGnatcatchers in inlandportions of the species'range is similarto that of coastalareas suchas Camp Pendletonand Fallbrook.In fact, densitiesof P.c. californica are generallylower farther inland (Mock et al. 1990; Atwood,unpubl. data). Applicationof densityvalues derived from coastalsites thus inflatesesti- matesof the true numberof CaliforniaGnatcatchers remaining in Riverside Countyand inlandSan Diego County. Althoughpopulation estimates provide a usefulstarting point for further investigationand continuedmonitoring, the exact numberof remaining CaliforniaGnatcatchers is, in manyways, a secondaryissue with regardto the species'status in the United States.Given that (1) approximately90% of southern California'shistoric distribution of coastalsage scrub has already been destroyedby urbanization(Hanes 1976, Kirkpatrick and Hutchinson1977, Mooney 1977, Westman1981, 1987, O'Leary 1990), (2) most remainingUnited Statespopulations of California Gnatcatchers are locatedon privatelands where urban development is imminent(Atwood 1990), and (3) increasing habitat fragmentation may inhibit the gnatcatcher'sdispersal and reduceits reproductivesuccess (e.g., Soul• et al. 1988, Bolgeret al. 1991), the long-termsurvival of P.c. californica in the United Stateswould be questionableeven if there were currentlymany thousandsof breedingpairs remaining north of Mexico.Without reversal of present land use trends, California Gnatcatcherswill be extirpatedfrom most or all of their range in the United Statesin the near future.

SUMMARY

Analysisby meansof a geographicinformation system of the extentof undevelopedlowland in coastalsouthern California, coupled with density indicesderived from recentsurveys of Camp PendletonMarine Corps Base and FallbrookNaval Weapons Station, indicates that 1811-2291 pairs of California Gnatcatchers remained in the United States in the late 1970s. Variousassumptions inherent in this analysisimply that thesevalues are inflated and that the species'current populationsize north of Mexico cannotsignificantly exceed 2000 pairs.Continuing loss and fragmentation of the CaliforniaGnatcatcher's required coastal sage scrub habitat, of which most remaining tracts occur on private propertiesthreatened by urban CALIFORNIA GNATCATCHER'S POPULATION SIZE development,will eliminatethe speciesfrom muchor all of its UnitedStates range in the immediatefuture.

ADDENDUM

On 1 August1991, publictestimony was received by the CaliforniaState Fish and Game Commissionregarding the petition to add the California Gnatcatcherto the state'slist of endangeredspecies. During this hearing, lobbyistsfor the southernCalifornia development industry presented the resultsof preliminarygnatcatcher population surveys conducted within the lastfew years(H. L. Jones, 1991, A rangewideassessment of the Califor- nia Gnatcatcher(Polioptila californica), unpublishedreport preparedfor the BuildingIndustry Association of SouthernCalifornia, Santa Ana). No details were given concerningmethods used in surveys,specific areas censused,or the rationaleused in extrapolatingestimated totals from actual observations.Because only observersemployed by the developmentindus- try havebeen allowed access to manyimportant tracts of privateland, these resultshave not been independentlyverified. Nonetheless, to provideall currentlyavailable information regarding the populationsize of California Gnatcatchersin the United States, I repeat Jones' estimateshere: Los AngelesCounty, 11-20 pairsactual census results, 20-30 pairsestimated total; Orange County, 245 pairs actual censusresults, 325-350 pairs estimatedtotal; San Diego County, 830 pairsactual census results, 1000- 1100 pairs estimatedtotal; Riverside County, censusdata unavailable, 300-400 pairs estimatedtotal. Jones concludesthat "we have conserva- tively estimatedapproximately 1645 to 1880 pairs of [California]gnat- catcherswithin the United States."This value is approximately10-20% lower than the maximumpopulation estimates (1811-2291 pairs)derived above.

ACKNOWLEDGMENTS

J. Bolsinger,C. Cady, B. Hamilton, D. Heinemann, and M. Thrift all provided valuableassistance. Comments from M. Evans,P. Mock, and P. Unitt improvedthe manuscript,as did discussionswith D. Bontrager.This work was supportedfinan- cially by the membershipof Manomet Bird Observatoryand by grants from the Natural ResourcesDefense Council and the ShirleyWells ConservationFund of the El DoradoAudubon Society, Long Beach,California.

LITERATURE CITED

AmericanOrnithologists' Union. 1989. Thirty-seventhsupplement to the American Ornithologists'Union Check-list of NorthAmerican birds. Auk 106:532-538. Atwood, J. L. 1980. The United Statesdistribution of the CaliforniaBlack-tailed Gnatcatcher. W. Birds 11:65-78. Atwood,J. L. 1988. Speciationand geographicvariation in black-tailedgnatcatch- ers. Ornithol. Monogr. 42. Atwood, J. L. 1990. Status review of the California Gnatcatcher (Polioptila californica). Unpubl. report, Manomet Bird Observatory,P.O. Box 1770, Manomet, MA 02345. CALIFORNIA GNATCATCHER'S POPULATION SlZE

Atwood, J. L. 1991. Subspecieslimits and geographicpatterns of morphological variationin CaliforniaGnatcatchers (Polioptila californica). Bull. S. Calif.Acad. Sci. 90: in press. Atwood,J. L., and Bolsinger,J. S. In press.Elevational distribution of California Gnatcatchers in the United States. J. Field Ornithol. Best, T. L. 1983. Morphologic variation in the San Quintin Kangaroo Rat (Dipodomysgravipes Huey, 1925). Am. MidlandNat. 109:409-413. Bolger,D. T., Alberts,A. C., and SoulS,M. E. 1991. Occurrencepatterns of bird speciesin habitatfragments: Sampling, extinction, and nestedspecies subsets. Am. Nat. 137:155-166. ERC Environmentaland EnergyServices Co. 1990. Phase I report, Amber Ridge CaliforniaGnatcatcher study. Unpubl. report preparedfor Weingarten,Siegel, FletcherGroup, Inc. ERC Environmentaland EnergyServices, 5510 Morehouse Dr., San Diego, CA 92121. Hanes,T. L. 1976. Vegetationtypes of the San GabrielMountains, Plant Communi- tiesof SouthernCalifornia (J. Lating,ed.), pp. 65-76. Calif. NativePlant Soc. Spec. Publ. 2. Kirkpatrick,J. B., and Hutchinson,C. E 1977. The communitycomposition of Californiacoastal sage scrub. Vegetatio 35:21-33. Lettieri-Mclntyreand Associates.1987. Draft environmentalreport for Amber Ridge,TM 4685, EAD Log No. 87-19-34. Unpubl.report. Lettieri-Mclntyre Assoc.,533 F St., San Diego, CA 92101. Mock, P. J., Jones, B. L., Grishaver,M., Konecny,J., and King, D. 1990. Home rangesize and habitatpreferences of the CaliforniaGnatcatcher in San Diego County. Presentedat the AmericanOrnithologists' Union and Cooper'sOrni- thologicalSociety Joint AnnualMeeting, Los Angeles,CA. Mooney,H. A. 1977. Southerncoastal scrub, in TerrestrialVegetation of California (M. G. Barbourand J. Major, eds.),pp. 471-489. Wiley, New York. Oberbauer,T. A. 1977. Countyof San Diego GeneralizedVegetation [map]. Map- ping Section,Dept. of Transportation,County of San Diego, 5555 Overland Ave., San Diego, CA 92123. O'Leary, J. E 1990. Californiancoastal sage scrub:General characteristicsand considerationsfor biologicalconservation, in EndangeredPlant Communities of SouthernCalifornia (A. A. Schoenherr,ed.), pp. 24-41. S. Calif. Botanists Spec. Publ. 3. Rea, A.M., and Weaver, K. L. 1990. The taxonomy,distribution, and statusof coastal California Cactus Wrens. W. Birds 21:81-126. RiversideCounty Planning Department. 1990. Newsletter,"Planning News," Vol. 1, No. 1, January. SoulS,M. E., Bolger,D. T., Alberts,A. C., Wright,J., Sourice,M., and Hill, S. 1988. Reconstructeddynamics of rapid extinctionsof chaparral-requiringbirds in urban habitat islands. Conserv. Biol. 2:75-92. Westman,W. E. 1981. Diversityrelations and successionin Californiancoastal sage scrub.Ecology 62:170-184. Westman,W. E. 1987. Implicationsof ecologicaltheory for rare plantconservation in coastalsage scrub, in Proceedingsof the Conferenceon Conservationand Managementof Rare and EndangeredPlants (T. S. Elias,ed.), pp. 133-140. Calif. Native Plant Soc., Sacramento. Accepted 25 July 1991 California Gnatcatcher Sketch by Lou Barnicle

10