Proceedings--Fire Effects on Rare and Endangered Species and Habitats Conference, Nov. 13-16, 1995 Coeur d'Alene, Idaho © IAWF, 1997, Printed in U.S.A

Effects of Fire on the Ecology of the California , california, in California Sage Scrub Communities

William O.Wirtz II1, Audrey L. Mayer2, Mary M. Raney1, Jan L. Beyers3 1Department of Biology,PomonaCollege,Department of Biology,609 N.College Ave.,Claremont,CA91711-6339 Tel.(909)621-8606;Fax (909)621-8878; E-mail [email protected]

2Department of Ecology and EvolutionaryBiology, University of Tennessee,Knoxville, TN 37996-1610 Tel. (423)974-6186: Fax (423)974-0978

3USDAForest Service,Pacific Southwest Research Station Tel. (909)680-1527;Fax (909)680-1501; E-mail [email protected]

Abstract. The occurs in the disap entific history with the Black-tailed gnatcatcher, P. pearing sage scrub communities of southern California and melanura, of the southwestern deserts, the California gnat- Baja California, Mexico. Fire is an integral part of this catcher is now recognized as avalid species (Atwood 1988, community, but its effects on the gnatcatcher are unknown. American Omithologists’Union 1989). Thespecies is pres- By determining density and productivity of ently distributed from Ventura County, California south on burned and unburned coastal and inland sites, we can to the tip of Baja California, Mexico, with the subspecies evaluate the relationshipbetween utilizationof coastal P.c. californica occurring along the coast from Ventura sagescrub and fire history in southern California.Coastal and Los Angeles counties south to approximately 30 N and inland sites which burned in 1990 have been monitored latitude in Mexico (Atwood 1991). It is rare to nonexist- since 1992;thereareunburnedcontrol areas near eachsite. ent in the northern part of its range due to extensive habi- About20pairsof gnatcatchers use one coastal control, and tat loss in recent years (Atwood 1991) and current popu- 14 pairs another, or about 2 pairs per ha. Burned coastal lation studies estimate that there are less than 2,000 pairs sites supported 4 pairs and2pairs respectively in 1994and left in California (Sahta 1991). Several studies have indi- 1995, or about 0.2 pair per ha. The mean percent of cated that the California gnatcatcher has an extremely low successfully reproducing pairs on burned coastal sites over reproductiverate, due to high nest failure and significant four years was 42.9, the meanon unburned sites was 50.4. nest and hatchling predation (Atwood1990). Nest para- On three unburned coastal sites, successful pairs produced sitism by the brown-headedcowbird (Molothrusater)may an average of1.83 young per temtory overfour years. On also decrease reproductive success (Atwood 1990). How- two burned coastal sites, the young produced increased ever, the leading cause of gnatcatcher decline may be a from 1.3 in 1992 to 2.25 in 1994, but dropped to 1.0 in 1995. direct result of the significant loss of suitable habitat in Approximately 15 pairs have territories near a burned the last 70 years (Atwood 1980, 1992; Rea and Weaver inland site,and immatures and adults forage from three of 1990)., and the California gnatcatcherwas Federally listed these territories into the bum, but no territories are re- as a threatened species in March 1993. stricted to the burned area. These 15 pairs have a mean In addition to the fragmentation suffered at human success rate of 37.7 percent and an average of 1.35 young hands in the last 50 years, the sage scrub community in per territory for four years. We believe that the extent of southern California has also been subjected to frequent fires vegetative recovery determines habitat suitability for breed- of both natural and anthropogenic origin. While fire is an ing pairs, and several parameters are being examined. integral part of this community, its effects on vertebrate Gnatcatcher pairs that are able to establish territories on populations have not been well-documented. We have burned areas breed at rates similar to those breeding on been studying the effects of fire on the avian community unburned sites. in sage scrub habitats since 1992. This paper considers only effects of fire on California gnatcatchers. By deter- mining the density andreproductivesuccessof gnatcatch- Introduction ers on burned sitesof known age postfiire, andon unburned controls, we can examine the relationship between gnat- “One of the most restricted in distribution of all the catcher utilization of sagescrubcommunities and fire his- in the United States is the Black-tailed Gnatcatcher tory in southern California. (Polioplila californica), which is found in certain arid, Relatively little is known about the effects of fire on brushy sections of thePacific slopeof southern and Lower the sage scrub plant community,especially within the range California” (Woods 1921). Lumped for much of its sci- of the California gnatcatcher. All sage scrub shrubs can 92 Wortz II, W.O., Mayer, A. L., Raney, M. M., and Beyers, J. L. resprout to some degree. Some genera, like Malosma. tified by monitoring territories to determine clutch size Rhus,and Salvia, exhibittheability to resprout vigorously (if nests could be found), fledging success, and fledgling after fire, some are less vigorous, like Artemisia and survival. We used inconspicuous markers to indicate Eriogonum, while others reproduce primarily from seed. the limits of the resident pairs activity. Then, after the The firehistoryof an area is likely to affect succession after breeding season, we carefully mapped and measured the fire,duetodifferences in fuelloads which in turnaffectthe area delineated by these markers. intensity of the fire (Malansonand Westman 1991). In- Reproductive success was defined for each pair, for tense fires may be more destructive, and leave less standing each season, as no attempt (NA), unsuccessful attempt vegetation to be a source of shelter or new seed, or leave (UA), or successful attempt (SA), and the number of fledg- fewer refugia for after the fire. lings produced per pair each season was also determined. No attempt is defined as no breeding behavior observed; Methods unsuccessful attempt is defined as breeding behavior ob- served but no surviving fledglings;and successful attempt is There are ecological differences between coastal and defined as fledglings produced which survived to disperse. inland gnatcatcher populations,and between different sage Vegetation at these sites was quantifiedvia line point scrubsubassociations(Kirkpatrickand Hutchinson 1980). samples in a subsample ofgnatcatcher territories (Beyers Theplant community along the southern Californiacoast and Pena 1995;Beyers and Wirtz this volume). is San Diegan, while that inland in western Riverside County is Riversidean (Kirkpatrick and Hutchinson Results and Discussion (1977). Since 1992, we have been studying a coastal population atCamp Pendleton, SanDiego County, where At Camp Pendleton, approximately 20 pairs of gnat- gnatcatcher surveys conducted in 1989 by the U.S. Fish catchers have utilized the unburned Basilone site, and and Wildlife Service located 175-200pairs of gnatcatch- approximately 14 pairs have utilized the unburned por- ers. In November 1990, two fires burned over 2,600 ha tion of the Uniform site, over four successive breeding of coastal sage scrub at Camp Pendleton. Our two study seasons. These 34 pairs are distributedat about 2 pairs areas, Basilone-San Onofre and Uniform, contain un- per ha in unburned habitat. Four pairs of gnatcatchers burned controls adjacent to habitat that burned in 1990. have utilized the burned San Onofre site and two pairs have And we have also been studying an inland population at established territoriesonthe burned Uniform site during this Lake Mathews in Riverside County, where fire burned period. Burned sites supportedabout0.2 pair per ha. portions of the habitat in 1990. The mean percent of successfully reproducing pairs All siteswere surveyedregularly forgnatcatchersand on burned areas over four years was 42.9, the mean on associated bird species. Reproductive success was quan- unburned areas was 50.4 (Figures 1 and 2). These fig-

Figure 1. Gnatcatcher reproductive success at Camp Pendleton (Agra= burned, Horno and Don= unburned) - Effects Fire on the Ecology of the California Gnatcatcher - 93

Figure 2. Gnatcatcher reproductive success at Camp Pendleton (San Onofre = burned, Basilone = unburned)

ures are greater than the 42% reported by Bontrager (1990) (Brassica nigra), and resprouts of native laurel sumac for pairs in Orange County. On tthree unburned areas (Malosma laurina), California sagebrush (Artemisia successful pairs produced an average of 1.83 young per californica),coyotebush (Buccharispilularis), deer vetch territory over four years (Figure 3). On two burned ar- (Lotusscoparius) andelderberry (Sambucusmexicanus). eas, the young produced increased from 1.3 in 1992 to Unburned areas at Pendleton are dominated by a mixture 2.25 in 1994, but dropped to 1.0 in 1995(Figure 3). of California sagebrush,coyotebush, laurel sumac, cacti Theplant community on theburned Pendleton sitesis (Upuntia sp.), deer vetch, elderberry, and seasonal her- dominated by mixed introduced grasses, black mustard baceous plants.

Figure 3. Gnatcatcher fledglings per pair at Camp Pendleton (Agra and San Onore= burned; Horno, Don, and Basilone= unburned) 94 Wirtz II, W.0,. Mayer, A. L., Raney, M. M., and Beyers. J. L.

Figure 4. Gnatcatcher reproductive success at Lake Mathews

At Lake Mathews, 15 pairs utilize the unburned area The plant community on more mesic sitesat Mathews and, while birds forage in burned riparian, no territories is dominated by California sagebrush and desert are located solely on the 5 year old burned area. These brittlebush (Encelia farinosa); this is the area typically 15 pairs have a mean success rate of 37.7 percent(Figure occupied by gnatcatchers. More xeric stands at Lake and an average of 1.35 young per territory (Figure 5), Mathews are nearly pure brittlebush; these areas arenever for four years. The mean success rate of bitrs at this site includedin gnatcatcher territories,and seldom utilized for is less than the 42% reported by Bontrager (1990). foraging. Plant succession on the burned area at Mathews

Figure 5. Gnatcatcher fledglings per pair at Lake Mathews - Effectsof Fire on the Ecology of the California Gnatcatcher - 95 has been limited to introduced grasses, with sparse resproutingin Riversideansage scrub because of the more resprouts of sagebrush and brittlebush; burned areas have intense fires which usually occur in this habitat (O’Leary not been included in gnatcatcher territories, and are sel- 1989). Increased pollution and grazing pressure may dom utilized for foraging. Vegetative recovery in the hinder resprouting and regrowth of Riversidean sage scrub burned riparian at Mathews has been rapid. Gnatcatchers and facilitate invasion by non-native herbs (O'Leary and frequently forage in recovering riparian at Mathews, and Westman 1988). Gnatcatchersrarely forageinto this area, also at San Onofre at Camp Pendleton. but occupied temtories are present within 0.25 km on While gnatcatcherdensity may initially be reduced by nearby slopes which are covered with Artemisia and fire, individuals forage in burned areas, and some pairs Encelia. Beyers and Pena (1995) report a positive rela- establish territories in burned areas, within three years tionship between shrub cover and gnatcatcher utilization postfiie. The presence of gnatcatchers is probably depen- on our study areas at Mathews. They (1995) also note dent on suitable vegetation, whether left behind by the that, on unburned areas, gnatcatcher territories include fire as refugia or as seedlings and resprouts respond to much more shrub cover and taller shrubs, than adjacent the fire. At Camp Pendleton, shrub regrowth has been areas not utilized. rapid on the burned San Onofre site, which currently sup- ports four breeding pairs. The San Onofre site is also ad- jacent to burned riparian,which has resprouted especially Summary and Conclusions vigorouslysince 1992. The burned area atUniform (Agra) was dominated by mustard in 1992, but by 1994 both We have observed 34 pairs of gnatcatchers in unburned crown sprouts and seedling shrubs were over one meter San Diegan sage scrub and 6 pairs on burned San Diegan tall. Two breeding pairs of gnatcatchers were found in sitesover five breeding seasons(Table 1). Unburned habi- this area in both 1994 and 1995. tat supports approximately 2 pairs per ha, while burned The gnatcatchers found on burned areas at Pendleton sites support 0.2 pair per ha (Table 1). Mean reproduc- presumably represent dispersal from unburned areas, tive success on unburned sites was 50.4 percent, that on which are as close as one kilometeraway. Oneof us (wow) burned sites was 42.9 percent (Table 1). The mean num- has observed immaturegnatcatchers flying across burned ber fledged per territory, over five seasons, was 1.83 on sites (both Agra and San Onofre)in July and August. The unburned sites and 1.87 on burned sites (Table 1). differences in reproductive success and fledglings per pair We haveobserved 15 pairs of gnatcatchers in unburned between the two burned areas is probably a function of Riversidean sage scrub over five breeding seasons, and differences in postfire plant succession. At San Onofre, found no territories inburned habitat (Table 1). Unburned riparian habitat supported much of the foraging activity habitat supported about 0.10 pair per ha, their mean re- of a pair that established itself between 1993 and 1994, productive success was 37.7 percent, and the mean num- but by 1995 this pair had moved largely into sage scrub ber of fledglings per territory was 1.35 (Table 1). regrowth. Unburned refugia from the 1990 fire initially Although fire in San Diegan sage scrub initially de- provided foraging areas for the other three pairs; today creases the number of gnatcatcher pairs and the repro- (1995) regrowth in these areas reaches 2m. The number ductive success of surviving pairs, there may be a rapid of fledglings produced per pair on the burned area is postfire increase in both gnatcatcherdensity and produc- equivalent to those produced on the unburned area at tivity in only a few years IF there is adequatevegetation Basilone. Although the burned area supports fewerpairs, to support foraging and breeding activities. Either refu- there is little effect on the productivity of resident pairs gia left by fire,or riparian habitat which recoversquickly after five years postfire. At Agra, where destruction of from fire, may increase the likelihood of gnatcatcher oc- vegetation left no refugia after the 1990 fire, a few un- cupation and breeding on burned areas in San Diegan sage burned bushes, plus rapidly resprouting species such as Malosma laurina and Malacothamnos now (1995) seem Table 1. Density and reproductive success of gnatcatchers at to provide critical foraging and nesting sites for the two coastal and inland sites. pairs that have established themselves. Unburned Burned A census by Tutton (1991) at Camp Pendleton esti- Coastal mated that 80% of the gnatcatcherpairs on the base were Total pairs 34 6 located on areas that were at least 16 years postfire. Our Pairsiha 2.0 0.2 Mean % repro. success 50. 4 42.9 data demonstrate an increased use of burned sites since Mean no. fledged/territory 1.83 1.87 1992. Tutton suggests that the effects of human impact Inland on gnatcatcher habitat cannot easily be separated from the Total pairs 15 0 effectsof fire,and it is possible that areas with higher fire Painiha 0.1 0 Mean % repro. success 37.7 0 frequency experience other forms of disturbance as well. Mean no.fledged/territory 1.35 0 At Lake Mathews, the burned area supports low grass and sparse shrub cover. There is generally little postfire 96 Wirtz II, W. O., Mayer, A. L., Raney, M. M., and Beyers, J. L.

scrub. Riversidean sage scrub apparently recovers much Malanson G. P. and W.E.Westman. 1991.Climatic change and more slowly from fire and is not suitable for gnatcatcher the modeling of fireEffects in coastalsagescruband chapar-

colonizationfor at least 5, andperhaps asmuch as10 ral.IN:Fire and theenvironment:ecologicalandcultural perspectives:Proceedingsof an internationalsymposium; S. years. C. Nodvin and T. A. Waldrop, eds. Knoxville, TN. Gen. Tech. Rep. SE-69. USDA Forest Service SoutheasternForest Acknowledgments.Thisstudy wasmadepossibleby Experiment Station. Asheville, NC. 429 pages, funding from the Pacific Southwest Research Station of the O'Leary, J. F. 1989. Californian coastal sage scrub; General USDA Forest Service, and the California Department of characteristics and considerations for biological conserva- Fish and Game Natural CommunitiesConservationPlan- tion. IN: Endangered Plant Communities of Southern Cali- ning program. Permissionwas granted by the Metropolitan fornia.A.Schoenherr,ed.SouthernCaliforniaBotanists Water District to work at the Lake Mathbws Ecological Spec. Publ.3:2441. Reserve,andwewereallowedtoworkatCampPendleton O'Leary, J. F. andW. E. Westman. 1988. Regionaldisturbance by permission from the Environmental and Natural Re- effects on herb succession pattems in coastal sage scrub. Journal of Biogeography 15:775-786. sources Management Office. Rea,A.M..andK.L.Weaver.1990.Thetaxonomy,distribution, andstatus ofcoastalCaliforniacactuswrens.WesternBirds 21:81-126 References Salata,L1991.A Status Reviewofthe California Gnatcatcher. U.S.FishandWildlifeService.pages3-8. American Ornithologists' Union. 1989. Thirty-seventh supple- Tutton,J.1991.A survey of the Califomia gnatcatcher and cactus ment to the American Ornithologists' UnionCheck-list of wrenonCampPendleton,SanDiegoCounty, California. NorthAmericanbirds. Auk106:532-538.1 Unpublishedreport, prepared fortheU.S.MarineCorps, Atwood,J.L.1980.TheUnitedStates disbibutionofthe Environmental and NaturalResourcesManagementOffice, California black-tailed gnatcatcher. WestemBirds 1155-78. Camp Pendleton, Californiaby the U.S.Fishand Wildlife Atwood, J. L. 1988. Speciation and geograpbc variation in Service,SouthernCaliforniaField Station LagunaNiguel black-tailed gnatcatchers. Ornithological Monographs 42. Office. November, 1991. Atwood,J.L.1990.StatusreviewoftheCaliforniagnatcatcher Woods,R.S.1921.Home life of the Black-tailed Gnatcatcher. Polioptila californica. Unpublished technical report, Manomet Condor 23:173-178. Bird Observatory, Manomet, Massachusetts. 79 pages. Atwood, J. L. 1991. Subspecies limits and geographic patterns of morphological variation in California Gnatcatcherss (Polioptila californica). Bulletin of Southern California Academy Sciences 90:118-133. Atwood, J. L. 1992. A maximum estimate of the California Gnatcatcher's population size in the United States Western Birds 23:l-9. Beyers, J. L.and G.C.Pena. 1995. Characteristics of coastal sage scrub in relation to fire history and use by California gnat- catchers. IN: Weise, D. R., and R. E. Martin(tech. coord.), The Biswell Symposium:Fire issues and solutions in urban interface and wildland ecosystems. Gen. Tech. Rep. PSW- GTR-158.Pacific SouthwestResearch Station, USDA Forest Service, Albany, CA. pages 153-154. Beyers, J. L., and W. O. Wirtz, II. (1996).Vegetative character- istics of coastal sage scrub sites used by California gnatcatch- ers: Implications for management in a fire-prone system. Bontrager. D. R. 1990. Unpublished report: Commentson the draft environmentalimpactreport(DEIR)TACAEIR3for the Foothill Transportation Corridor (FTC). Letter to Mr. Steve Letterly, Transportation Corridor Agencies, Costa Mesa, California; October.1990. Kirkpatrick,J.B.andC.F. Hutchingson. 1977. The community composition of Californian coastal sage scrub.. Vegetatio 35:21-33. Kirkpatrick, J. B. and C. F. Hutchinson. 1980. relationships of Californian coastal sage scrub and some of its component communitiesand species. Journal of Biogeogra- phy 7:23-28.