HABITAT PREFERENCES OF WINTERING DIURNAL RAPTORS IN THE SACRAMENTO VALLEY

GERALD S. WILKINSON, Department of Biology, C-016, Universityof Californiaat San Diego, La Jolla, California 92093 KIM R. DEBBAN, 3407 Wood Acres Drive, Boise, Idaho 83705

As slowly reproducingtop carnivores, most raptors can be con- sideredpotentially endangered (Miller and Botkin 1974). Therefore, accurate data on population trends and their causes, such as pesticideaccumulation or habitat alteration, are urgently needed. Olendorff and Kochert (unpublishedmanuscript) recently stressed the importance of preservingbreeding habitat for successfulraptor management and conservation. However, considerablyless em- phasis has been placed on wintering habitat, even though most North American raptorsare migratory(Bent 1937, 1938). Relatively little is known about wintering habitat preferencesfor any raptor species(but see Weller 1964, Schnell 1968 and Koplin 1973); con- sequently, it is difficult to predict the effects of wintering habitat alterationson raptor populations. Fifteen diurnal raptors--White-tailed Kite ( leucuru$), Marsh (Circuscyaneu$), American (Falco$parveriu$), (F. colurnbariu$),Prairie (F. mexicanus), (F. peregrinu$), Red-tailed Hawk (Buteo jamaicen$i$), Rough-leggedHawk (B. lagopus),Ferruginous Hawk (B. regalis) Red-shoulderedHawk (B. lineatu$),Sharp-shinned Hawk ( $triatu$), Cooper's Hawk (A. cooperii), Golden Eagle (Aquila chrysaeto$), Bald Eagle (Haliaeetu$ leucocephalu$)and Turkey (C'atharte$aura)--winter regularlyin or migratethrough the SacramentoValley of California.High winter concentrationsof some of these raptorsoccur locallyin the valley. To discoverif wintering raptorsdo have habitatpreferences we monitoredpopulation fluc- tuationsof all diurnalraptors in two habitatsin a portionof the central SacramentoValley by conducting18 censusesover 190 days from 26 October 1976 to 2 May 1977.

STUDY AREA AND METHODS

The censusarea comprises43 km2 and its southernboundary is located 4.8 km north of Davis, Yolo County, California. Rotating cultivationof such cropsas sugarbeets, corn, tomatoesand wheat characterizesthe centralportion of the valley. At any time duringthe winter months 25 to 50% of the fieldsare fallow. The valley floor is flat and treelessexcept along windbreaksand watercoursesor near farmhouses.

Western 11: 25-34, 1980 25 WINTERING DIURNAL RAPTORS A privatehunting club maintains 7.7 km2 of thestudy area by floodingintermittently and planting cultivated oats. A localtomato canneryuses 1.3 km 2 as a sprayfield for their waste water. We con- siderthe sprayfield tobe a separatehabitat from the agricultural fields becausethe compositionof itsvegetation is moreheterogeneous. Watergrass(Echinochloa crusga!!i) ismost common although Prickly Lettuce(Lactuca serfJoia), Lambs-quarters (Chenopodium album), Alkali Mallow ($ida !eprosa),Common Spikeweed(Hemizonia pungens),Curly Dock (/•urnex crisp us) and Yellowstar Thistle (Cen- taureasolstitialis) occur frequently. The cannery fenced and posted thisarea as a raptorpreserve in 1975and occasionally mows it for hay.

I KM

R D 28H

Figure1. Routedriven on each census north of Davis, Yolo County, California. The studyarea extends 0.8 km on each side of the route; arrows signify direction oftravel. Thecircle denotes the starting point and the square (intersection roads 28H and 103) denotes the ending point.

26 WINTERING DIURNAL RAPTORS

We drove a 37.5 km route (Figure 1) during each censusand recorded all raptors seen within 0.8 km of the road. All raptors sightedcould be identifiedto speciesat thisdistance using binoculars. These observationallimits defined the perimeterof the study area. An average speed of 50 km/hr was maintained for spotting; however, all sightingswere verified with our vehicle at a standstill. The location (as well as sex and age when possible)of each raptor sightedwas recordedby markingunique symbolsfor each species' age/sex classat appropriate coordinateson gridded maps of the study area. Duplicationof sightingwas uncommonbecause most birds sighted were perchingon telephonepoles or sittingon the ground, the cen- sus area was large with little overlap, and most censuseslasted ap- proximately 2 hours (mean censustime _+ 2 SD was 128 _+ 45 minutes). However, any birdswhich were suspectedto be recounts due to their direction of movement were not counted. Since there are few trees within the study area, most raptorspresent were prob- ably observedon each census. We recorded time of day, minimum and maximum wind velocities and ambienttemperature before and after each census.Three of the 18 censuseswere conductedin the morning, 2 during mid-day and 13 during late afternoon.

RESULTS

Each of the four commonresident raptor species showed marked populationchanges over the study period (Figure2). If the popula- tion increaseswere due to the influx of winteringpopulations, then the winter censustallies should nonrandomly increase over the resi- dent populationlevel. We used the mean of all censustallies as an estimateof the residentpopulation level for each species.Then, we determinedif the fluctuationswere random or nonrandomby com- paring individualcensus tallies for each specieswithin and outside the raptor preserveto each species'estimated local population level. Outsidethe preserve,Marsh Hawk and White-tailedKite populations showed nonsignificantchanges, whereas Red-tailed Hawk and AmericanKestrel populations showed marked changes (p < 0.001 in each case, chi-squared test for goodnessof fit, Sokal and Rohlf 1969). Within the raptor preserveall four speciesshowed significant population changes (p<0.001 for all cases, chi-squaredtest for goodness of fit). These results indicate that these four resident specieshave largerwintering than residentpopulations and suggest that Marsh Hawk and White-tailedKite winteringpopulations con- centrate in the raptor preserve.

27 2.0 MARSHHAWK

1.5

•.o

0.5

J F M A CENSUS DATE

2.0 WHITE-TAILED KITE

1.5

'•.o

0.5

J F M A CENSUS DATE 2.0 AMERICAN KESTREL

1.5

z 0.5

I I i I i I I N D J F M A CENSUS DATE 2.1 RED-TAILED HAWK

1

z 0.5

N D J F M A CENSUS DATE

Figure2. Densityfluctuations ofthe four most common resident diurnal raptors inside the raptorpreserve (dark trace) and outsidethe raptorpreserve (light trace). Hash marks above abscissaindicate census dates. WINTERING DIURNAL RAPTORS

To determine more preciselythe habitat preferencesof the five winter residentraptors--Red-tailed , Rough-leggedHawks, Marsh Hawks, White-tailed Kites and American --the den- sities (birds/kin2) of each specieswithin the raptor preserve were compared to densitiesoutside the preserve (cf. Figure 2) with the Wilcoxon test for paired comparisons(Siegel 1956; all probabilities are computed using a one-tailed test). Rough-legged Hawks (p<0.05), Marsh Hawks (p<0.001) and White-tailed Kites (p<0.001) preferentially used the raptor preserve whereas Red- tailedHawks showed no statisticalpreference (p = 0.158) for the rap- tor preserve over the rest of the study area. American Kestrels showed a statisticalavoidance of the preserve(p•. 0.001). To estimatethe effectsof weather on our sightingprobabilities for each species,we performed Spearman rank correlations(Siegel 1956) betweencensus totals and mean temperatureand wind veloci- ty (cf. Figure 3). Marsh Hawk and White-tailed Kite observations were not significantlycorrelated with wind velocities,whereas Red- tailed Hawk and AmericanKestrel observations showed significant negative correlations (rs-0.57, p•.0.05; rs=0.59, p<0.02, respectively).American Kestrel observationswere not significantly correlatedwith temperaturewhereas Marsh Hawk, Red-tailed Hawk and White-tailedKite observationsshowed significantpositive cor- relations (rs=0.60, p•.0.02; rs-0.77, p<0.001; rs--0.66, p <0.01, respectively).

3O

N D J F M A CENSUS DATE

Figure3. Meanwind and temperature readings for each census, October 1976 to May 1977.

3O WINTERING DIURNAL RAPTORS

SPECIES ACCOUNTS

Nestingrecords of each speciesas well as age/sex information,if known, and dates of observations of uncommon species are presentedhere. In addition, populationstatus in the study area and the immediatevicinity, based in part on Grinnelland Miller (1944), is included. Status symbolsrepresent transients (T), winter residents (WR), and breeding residents(B).

WHITE-TAILED KITE (WR,B): This specieswas commonly observedin sugarbeet fields south of Road 28H after 1 March. No nest was discoveredin the study area althoughkites were observedthere throughoutthe breedingperiod. MARSH HAWK (WR,B): Until 11 March we observeda mean of 5.6_+ 2.9 (_+2 SD) adult males. This relativelylow variancesuggests that most of the hawkswere not transientsbut winter residents. Unfortunately, first year birds (ventrally bully to cin- namon without streaking)were not distinguishedfrom adult females(ventrally bully withstreaking) until 22 January.However, mostof the wintervisitors were either adult femalesor immatures.A communalroost was used duringJanuary withinthe study area. At leastfour pairsattempted to nestwithin the studyarea, all in sugarbeet fields. Three of these pairs contained immature birds. All four nestsfailed when the beet fields were plowed. AMERICAN KESTREL (WR,B): Of the 345 kestrelsobserved through 19 March, 22% were males, 61% were females and 17% were unknown. This relative abun- dance of femalesduring winter agreeswith the findingsof Koplin (1973) and Mills (1975, 1976). After 19 March we observed a total of 19 kestrels:67% males, 22% females and 11% unknown. We believe this reversal in the relative abundances of each sex duringearly springis due to the involvementof the femalesin incubation.A minimum of two kestrelpairs nestedin the study area. MERLIN (T,WR?): A solitaryMerlin wasseen six timesbetween 12 Januaryand 14 April at one of two locationsalong road 103 betweenroads 27 and 25 (cf. Figure 1). On each occasionthe falcon resembledan adult male Taiga Merlin (F. c. colurnbarius, Temple 1972). PRAIRIEFALCON (T): The threesightings occurred on 11 November,4 Januaryand 22 Januaryat differentlocations in the studyarea. Age and sex were not determined on any sighting. RED-TAILED HAWK (WR,B): Of the 366 sightings,35% were adults,30% were im- matureplumaged and 35% were not aged. No nestswere found in the studyarea. A melanisticbird withjesses was occasionally observed over a periodof severalmonths. ROUGH-LEGGED HAWK (WR): A total of 41 was observed over 15 censusesbe- tween 16 Novemberand 6 April. A mean of 2.7 _+3.6 (_+ 2 SD) hawkswere observed on each of thesecensuses. Twelve Rough-legs,10 of which were flyingin a group, were observedwithin the study area on 26 February, a noncensusday. One, which wasrecognizable by plumagepattern and wasthe lastbird to depart, wasobserved on the same telephonepole regularlyover severalmonths. FERRUGINOUS HAWK (T): All six were observedduring five censusesbetween 30 Novemberand 23 February.We have no evidencethat any birdremained in the study area for longer than one week. RED-SHOULDERED HAWK (T,B): Solitary birds were observed on 22 and 28 January, and two were observedon 4 February. All sightingswere of immature plumagedbirds. SWAINSON'S'HAWK(Buteo swainsoni; T,B): Two were observed on 6 April,10 on 14 April and one on 2 May. Fourteenwere reported near the studyarea in mid-April. 31 WINTERING DIURNAL RAPTORS

Three were observedin the studyarea on 28 May and alsoon 22 July. At leastone pair attemptedto nestwithin the studyarea. SHARP-SHINNED HAWK (T): An adult was observedin the orchardsouth of Willow Slough on 22 January. GOLDEN EAGLE (T): A first-yearbird (whitein wingsand tail) was observedflying southeaston 19 March just north of the raptorpreserve. A Rough-leggedHawk stoopedat the eaglefor severalminutes. TURKEY VULTURE (T): Two vultureswere observedon 11 March, one on 25 March and two on 14 April. All sightingswere of soaringbirds. SHORT-EARED OWL (Asioflarnrneus; WR,B): Althoughnot enteredin our census tallies,these owls were seen occasionally at dusk.All sightingswere either in the raptor preserveor in the sugarbeet fields south of road28H. We suspectthat the owlscom- munallyroosted on the raptorpreserve on someoccasions. A highof 20 wasseen overthe raptorpreserve on 23 January,a noncensusday. At leastone pair nestedin the sugarbeet fields south of road 28H becausea fledgling(presumably dead from starvation)was found there within a week after the fieldswere plowed. BURROWING OWL (Athene cunicularia;WR,B): This owl also was not censused regularly,but at leasttwo pairs nested within the study area. One nest had seven and the other had five fledglings.

DISCUSSION AND CONCLUSIONS Most winteringraptors in the centralSacramento Valley arrive be- tween October and November (Bent 1937, 1938). These birds leave between March and April although each speciesmay show some variation. Since our data extend from late October until early May, we believe that we have accurate tallies of all winter residents. However, we may have misseda few early fall migrants. Several of the correlations we discovered between species numbersand either temperatureor wind velocityindicate that these two environmentalfactors influence raptor movementsdirectly or in- directlyvia the behavior of prey species.The correlationsbetween Marsh Hawk, Red-tailed Hawk and White-tailed Kite census totals and ambient temperature suggestthat these species either move locally, migrate or become less detectableduring periods of cold weather. The absence of a correlation between American Kestrel numbers and temperature suggeststhat American Kestrels do not move in responseto temperature. This result is expected because thesefalcons do not rely on thermalsto forage. Correlationsbetween Red-tailed Hawk and American Kestrel census totals and wind velocitysuggest that thesetwo speciesalso respond to highwinds by moving locally and/or become less detectableby perchingin pro- tected locations.This interpretationagrees with our casualobserva- tions. When the wind was strong (approximately20 km/hr or stronger),Red-tailed Hawks and AmericanKestrels almost always were observedperching, whereas Marsh Hawks and White-tailed Kites often were foraging. Since Marsh Hawks, Rough-leggedHawks and White-tailedKites preferentiallyfrequented the raptorpreserve during the day (and 32 WINTERING DIURNAL RAPTORS probablyShort-eared Owls did so at dawn and dusk), one or more foodssources may be commonlyused by theseraptors. All of these speciesare known to prey on the CaliforniaVole (Microtuscalifor- nicus;Craighead and Craighead1956, Brownand Amadon 1968, Warner and Rudd 1975). Krebs (1966) statesthat CaliforniaVoles tendto thriveunder irrigated agricultural conditions. Since the raptor preserveis frequentlysprayed with wastewater, grassesand herbs are usuallygrowing and provideabundant rodent food. Water Pipits (Anthusspinoletta), Brewer's Blackbirds (Euhpagus cyanocephalus),Killdeer (Charadrius uociferus) and Savannah Sparrows(Passerculus sandwichensis) occur frequently and abun- dantly in the preserveand may provide alternatefood sources. Marsh Hawks and Short-earedOwls commonlytake avian prey of thissize (Craighead and Craighead1956, Pageand Whitacre1975). The unusuallydry winterof 1976-77 may havecaused a reduction in birdand rodentpopulations in agriculturalfields, thereby intensify- ing raptorcongregation at the preserve.In a normalrainfall year, the preservemight be usedless frequently. However, on 22 December 1977 more than 15 White-tailedKites, 6 Rough-leggedHawks and many Red-tailedHawks and Marsh Hawks were observedon the preserveeven though 1977-78 was an above normal rainfall year. These figuresare comparableto our censustallies of the previous year, hence the droughtmay not have had a significanteffect. Of course, such an interpretationmust be consideredwith caution becauseeffects of a droughton raptor populationsmay take several yearsto disappear. Three of the central SacramentoValley winter resident diurnal raptors (Rough-leggedHawk, Marsh Hawk and White-tailedKite) preferentiallyfrequent the raptor preserve. Recognition of this preference is important becausealteration, such as plowing, of preferred winter habitats can cause local movements of raptor populationsand may createcritical energetic demands if prey is not abundantin nearbyareas. We believethat a raptorpreserve, such as the sprayfieldin our studyarea, is an effectivemanagement area for those specieswhich frequent it, insuringsuitable winter habitatfor those speciesin addition to providingan excellent opportunityto view relativelyuncommon birds in high concentrations.

ACKNOWLEDGMENTS We would like to thank Cameron Barrowsfor assistingwith plant identification,John Rice for suggestingsome of the statisticalpro- cedures,both David Winkler and Tim Manolisfor providingmany usefulcomments on the manuscriptand Pete Bloom for excitingus with raptor biology.

33 ß WINTERING DIURNAL RAPTORS

LITERATURE CITED

Bent, A.C. 1937, 1938. Life historiesof North Americanbirds of prey. U.S. Natl. Mus. Bull. 167, 170. Brown, L. and D. Areadon. 1968. Eagles, hawks and falconsof the world. Vol. 2. McGraw-Hill, , NY. Craighead,J.J. and F.C. Craighead,Jr. 1956. Hawks, owlsand wildlife.Stackpole Co., Harrisburg,PA. Grinnell, J. and A.H. Miller. 1944. The distributionof the birds of California. Pac. Coast Avif. 27. Koplin, J.R. 1973. Differentialhabitat use by sexesof AmericanKestrels wintering in northern California. Raptor Res. 7:39-42. Krebs,C.J. 1966. Demographicchanges in fluctuatingpopulations of Microtu$califor- nicu$. Ecol. Monogr. 36:239-273. Miller, R.S. and D.B. Botkin. 1974. Endangeredspecies models and predictions. Am. Sci. 62:171-180. Mills, G.S. 1975. A winterpopulation study of the AmericanKestrel in centralOhio. Wilson Bull. 87:241-247. Mills, G.S. 1976. American Kestrei sex ratios and habitat separation. Auk 93:740-748. Page, G. and D.F. Whitacre. 1975. Raptor predationon winteringshorebirds. Con- dor 77:73-83. Schneil, G.D. 1968. Differentialhabitat utilizationby winteringRough-legged and Red-tailed hawks. Condor 70:373-377. Siegel, S. 1956. Nonparametricstatistics for the behavioralsciences. McGraw-Hill, New York. Sokal,R.R. and F.J. Rohif.1969. Bi6metry.W.H. Freeman,San Francisco. Temple,S.A. 1972. Sex and age characteristicsof the North AmericanMerlins. Banding 43:191-196. Warner, J.S. and R.L. Rudd. 1975. Hunting by the White-tailed Kite (Elanus leucuru$). Condor 77:226-230. Weller, M.W. 1964. Habitat utilizationof two speciesof buteoswintering in central Iowa. Iowa Bird Life 34:58-62. Accepted 9 November I979

Sketch by Cameron Barrows 34