Elevational Distribution of California Gnatcatchers in the United States

J. Field Ornithol., 63(2):159-168

ELEVATIONAL DISTRIBUTION OF CALIFORNIA GNATCATCHERS IN THE UNITED STATES

JONATHANL. ATWOODAND JEFFREYS. BOnSINOER MahometBird Observatory P.O. Box 7770 Mahomet, Massachusetts023d5 USA

Abstract.--Knownoccurrences of CaliforniaGnatcatchers (Polioptila californica) in the United Stateswere mappedrelative to the 250-m, 500-m and 750-m elevationcontours of southern California. Of 324 siteswhere this seriouslythreatened species was recordedfrom 1960 to 1990, 84% were located below 250 m elevation, and 97% occurredbelow 500 m. A total of 69 historic(pre-1960) sitesshowed a similar elevationalpattern. These resultsindicate that previousestimates of the rangeof the northern,nominate subspecies were probablyexcessive, and that protectionof coastalsage scrub habitat locatedabove 250 m (in regionswithin 35 km of the coast)or 500 m (inland regions)may not significantlycontribute to California Gnatcatcherconservation efforts. Furthermore, using the 250-m and 500-m elevationcontours as approximateindices of the potential extent of the species'range in northwestern Baja California, approximately65-70% of the historicdistribution of P. c. californicamay have been located in the United States.

DISTRIBUCION ALTITUDINAL DE POLIOPTILA CALIFORNICA EN LOS ESTADOS UNIDOS Sinopsis.--Selocalizaron en mapas los lugaresen dondeha sidoinformada en territorio de los EstadosUnidos la Perlita de California (Polioptilacalifornica) en relaci6na loscontornos de elevaci6nde 250 m, 500 m, y 750 men el sur de California.De los 324 sitiosen donde estaespecie (seriamente amenazada) fue encontradaentre 1960 y 1990, el 84% correspondi6 a localidadespor debajode los 250 m de elevaci6n,y el 97% por debajode los 500 m. Un total de 69 registroshist6ricos (antes de 1960) muestranun patron similar. Estosresultados indican que los estimadosde la distribuci6nde la subespecienominal (del norte), fueron probablementeexcesivos, y que la protecci6ndel habitatde matorralcostero localizado por debajode los 250 m (en regionesdentro de un radio menor a los 35 km de la costa),o 500 m (en regionesinteriores) no contribuir•n significativamentea los esfuerzos de conservaci6n de esta especie.AdemJs, usando los contornosde 250 mm y 500 m como indicadores potencialesde la distribuci0nde estaespecie en el noroestede Baja California, aproxima- damenteel 65-70% de la distribuci6nhist6rica de P.c. californicaprobablemente ha estado localizada en los Estados Unidos.

The California Gnatcatcher (Polioptila californica)was recently rec- ognized as specificallydistinct from the Black-tailed Gnatcatcher (P. melanura) of the Sonoran and Chihuahuan deserts of North America (American Ornithologists' Union 1989, Atwood 1988). Although com- monly and widely distributedin many of the lowland, arid regionsof Baja California, the currentUnited Statesrange of the speciesis extremely localized,being entirely restrictedto small fragmentsof coastalsage scrub that occurin the increasinglyurbanized southernCalifornia countiesof Los Angeles, Orange, Riverside and San Diego (Atwood 1980, 1990). Kirkpatrick and Hutchinson (1977:21) describedcoastal sage scrub as "one of the least known and fastestdisappearing types of vegetationin Galifornia," and Westman (1987:134) believed it to be "one of the most endangeredhabitat types in the nation." Federal and State endangered

159 160] J. L. Atzvoodand J. S.Bolsinger J.Field Ornithol. Spring 1992 specieslisting of the California Gnatcatcher'snorthern subspecieshas beenrecently petitioned by the Natural ResourcesDefense Council and Manomet Bird Observatory. The California Gnatcatcher'snorthern, nominate subspecies,which occursfrom coastalsouthern California south to 30øN latitude in Baja California (American Ornithologists'Union 1957, Atwood 1991), is sim- ilarly threatenedin its Mexican range by habitat losscaused by agricul- ture, grazing and urban development(Best 1983, P. A. Bowler, pers. comm.).Michael BrandmanAssociates (1991) reportedthat no California Gnatcatcherswere found in 1990 betweenTijuana and Santo Tomas, a region of northwesternBaja California that representsapproximately 40% of P. c. californica'shistoric, latitudinal range in Mexico. Atwood (1980, 1988) previouslystated that P. c. californicawas dis- tributedmainly below elevations of 600 or 1000 m. Thesevalues, however, were basedon the generallimits of coastalsage scrub vegetation as defined by plant ecologists(Munz and Keck 1959), and failed to considerthe possibilitythat California Gnatcatchersmight not occurthroughout all portionsof this floristicallyvariable (DeSimone 1989, O'Leary 1990, Westman 1983) plant community.In the present study we use current distributional information to examine the elevational limits of California Gnatcatcherdistribution in the United States.Also, we apply theseresults to the questionof whetherthe historicrange of P. c. californicawas located primarily in southernCalifornia or in Baja California, a potentially important issuein future discussionsof whether this subspeciesshould be grantedprotection under the United StatesEndangered Species Act.

METHODS Data concerninghistoric (pre-1960) and recent (1960-1990) records of California Gnatcatchers in southern California were obtained from museumspecimens, literature accounts,public environmentalreview doc- uments,field notesof ornithologistscurrently activein southernCalifor- nia, and the Natural Diversity Data Baseof the California Department of Fish and Game (Atwood 1990). These data permit a more extensive analysisof California Gnatcatcherdistribution in the United Statesthan that providedby Atwood (1980). Observationswere locatedto the nearest 1-min block of latitude and longitude,hereafter referred to as a "site." Latitude-longitudecoordinates for historicrecords, which generallylacked preciselocality data, could in most casesonly be approximated. Visual presentationof data and calculationof land areaswere accom- plished using the geographicinformation systemCAMRIS (Computer Aided Mapping and ResourceInventory System;Ecological Consulting, Inc., Portland, Oregon). The 250-m, 500-m, and 750-m elevationalcon- tours were digitized from United States GeologicalSurvey (USGS) 1:100,000scale topographic maps; vegetation data for southernOrange County were based on the 1:125,000 scale Corona Quadrangle map prepared by the United StatesForest Service(1940). Unpublished documentscited in the text have been depositedin the Vol.63, No. 2 CaliforniaGnatcatcher Distribution

Wilson OrnithologicalSociety's Josselyn Van Tyne Memorial Library, Universityof Michigan Museumof Zoology,Ann Arbor, Michigan 48109.

RESULTS AND DISCUSSION The distribution of California Gnatcatchers in the United States is definedby relatively narrow elevationallimits (Fig. 1). Of 324 sitesof recent occurrence,272 (84%) were located below 250 m elevation, 315 (97%) below 500 m, and 324 (100%) below 750 m (Table 1). The elevationsof 69 approximately-locatedrecords of historicoccurrence were consistentwith those of recent observations;65 (94%) of these sites occurred at elevations <500 m. In areasof LosAngeles County, OrangeCounty, and San Diego County located <35 km from the coast, 269 of 289 recent sites (93%) occurred at elevations<250 m. In Riverside County, located >35 km inland, 24 of 32 recentsites (75%) occurredat <500 m elevation,but only 1 (3%) occurredbelow 250 m. Destructionof coastalsage scrub vegetation from mostlowland portionsof this countyhas probably biasedrecent observations of California Gnatcatchers in this area toward elevations >250 m; nonetheless,the distributionof historicrecords from RiversideCounty suggeststhat the upper elevationallimit of P. c. califorrticain inland areas is higher than in coastalregions (Table 1). The exactecological factors controlling the distributionof the California Gnatcatcher'snorthernmost, nominate subspecies are presntlyunknown. Throughout its range P. c. califorrticais obligatelyassociated with coastal sagescrub (Atwood 1980, 1988). Many areas of coastalsage scrub do not support California Gnatcatchers,however, and the range of P. c. califorrticais dramatically smaller than might be predictedfrom maps delimiting the distributionof this vegetationtype (e.g., Mooney 1977). For example, the Venturan coastalsage scrub association(Kirkpatrick and Hutchinson1977; Westman 1983), althoughincluding areas of known California Gnatcatcheroccurrence in southernVentura County and coastal Los AngelesCounty, was alsodescribed as extendingnorth into Santa Barbara and San Luis Obispo counties,well beyondthe northernmost limit of P. c. califorrtica(American Ornithologists'Union 1957, Atwood 1991). Even within the subspecies'range, not all areas of coastalsage scrub are occupiedby gnatcatchers.For example, sizable portionsof coastal sagescrub exist in southernOrange County at elevations>250 m; however, few California Gnatcatcher records are known from most of these areas (Fig. 2), despiteconsiderable effort to locate the species(D. R. Bontrager,pers. comm.). Nonetheless,elevation appears to be at least one factorthat may serve to crudelyevaluate gnatcatcher habitat suitability. We suspectthat some type of transitionin the structureand/or speciescomposition of coastal sagescrub occurs near 250 m elevationin regionslocated <35 km from the coast,and that areas of this vegetationtype that are locatedabove this elevationare generally unsuitablefor California Gnatcatchers.In 162] d. L. Atwoodand J. S.Bolsinger J.Field Ornithol. Spring 1992

AngelesCo. •an BernardinoCo. C0.i

co.i

•i•i•?!•RlversldeCo.

SOUTHERN CALIFORNIA

0 10 20 30 40 50 60 70 ...... Kil•rne•ers ' ' •glcvation< 250 m :i:'•Elevation250-500 m '/•Elevation500-750 m Elevationß750 m GnatcatcherSite i I i 117 ,50

FIGURE 1. Distribution of recent and historic California Gnatcatcher records relative to threemajor elevation contours. Sites are mappedto the nearest1-min blockof latitude- longitude.

more inland regionsthis postulatedhabitat changeoccurs near 500 m elevation. Further studyof the preciseecological requirements of ?. c. californica is needed,including possible floristic changes that may occurin coastal sage scrub at different elevations.Kirkpatrick and Hutchinson (1977, 1980) and Westman(1981) did not identifyelevation as a significant predictorof mostdominant plant speciesin the coastalsage scrub community. Many of the samplinglocalities used by theseauthors were not within the range of the California Gnatcatcher,however, and theseanal- ysesmay have failed to detectecological patterns relevant to the distributionof P. c. californica.Furthermore, although elevation may provide a broadindex of habitat suitability,many areasof coastalsage scrub that are locatedbelow the 250 m or 500 m "limits" are alsonot occupiedby gnatcatchers.Reasons for the species'absence from suchsites are presently unknown,but may include a variety of factorssuch as slope,isolation Vol.63, No. 2 CaliforniaGnatcatcher Distribution [ 163

TABLE 1. Elevational distribution of recent and historic sites of California Gnatcatcher occurrence in southern California.

Elevation b County Typea <250 m 250-500 m 500-750 m Los Angeles Recent (n = 11) 91 9 0 Historic (n = 19) 63 32 5 Orange Recent (n = 68) 88 12 0 Historic (n -- 2) 100 0 0 Riverside Recent (n = 32) 3 72 25 Historic (n = 11) 45 27 27 San Bernardino Recent ------Historic (n -- 9) 0 100 0 San Diego Recent (n = 213) 94 5 1 Historic (n -- 23) 91 9 0 Ventura Recent ------Historic (n = 5) 60 40 0 Total Recent (n -- 324) 84 13 2 Historic (n -- 69) 62 32 6

• Recent(1960-1990) and Historic (pre-1960) sitesof California Gnatcatcheroccurrence. Total numberof 1-min blocksof latitude-longitude(n) givenin parentheses.Locations of historicsites are approximate;all recentsites were exactly locatedon USGS topographic maps. • Percent of total sites (n).

from adjacentpatches of coastalsage scrub, fire history,bordering habitat type, or Brown-headedCowbird (Molothrusater) abundance. Despite the fact that questionsremain regarding the details of Cali- fornia Gnatcatcherhabitat requirements,available data nonethelesshave important implicationsfor conservationplanning effortsin southernCal- ifornia. In Orange County and the coastalareas of Los Angelesand San Diego counties,coastal sage scrub located above 250 m is unlikely to supportpopulations of P.c. californica.In RiversideCounty and inland portionsof Los Angelesand San Diego counties,California Gnatcatchers generally do not occur higher than approximately 500 m. Protectionof coastalsage scrub abovethese elevations,although potentially valuable from the standpointof maintaining connectivecorridors between other- wise isolatedpatches of lower elevationhabitat, may contributelittle to the long-term maintenanceof viable gnatcatcherpopulations. Much of southernCalifornia's remaining coastalsage scrub, especially that occurringat low elevations,is privatelyowned and subjectto intense developmentpressures (Atwood 1990). Although a growing awareness of the conservationneeds of the coastalsage scrub ecosystem has prompted increasinglyfrequent attempts to mitigatefor habitat losscaused by con- structionactivities, such efforts have often failed to protect adequately 164] J. L. Atwoodand J. S.Bolsinger J.Field Ornithol. Spring 1992

0 1 2 3 4 ,5 6 ...... Kilo•eter• ' ""•CoastalSageScrub CAGNSite < 250rn •]CAGNSite > 250rn 250 rn

33 40 500 rn

•250

117 42 117 40

FIGURE 2. Distribution of coastal sage scrub in southern Orange County relative to elevation. Bold lines indicate 250-m and 500-m elevation contours.Vegetation map based on United StatesForest Service (1940), with coastalsage scrub defined as areasdominated by Artemisiacalifornica, Salvia mellifera, Salvia apiana or Eriogonumfasciculatum. Known sitesof California Gnatcatcher(CAGN) occurrencemapped as 1-min blocksof latitude- longitude. known gnatcatcherpopulations (Atwood 1990). Housing constructionin southernCalifornia is generallyfocused in lowland areas,whereas sites designatedfor mitigation are often situatedin areas of higher elevation that are more costlyto develop,and lesssuitable for gnatcatchers.Con- sequently,areas of low elevationcoastal sage scrub that are likely to supportCalifornia Gnatcatchersare oftendestroyed during development projects,while areas of coastalsage scrub located at higher elevations, which are occupiedby few, if any, gnatcatchers,are preserved(Atwood 1990). Vol.63, No. 2 CaliforniaGnatcatcher Distribution [ 165

Los Angeles -33

;UNITED.....

San Diego -32

Ensenada

-31

Polioptila c. californica

0 20 40 60 80 100 1•0 • ' '• Distribution'Kilon•eter• ' Elevation< •50 m "•:•rlcvation250-500m Sanquintin

--116

FIGURE3. Maximum historic distribution of Polioptilacalifornica californica, based on the 250-m and 500-m elevationcontours and range descriptionsprovided by Grinnell and Miller (1944), American Ornithologists'Union (1957) and Atwood (1991). See text for further discussion.

The broad elevationalpatterns of California Gnatcatcherdistribution alsoprovide a methodof addressingan issueof relevanceto the proposed endangeredspecies listing of the northern, nominate subspecies.The United StatesEndangered Species Act is seldomapplied to speciesor subspeciesthat occuronly marginallyin the United Statesbut which have 166] J. L. Atwoodand J. S.Bolsinger J.Field Ornithol. Spring 1992 widespread,stable distributions in other countries.Should P. c. californica be considereda Mexican subspeciesthat only marginally extendsinto the United States, or vice versa? California Gnatcatchersin northwesternBaja California are, like those of southernCalifornia, obligatelyassociated with coastalsage scrub (At- wood 1980, 1988); the southern distributional limit of P. c. californica (American Ornithologists'Union 1957, Atwood 1991) is coincidentwith the southernterminus of this vegetationtype (O'Leary 1990). Although the distributionof coastalsage scrub in Mexico is characteristicallypatchy and fragmented,there are no data to suggestthat the degreeof patchiness of this habitat differedhistorically between the United Statesand Mexico (Kirkpatrick and Hutchinson1980; Atwood,unpubl. data; P. A. Bowler, pers. comm.). In fact, Mooney and Harrison's (1972) descriptionsof coastalsage scrub near SanTelmo, Baja California (3 IøN latitude)suggest that the distribution of this vegetationtype in Mexico is controlledby similar ecologicalfactors as in the United States.Therefore, we usedthe elevational characteristicsof P. c. californica'sdistribution in southern California to estimatethe maximum, historic extent of the subspecies' rangein Baja California, and to comparethis valuewith a similar estimate from the species'range in the United States(Fig. 3). Calculationsof land areasthat have been definedby elevationcontours obviously overestimate the extent of coastalsage scrub. Nonetheless, we believethat suchmea- surementsprovide an index of the relative historic abundanceof this habitat type, and probably of P. c. californica,in the United Statesvs. Mexico. Geographicinformation systemanalysis indicated that approximately 6898 km2 of low elevation(<250 m) land existsin the southernCalifornia rangeof P. c. californica,defined by Grinnell and Miller (1944) as "coastal southern California from the Mexican line northwest to the lower Santa Clara Valley, Ventura County, and eastwardto San Gorgonio Pass." From the Mexican border southto the subspecies'southern limit at 30øN latitude (American Ornithologists'Union 1957, Atwood 1991), approximately 2869 km2 of land <250 m elevationis presentwest of the Sierra San Pedro Marfir. Using a more liberal definition of the California Gnatcatcher'selevational limits, approximately 12,930 km2 of land < 500 m elevationwas calculatedas occurringwithin the rangeof P. c. californica in the United States, as opposedto 6934 km2 within the subspecies' Mexican range. These resultsare consistentwith Kirkpatrick and Hutchinson's(1977: 21) statementthat coastalsage scrub covers "its largestarea betweenSan Diego and Ventura where it extendsinland as far as San GorgonioPass." In northwesternBaja California, low elevationland that is most likely to support California Gnatcatcherpopulations is restrictedto a narrow coastalstrip that is borderedto the eastby the abruptly-risingSierra San PedroMarfir. In conjunctionwith Atwood's(1991) recommendationthat the presentlyaccepted (American Ornithologists'Union 1957) distributional limits of P. c. californicabe maintained, these data indicate that Vol.63, No. 2 CaliforniaGnatcatcher Distribution [167

65-70% of the historic range of the California Gnatcatcher'snominate subspeciesoccurred north of Mexico.

ACKNOWLEDGMENTS

Barbara Hamilton, Dennis Heinemann and Mike Thrift all made valuable contributions to data collectionand analysis.Individuals too numerousto mention provided valuable, unpublishedfield information. Discussionswith Dave Bontrager and Pat Mock provided valuable insights;the manuscriptwas improved by the commentsof Kimball Garrett. Financial supportwas providedby Manomet Bird Observatory,the Natural Resources DefenseCouncil, the Shirley Wells ConservationFund of the El Dorado AudubonSociety and the EndangeredHabitats League.

LITERATURE CITED

AMERICAN ORNITHOLOGISTS' UNION. 1957. Check-list of North American birds. 5th ed. AmericanOrnithologists' Union, Washington,D.C. 691 pp. ß 1989. Thirty-seventhsupplement to the American Ornithologists'Union Check- list of North American birdsß Auk 106:532-538. ATWOOD,J. L. 1980. The United Statesdistribution of the California Black-tailed Gnat- catcher. Western Birds 11:65-78ß

ß 1988. Speciationand geographicvariation in black-tailedgnatcatchersß Ornithol. Monogr. 42:1-74. ß 1990. Statusreview of the California gnatcatcher(Polioptila californica). Unpub- lishedtechnical report, Manomet Bird Observatory,Manomet, Massachusetts.79 pp. 1991. Subspecieslimits and geographicpatterns of morphologicalvariation in California Gnatcatchers(Polioptila californica). Bull. S. Califß Acad. Sci. 90:118-133. BEST,T. L. 1983. Morphologicvariation in the San Quintin KangarooRat (Dipodomys gravipesHuey 1925). Am. Midi. Natur. 109:409-413. DESIMONE, S. A. 1989. Classificationand environmental relationshipsof coastal sage scrubat Starr Ranch Sanctuary,Orange County, CaliforniaßM.A. thesis,California State Univß, Fullerton, California. GRINNELL,J., ANDA. H. MILLERß 1944. The distributionof the birds of CaliforniaßPac. Coast Avif. 27:1-608. KIRKPATRICK,,J. B., AND C. F. HUTCHINSONß1977. The communitycomposition of California coastalsage scrubß Vegetatio 35:21-33. , AND--. 1980. The environmentalrelationships of Californian coastalsage scruband someof its componentcommunities and species.J. Biogeog.7:23-38ß MICHAELBRANDMAN ASSOCIATESß 1991. Unpublishedjob report:A rangewideassessment of the California Gnatcatcher(Polioptila californica). Prepared for the Building Industry Associationof SouthernCalifornia, Santa Ana, California;July, 1991. MOONEY,H. A. 1977. Southerncoastal scrubß Pp. 471-489, in M. G. Barbour and J. Major, eds.Terrestrial vegetationof CaliforniaßWiley, New York, New York. , ANDA. r. HARRISON.1972. The vegetationalgradient on the lower slopesof the Sierra San Pedro Martir in northwestBaja California. Madrofio 21:439-445. MUNZ, P. A., ANDD. D. KECK. 1959. A California floraßUniv. CaliforniaPress, Berkeley, California. 1681 pp. O'LEARY,J. F. 1990. Californian coastalsage scrub: general characteristics and consid- erationsfor biologicalconservationß Pp. 24-41, in A. A. Schoenherr,ed. Endangered plant communitiesof SouthernCalifornia, SouthernCalifornia BotanistsSpec. Publ. No. 3. Rancho Santa Ana Botanic Garden, Claremont, Californiaß UNITEDSTATES FOREST SERVICE. 1940. Map of the vegetationtypes of California, Corona QuadrangleßCalif. Forest and Range Exp. Station,Cleveland Natl. Forest,California. WESTMAN,W. E. 1981. Factors influencing the distribution of speciesof Californian coastalsage scrub. Ecology62:439-455. ß 1983. Xeric Mediterranean-type shrublandassociations of Alta and Baja Cali- fornia and the community/continuumdebateß Vegetatio 52:3-19. 168] J. L. Atwoodand J. S.Bolsinger J.Field Ornithol. Spring 1992 ß 1987. Implicationsof ecologicaltheory for rare plant conservationin coastalsage scrubßPp. 133-140, in T. S. Elias, ed. Proceedingsof the Conferenceon Conservation and Managementof Rare and EndangeredPlantsß California Native Plant Society, Sacramento, California. Received28 Feb. 1991; accepted13 Aug. 1991.

INTERNATIONAL CRANE SYMPOSIUM

The National Audubon Societyis sponsoringan internationalsymposium entitled "People, Water and Wildlife: Human PopulationImpacts on Cranes" on March 23-24 in Kearney, Nebraska.The symposiumwill highlightcrane and water expertsfrom the U.S. and Asia, including Dr. Jim Lewis, National Whooping Crane Coordinator for the U.S. Fish and Wildlife Service,Ken Strom, Manager of Audubon'sRowe Sanctuary,Dr. Jonathan Cole, researcherfor the Institutefor EcosystemStudies, Bijay Malla from Nepal'sKing Mahendra Trust for Nature Conservation,Dave Fergusonof the U.S.F.W.S. International Office, Gary Lingle of the Platte River Whooping Crane Critical Habitat MaintenanceTrust, Jim Harris of the International Crane Foundation, Ross Lock of Nebraska Game and Fish Department, and crane expertsfrom Russia, India and Pakistan.Audubon staff members will also discussthe issueof demographicpressures on wildlife and identify sustainable solutions.For additionalinformation, contact Jim Connolly, National Audubon Society, 666 PennsylvaniaAvenue, S.E., Washington, D.C., 20003. Telephone: (202) 547-9009. Fax: (202) 547-9022.