Barred Owl Habitat Use as Determined by ... - Semantic Scholar
Barred Owl Habitat Use as Determined by Radiotelemetry Author(s): Thomas H. Nicholls and Dwain W. Warner Reviewed work(s): Source: The Journal of Wildlife Management, Vol. 36, No. 2 (Apr., 1972), pp. 213-224 Published by: Allen Press Stable URL: http://www.jstor.org/stable/3799054 . Accessed: 03/02/2012 16:55 Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].
Allen Press is collaborating with JSTOR to digitize, preserve and extend access to The Journal of Wildlife Management.
http://www.jstor.org
BY RADIOTELEMETRY BARREDOWL HABITATUSE AS DETERMINED THOMAS H. NICHOLLS,Department of Ecology and Behavioral Biology, University of Minnesota, Minneapolis DWAIN W. WARNER,Bell Museum of Natural History, Universily ofMinnesota, Minneapolis
Abstract: Radiotransmitters were successfullyplacedon 10 barredowls (Strixvaria) during1965-66, systemfor a totalof 1,182 and the movementof the owlswerefollowedwith an automaticradio-tracking days. Morethan 28,000 owl locationswere sampledfrom some two millionlocationsrecorded.These were programmed for computeranalysisto deterlocations,along with habitatavailabilityinformation, mineintensityof habitatuse. Barredowls showedconsistentandhighlysignificantpreferenceor avoidancefor differenthabitatsdespitechangingseasons,phenology,weatherconditions,and years. The order of habitatpreferencein decreasingintensityof use uZasoak ( Quercusspp.) woods,mixedhardwoodsand alder (Alnus spp.) swamps,marshes, conifers,whitecedar (Thufaoccidentalis)swamps,oak-savannas, and openfields.
Thispaperreportsthe use of radiotelem- Traditionalmarking methods seldom etry to determinethe intensityof use of producethe continuous,minute-to-minute varioushabitattypesby barredowls. The datanecessaryforevaluatingshort-duration recentlyacceleratedinterestin predatory movementsandactivitiesof wild birdsthat animalsand the difficultiesinvolvedin the will accuratelyprovideinformationon instudy of generally nocturnaland little tensityof habitatuse. In this study,radio weredevelopedfor andplaced known avian predatorspresenteda dual transmitters challengeto the techniqueof radio-track-on 10 barredowls thatwere capturedwith of the owls Japanesemistnets. Movements ing. University of Minwere monitored by the Radio-trackingtechniques have been Automatic Radionesota's Cedar Creek used most extensivelyin studiesof mamet al. 1965). Of tracking Station (Cochran mals,whichgenerallyhavea greatercapaclocations ( fixes ) some two million owl ity for carryingradio transmittersthan 28,000 were used in recorded, more than birds. Severalinvestigatorshave tried to equipbirdswith miniatureradiotransmit- data analysis. These locationswere used tersfor field studies,but few have doneso to determinethe intensityof use of seven successfully( Southern1965, Graberand differenthabitattypesduringall seasons. The authorsthankDr. D. B. Siniff,who Wunderle1966,Cochranet al. 1967). Habitatuse by birdshas beenstudiedby wrotethe computerprogramsused in data visualobservation, by recaptureof marked analysis;L. B. KuechleandC. Ransom,who birds,andby the appearance in studyareas built the radio transmittersused in this of species intermittently capturedby var- study;D. B. Lawrencefor the aerialphotoious samplingmethods. In addition to graphs;and MarieA. Gravdahl,who predetermininghabitatuse, data from these pared the line drawings.Appreciationis types of studies have been the basis for extendedto Dr. W. H. Marshall,Director major interpretations of responseto en- of the Universityof MinnesotaCedarCreek vironmental change,homerange,territory, NaturalHistoryArea;to A. Peterson,Resimigration,dispersal,and populationstruc- dent Managerof CedarCreek;to P. Rice, ture. our technician;and to the manypersonnel of CedarCreekwho gave freely of their 1 Present address: North Central Forest Experi- help. Theauthorsacknowledge the support ment Station, USDA Forest Service, Folwell Aveof Health Trainof the National Institutes nue, St. Paul, Minnesota. 213
S^,;,t,
214
......................
,
I
Journal of Wildlife Management, Vol. 36, No. 2, April 1972
ing Grant5 T1 AI 188 to Dr. D. W. Warner and the U. S. Atomic Energy Commission GrantCOO-133b67to Dr. J. R. Tester.
OCCURRENCEOF SEASONS CEDARCREEKNATURALHISTORYAREA MINNESOTA
STUDY AREA
Thisinvestigationwas carriedout on the CedarCreekNaturalHistoryArea,30 miles northof Minneapolis,Minnesota,in Anoka and Isanticounties. The some 5,000acres areownedandadministered by the University of Minnesota.Habitatwithinthe study areahas been describedin detailby Pierce ( 1954), Brayet al. ( 1959), andothers.The topographyis flat to gently rolling,with sandyuplandareasinterspersed with bogs, cattail ( Typha latifolia) marshes,swamps, four small lakes, a creek, and a drainage ditch. Upland areas typicallyhave sandy soil andwereclassifiedby Brayet al. (1959) as tallgrassprairie,deciduous-angiosperm EARLY LATE EARLY LATE savanna and forest, and mixed coniferSPRING SPRING SUMMER FALL FALL WINTER angiospermforest. The predominantupland trees are oaks (Quercus macrocarpa, Fig. 1. Occurrenceof seasons-Cedar CreekNaturolHistory Q. ellipsoidalis,Q. alba, Q. rubra) andpines Area. (Pinus strobus, P. resinosa, P. banksiana),
with scatteredsugarmaple (Acer saccharum), whitebirch(Betula papyrifera), and basswood( Tilia americana) . Thereis often a dense hazel (Corylus cornuta, C. americana) understory.Small farms and marginal croplandsurroundthe area. There were some cultivatedfields of rye (Secale cereale), alfalfa (Medicago sativa), corn (Zea ma?ys),and soybeans(Glycine max). Manyfields were not cultivated,and most fieldswereseparatedby rowsof treeswith understoriesof wild grape (Vitis spp.), Rubus species, and chokecherry( Prunus vtrgtnta) . .
*
.
\
The lowland areas have muck or peat soils usually poorly drained. These are normallydominatedby white cedar,tamarack ( Larix laricina), and alder (Alnus rugosa), withsuchassociatedspeciesas red osierdogwood(Cornusstolonifera), willow
(Salixspp.), white birch,and aspen ( Populus tremuloides). The generalclimatological characteristics of the CedarCreekNaturalHistoryArea have been recordedin Baker and Strub ( 1965) and Pierce ( 1954) . The general seasonal and phenologicalcharacteristics occurringduringthis study are illustrated in Fig. 1. Phenology Early Spring. Owls are nesting, occasional snow stormsoccur, killing frost is common,ice and snow are melting, the groundis bare, grassesare beginningto grow, tree buds expandand leaves grow to about one-halfsize, frogs start calling, and migratorybirdsreturn. Late Spring. Youngowls are about to
USE * Nicholls and Warner BARRED OWLHABITAT
215
leaveor have left the nests,tree leavesare fully grown, herbs and shrubs are fully developed,and mosquitoesand wood ticks (Dermacentorvariabilis) are numerous. Summer. Owls are in family groups, adult owls start moltingnear the end of summer, mosquitoesare present, wood ticks disappearabout mid-July,deer flies are ( Chrysopssp.) areactive,grasshoppers numerous,tadpoleshatch,somebirdsstart berries migration in AugustandSeptember, Fig. 2. Left: owl transmitteras it appeared before back and nuts maturein late summer,and July strap wos attachedand acrylicapplied. Right: the 70-gram temperatures rangefrom43 to 88 F. transmitter-harness used on barred owls. The transmitter Early Fall. Adult owls completemolt, operated 175 to 200 doys with a continuoussignal and had range of up to 3 miles when a MalloryZM12batterywas killingfrosts begin, mosquitoesand deer aused for power. flies disappear,leaves begin to turn color (most leaves are fallen by mid-October exceptfor some dead leaves persistingon with the backstrapon the midlineof the the oaks), and most of the Passerinebirds backbetweenthewings. Theharnessesand transmitters on all birds were completely leavethe area. an advantagein cold covered by feathers Late Fall. Waterfowlmigratethrough were kept weather because transmitters thearea,watersbeginto freeze,snowstorms warm. start,and saw-whetowls migratethrough the area. AutomaticRadio-trackingSystem Winter. Owls increase their daytime The Universityof Minnesota'sCedar hunting,owls begin courtshipactivitiesin System(Cochranet Creek Radio-tracking Februaryand March,most of the water rotatingdirectional al. 1965) utilizes two areasare frozen,snowstormsare common, Each radiotransmitted receiving antennas. andsubsnowcoveris normallycontinuous, When the radio at a different frequency. are commonin January zero temperatures the direction of an owl antennas pointed in andFebruary. with a transmitter,the radio frequency energy emittedfrom the transmitterwas MATERIAL AND METHODS detectedby receiversat a centrallaboratory. Owl Transmitter Permanentrecordsof receivedsignals,as was de- displayedby indicatorlights on receivers, A successfultransmitter-harness veloped( Nichollsand Warner1968) from were made by photography.The data on a solid strandof No. 12 or No. 14 cop- the filmwerereadandtabulatedto conform per electricalwire formedinto a harness to a samplingintervalthat normallywas with a neck loop and body loop to which one fix every 15 minutesduringdarkness transmitterparts were attached(Fig. 2). and one fix every30 minutesduringday"Transmitter partsandbatterywereencap- time,orasnearlyas possibleto thisinterval. du- Degreebearingsobtainedfromtheantennas sulatedin a waterproof,cold-resistant, rableacrylic... hardenoughto withstand permittedowls to be locatedon mapsby code The date,identification biting and clawing."The transmitterwas triangulation. locatednearthe frontof the owl'ssternum, foreachowl,time,anddegreebearingfrom
216
Journal of Wildlife lManagement,Vol. 36, No. 2, April 1972
if both an openfield and an each tower were punched on computer For example7 cardsand analyzedby a digitalcomputer oak woods appearedin the same 1.6-acre that calculated movement and location square and the computerlocation map and drewa mapof movements showedmostlocationsoccurringin the oak parameters for each animal (Siniff and Tester 1965> woods, the squarewas coded as an oak for woodshabitat.The habitatinformation Siniff1966). systemusing tri- eachsquare,alongwith all locationsoccurIn any animal-tracking be- ring in each squarefor each barredowl) angulationtheremay be discrepancies tween the true locationand the locations were programmedfor computeranalysis. Importantsources Theoutputfromtheprogramwaspresented found by triangulation. of errorsand limitationsof the automatic as the numberand percentageof locations systemhave been discussed occurringin each habitat type for any radio-tracking by HeezenandTester( 1967)andSargeant desiredunit of time et al. ( 1965). Cochranet al. ( 1965) deter- Habitats were categorizedinto seven mined that combinedsystemand reading types fouruplandtypes (oak woodsnoakerrorswere no greaterthan +0.3° with savannaopen field, and mixedhardwoods strongsignalson a calm day, and +0.5° and conifers) and three lowland types (white cedar swamp, alder swamp, and with strongsignalson a windyday. marsh). Brief descriptionsof these types HabitatAnalysis follow. The oakwoodstype usuallyconsistedof TheCedarCreekareawasgriddedintoa of two or threespeciesof oak ln mixture a (0.002S 1.6-acre 2,080, of systemconsisting high densities.Treesvariedin to moderate numwere squaremile) squares.Squares to 65 feet. Few shrubsand 20 from height beredfrom 1 to 2,080for computeridengrew underthese dellse plants herbaceous tification.Choiceof squaresize was arbihabitatconsisted oak-savanna The stands. trary,but 1.6-acresquareswereconvenient densitieswere where trees oak scattered of in termsof accuracyof the radio-tracking otherherand grasses that so enough low systemand alsopermitteddetailedclassifidomactual the were vegetation baceous cation of habitattypes. Squaresize was hardmixed largeenoughso thateachsquarecontained inantsof the community.The inter- woodandconifertype,dependinguponthe at least one possibledegree-bearing area, had variouscombinationsof sugar section. basswood red oak, white birch, maplen type habitat a assigned was Eachsquare pin oak,and white)red, and jack northern that was determinedfrom aerial photographs,vegetationmaps7andfield observa- pine. Few shrubsand herbaceousplants morethanone habitat grewunderthemixedhardwoodandconifer tions. Occasionally, type was presentin a 1.6-acresquare. To canopy. Open-fieldhabitatsconsistedof determinewhichhabitattype was used by pastures,cultivatedfields,andprairieswith clumpsof trees. Mostopenfields the owl in such cases, a computer-drawnoccasional by woodedareas. bordered were mapwas madeof all locationsfor eachowl andplacedovera vegetationmapof Cedar Whitecedarwas the dominantlowland Creekthat also had the 1.6-acregrid. The foresttree. In someareaswhitecedaroversquarewas coded for the most important lapped and mixed with tamarack.White in latesuccessional habitatin termsof the owl'spreference,as cedarreplacestamarack indicatedby the computermapof locations stages at Cedar Creek. The more dense
USE * Nicholls and Warner 0WL HABITAT BARRED
217
stands of white-cedarwere effective in type by the totalnumberof observedfixes reducingthe amountof light reachingthe obtainedfor each owl. of owl fixeswith forestfloor,andonlya few shrubsandherbs Dataon the distribution toleratedthese low light conditions.The respect to different types of available alder swamp habitat consistedof dense habitatwere tested for significantdifferstands of alder in moist lowland areas. ences by chi-squaremethodsof analysis. Somealderstandswereso densethatit was Theprobabilitylevel of 0.05percentor less difficultto walk throughthem. An occa- was used to indicate significance. The testwasstated sionaltamarackwas found in this habitat hypothesisforthe chi-square type. Marshareaswere characterizedby as follows: If an owl entered different open wetlandswith cattails,sedges, and habitatsby chance alone, the numberof grasses,generallywithouttrees but some- radiofixes in each habitatwill be proportimeshavingscatteredclumpsof red osier tional to availabilityof the habitattypes within the home range of the individual dogwood,alder,andwillowspecies. A determinationof home range was owl. To reduceerrorsin location,fixes were necessaryin evaluatinghabitatuse. The termhomerange,as usedin thisstudy,was taken only when owls were not flying, defined as that area used, traversed,or becausethe two antennasseldomreceived by an owl duringits signalsat exactlythe sametime. Whenthe regularlysurrounded mating, samplingintervalcouldnot be maintained normalactivitiesof food-gathering, nesting,caringforyoung,andseekingshel- because of random uncontrollabledata ter. The areaof eachowl'shomerangewas gaps,we consideredthat the percentageof determinedby multiplyingby 1.6 (acres) time unknownwas unbiasedtoward any thenumberof squaresin whichoneormore specifichabitattype. locations were obtained. Home range boundarieswere determinedby drawing RESULTS AND DISCUSSION linesaroundthe outermost1.6-acresquares Barredowlsadaptedquicklyto the transwithfixesin them. Thosefew squareswithand it did not appear to out fixes,but insidethe homerangebound- mitter-harness ary,wereconsideredpartof thehomerange influencetheir naturalactivitiesor cause by injuriesevenafterprolongeduse. Onlyone becausethey wereregularlysurrounded barredowl died duringthe study and its squareswith fixes. death was believedto be due to natural owl's home of each acreage After the as Severalowlscarriedtransmitters causes. the totalnumberof rangewas determined, acres of each of the seven habitattypes long as 180 days before the transmitters presentwithin the home range was also failed. determined.Habitatpreferenceand inten- Ten barredowlsweresuccessfullyradiosity of use were determinedby comparing trackedfor a total of 1,182days. During the observednumberof fixes occurringin this time, 28,338 locations (fixes) were eachhabitattype andthe expectednumber sampledfromsome two millionowl locarecordedby the radioof fixes that would have oucurredif owls tions automatically had enteredthe differenthabitattypes by trackingsystem(Table1) . Sufficientdata chancealone.Theexpectednumberof fixes forhabitatanalysiswereobtainedfrom9 of was obtaindby multiplyingthe percentage the 10 owls and comprisedsome 26,841 of the totalacreagepresentforeachhabitat radio locations.The averagehome range
218
Journal of Wildlife Management, Vol. 36, No. 2, April 1972
Table 1. Summaryof tronsmitterdays, dota doys, and numberof fixes obtained from burred owls during 1965-66 on the Cedor CreekNaturol HistoryArea. BARRED OWL
DATE RELEASED WITH TRANSMrrTER
END OF TRACKING PERIOD
TRANSMITTER DAYSa
DATA DAYSb
APPROX:[MATE NUMBEROF FIXESSAMPLED
701 702 7()3 704 707 709 710 711
S19-65 3-22-66 7-156S 11-12-65 2-15-66 2-23-66 3- 8-66 3-14-66
8- 9-65 S-19-66 9-20 65 2-13-66 W15-66 9-11-66 9-11-66 5-17-66
84 58 67 94 181 200 186 64
77 56 65 92 65 197 185 14
2,251 1,027 2,479 4,653 2,122 6,431 5,043 400
712 714
3-14-66 3-22-66
5- S-66 10- W66
52 196
51 99
2,587 1,345
1,182
901
28,338
Total a b
NOTES
Stoppedtransmitting Stoppedtransmitting Stoppedtransmitting Owl died Stoppedtransmitting Stoppedtransmitting Stoppedtransmitting Owl often out of receiver range Owl left area Stoppedtransnaitting
Transmitter days = the number of days the transmitter was functional. Data days _ the number of days in which usable locations were obtained.
size for the nineowlswas 565 acres. Home of fixes, dle particularwoodsthe owl was rangesize variedfrom213 to 912 acres. using. For example,the heavilyused oak Computermaps of owl locationsand woodsin the northwestcornerof Fig. 3 was movementsweremadefor all owls. Fig. 3 filledin withfixes. A pictureof thatwoodis an exampleof such a map for a 45-day lot can be seen at the top of Fig. 4 for period from August24 to September20, comparisorl.The open fields north and 1965,for barredowl No. 703. Each plus southof this oakwoodsreceivedlittle use. markon the mapdenotesone or moreradio The large oak woods to the south of the fixesobtainedfor the owl in thatparticular small open field in the middle of Fig. 4 location.Linesbetweenplusmarksindicate was intensivelyused, as can be seen by movementsbetweensuccessivelocations. the numerousplus markson the computer Computermaps,whenplacedovervege- mapin Fig. 3. tation maps of the study area ( Fig. 3 ), Comparison of mapsin Figs.3 and5 can illustratehow intensivelythe owl used dif- be madeby keepingin mindthe reference ferent areas within its home range. The pointsof the radio-tracking towers)Cedar woodedareaswere almostcompletelyout- Creek,CedarBog Lake, and the various lined with radio fixes, whereasthe open characteristicpatterns of the deciduous fields and marshon eitherside of Cedar woodlots.Barredowl 703 had a 258-acre Creekhad few fixes. In some cases the homerangefor the 6S-dayperiodbetween map areas representingdeciduouswoods July 1S and September20, 1965 (Fig. 5). were completelycoveredwith overlapping Eachsquareis 1.6 acresin size, and numplus symbolsdenotingowl locations.The bers indicatethe total radio fixes falling lines crossingthe open areasshowedthat within each square. A large numberof owls frequentlyflew back and forth be- radiofixesoccurredin the deciduouswoods tween differentwoodlotswithinthe home comparedwith the few fixes in the white range. By knowingthe characteristic out- areas,whichdenoteopenfieldsandmarshes line of each woods, it was often easy to (Fig. 5). Numerousradiofixes were obdetexmine, fromthe computermappattern tainedin someedge squares.Forexample,
s-l
El
at
+:f:f:i:t00
^"
...... ...
.E
E
.
S;Dt7
....
E
E
-
it:
E.:
£
'
sw
D
.; E
:
s
.:
:L:
-Li....-v: N? b
.
-::
::i:: :
-::
, :
j
E
i
*
;
7
f
f
1 2 2 1S
7
0
219
USE * Nichollsand Warnet OWLHABITAT BARRED
|
y .
ffic
,4
.,_
Lake 1b<M9
- B
M
, 1§ 1 2
2 2 3 22
1 1 101 }t1
6 128
a
map of locationsand movements Fig. 3. A computer-drawn of barredowl 703 fromAugust24 to September20, 1965. A total of 1,055 fixes were used in compilingthe map. White Fig. 5. The home ronge of barredowl 703 fromJuly 15 to areas on map are wooded, and shoded areas are open fields September20, 1965, was 258 acres. Eachsquareis 1.6 acres or marshes. Note the heavy use of wooded areas and the in size; the numbersin squaresindicatethe total numberof little use of open areas and marshes. radio fixes folling within eoch square. The white areas on the mop indicateopen fields or marshes. Note how little use was made of these areas comparedwith the intensivelyused the largestdeciduouswoods (woodlot E, deciduouswoods. >
Fig. 5) in the home rangehad both field andoakwoodshabitatsin someof the same squareson its northernedge. The question to be answeredwas whetherthe fixes in thosesquareswerelocatedin the openfield or in the oakwoods. Computermapsmade at 10-dayintervalsduringeachowl'stracking periodansweredthis questionfor they showedexactlywherethe owl was. In the examplecited,morethan98 percentof the fixes in the edge squaresoccurredin the oak woods. This explainedthe straight
g
mule
l/4
1
borderoutlinedby fixes along the open field and the northernedge of woodlotE (Fig.3). An aerialview of the samewoodsfield edge is seen in Fig. 4. Less use was made of the white cedar partof the home swampin the northeastern range. A mixed hardwoodand conifer ..
:
...
:
.. cy
{
Si.d
iSE:-E
i
7
:f.i.E:f:E,
0
Sv
'
&0
,;:S
:4::
:
:
:
--
E
*6^ia
A .
:::::
':i:
:::::LE.i:,
j , §8
Tower \
,
> § }/4
(
-t . '::0;
0:
_
.
;:
;:
:
A
:
1 0"^
-0-'-
, m!le l, I >,c89
'.;%egi;
S*iM3 ; fj
Je5ai3 :0
Fig. 6. Computermop of the fixes and movementsof barred owl 709 fromApril9 to April 19, 1966, based on 480 fixes. Fig. 4. Aerial view of open-field and oak woods hobitats. The intensivelyused area to the west of the loke is a mixed The objectsin the center of the open field are experimental hardwoodand coniferhabitat that is surroundedby a littleusedwhite cedor swamphabitat. duckpens.
220
Journal of Wildlife AIanagement, Vol. 36, No. 2, April 1972
habitat(woodlotC, Fig. 5) in the middle of a lowlandcedarswampreceivedintensive use. By comparison,squareshaving only white cedar in them had few fixes. Computermaps showed that most fixes occurredin themixedhardwoodandconifer habitatin those edge squaresalso having the whitecedarhabitat.Thesamerelationship was true for the deciduouswoods (woodlot F, Fig. 5) to the south of the white cedarswamp. Few or no fixeswere obtainedin the openmarsharealocatedon eitherside of CedarCreek. A differentbarredowl intensivelyused a mixedhardwoodand coniferhabitat( Fig. 6). A 10-daycomputermap of the movementsof No. 709 fromApril9 to April19, 1966,is shown. The numerousfixesto the west of CedarBog Lakeoutlinedan almost circularisland of mixed hardwoodsand conifersthatwascompletelysurrounded by a lowlandwhite cedar swamp. An aerial photographof the same area showedthe circularislandto the upperleft of the lake, surroundedby the cedar swamp, which receivedlittle use (Fig. 7). Barredowls showeddefiniteand significant (P < 0.05) preferenceor avoidanceof differenthabitattypes. An exampleof how habitatpreferencewas obtainedfor each
Fig. 7. Aerial photographof Cedar Bog Lakeshowingthe circulormixed hardwoodand coniferisland, adjacentto the lake, which was completelyoutlined by fixes, as seen in Fig. 6.
individualowl,by usingthe chi-squaretest to determine significantdifferencesbetween observedand expectednumberof radio fixesforeachhabitattype,is shownin Table 2. The resultsof these tests, summarized in Table3, showedwhichavailablehabitats were preferredby individualowls. The resultswere consistentfor all owls, with the exceptionof No. 707. This owl made intensiveuse of the oak-savanna habitatin contrastto five other owls that had this habitatin their home rangesbut did not makeintensiveuse of it. A total of 52 habitatcomparisons were
Table 2. Order of hobitat preferenceas indicated by the chi-square test for significant differences between observed and expected numberof radio fixes in each hobitat type for barred owl 709 {sex unknown)from February23 to September 11, 1966. HABITAT TYPE
Mixedhardwoodsand conifers Oak woods Wkte cedar swamp Adderswamp Marsh Open field Total
PERCENTAGE OF ACREAGE TOTALACRES AVAILABLE AVAILABLE
NUMBEROF RADIOFIXES ObSerVeda
EXPeCtedb
X2 VALUE
1,786.90*** 644.03*** 116.96*** 331.64*** 683.99*** 734.01***
126 279 238 42 97 130
14 30 26 5 11 14
2,081 2,869 1,296 4 52 148
903 1,93S 1,677 323 709 903
912
100
6,450
6,450
***P
Allen Press is collaborating with JSTOR to digitize, preserve and extend access to The Journal of Wildlife Management.
http://www.jstor.org
BY RADIOTELEMETRY BARREDOWL HABITATUSE AS DETERMINED THOMAS H. NICHOLLS,Department of Ecology and Behavioral Biology, University of Minnesota, Minneapolis DWAIN W. WARNER,Bell Museum of Natural History, Universily ofMinnesota, Minneapolis
Abstract: Radiotransmitters were successfullyplacedon 10 barredowls (Strixvaria) during1965-66, systemfor a totalof 1,182 and the movementof the owlswerefollowedwith an automaticradio-tracking days. Morethan 28,000 owl locationswere sampledfrom some two millionlocationsrecorded.These were programmed for computeranalysisto deterlocations,along with habitatavailabilityinformation, mineintensityof habitatuse. Barredowls showedconsistentandhighlysignificantpreferenceor avoidancefor differenthabitatsdespitechangingseasons,phenology,weatherconditions,and years. The order of habitatpreferencein decreasingintensityof use uZasoak ( Quercusspp.) woods,mixedhardwoodsand alder (Alnus spp.) swamps,marshes, conifers,whitecedar (Thufaoccidentalis)swamps,oak-savannas, and openfields.
Thispaperreportsthe use of radiotelem- Traditionalmarking methods seldom etry to determinethe intensityof use of producethe continuous,minute-to-minute varioushabitattypesby barredowls. The datanecessaryforevaluatingshort-duration recentlyacceleratedinterestin predatory movementsandactivitiesof wild birdsthat animalsand the difficultiesinvolvedin the will accuratelyprovideinformationon instudy of generally nocturnaland little tensityof habitatuse. In this study,radio weredevelopedfor andplaced known avian predatorspresenteda dual transmitters challengeto the techniqueof radio-track-on 10 barredowls thatwere capturedwith of the owls Japanesemistnets. Movements ing. University of Minwere monitored by the Radio-trackingtechniques have been Automatic Radionesota's Cedar Creek used most extensivelyin studiesof mamet al. 1965). Of tracking Station (Cochran mals,whichgenerallyhavea greatercapaclocations ( fixes ) some two million owl ity for carryingradio transmittersthan 28,000 were used in recorded, more than birds. Severalinvestigatorshave tried to equipbirdswith miniatureradiotransmit- data analysis. These locationswere used tersfor field studies,but few have doneso to determinethe intensityof use of seven successfully( Southern1965, Graberand differenthabitattypesduringall seasons. The authorsthankDr. D. B. Siniff,who Wunderle1966,Cochranet al. 1967). Habitatuse by birdshas beenstudiedby wrotethe computerprogramsused in data visualobservation, by recaptureof marked analysis;L. B. KuechleandC. Ransom,who birds,andby the appearance in studyareas built the radio transmittersused in this of species intermittently capturedby var- study;D. B. Lawrencefor the aerialphotoious samplingmethods. In addition to graphs;and MarieA. Gravdahl,who predetermininghabitatuse, data from these pared the line drawings.Appreciationis types of studies have been the basis for extendedto Dr. W. H. Marshall,Director major interpretations of responseto en- of the Universityof MinnesotaCedarCreek vironmental change,homerange,territory, NaturalHistoryArea;to A. Peterson,Resimigration,dispersal,and populationstruc- dent Managerof CedarCreek;to P. Rice, ture. our technician;and to the manypersonnel of CedarCreekwho gave freely of their 1 Present address: North Central Forest Experi- help. Theauthorsacknowledge the support ment Station, USDA Forest Service, Folwell Aveof Health Trainof the National Institutes nue, St. Paul, Minnesota. 213
S^,;,t,
214
......................
,
I
Journal of Wildlife Management, Vol. 36, No. 2, April 1972
ing Grant5 T1 AI 188 to Dr. D. W. Warner and the U. S. Atomic Energy Commission GrantCOO-133b67to Dr. J. R. Tester.
OCCURRENCEOF SEASONS CEDARCREEKNATURALHISTORYAREA MINNESOTA
STUDY AREA
Thisinvestigationwas carriedout on the CedarCreekNaturalHistoryArea,30 miles northof Minneapolis,Minnesota,in Anoka and Isanticounties. The some 5,000acres areownedandadministered by the University of Minnesota.Habitatwithinthe study areahas been describedin detailby Pierce ( 1954), Brayet al. ( 1959), andothers.The topographyis flat to gently rolling,with sandyuplandareasinterspersed with bogs, cattail ( Typha latifolia) marshes,swamps, four small lakes, a creek, and a drainage ditch. Upland areas typicallyhave sandy soil andwereclassifiedby Brayet al. (1959) as tallgrassprairie,deciduous-angiosperm EARLY LATE EARLY LATE savanna and forest, and mixed coniferSPRING SPRING SUMMER FALL FALL WINTER angiospermforest. The predominantupland trees are oaks (Quercus macrocarpa, Fig. 1. Occurrenceof seasons-Cedar CreekNaturolHistory Q. ellipsoidalis,Q. alba, Q. rubra) andpines Area. (Pinus strobus, P. resinosa, P. banksiana),
with scatteredsugarmaple (Acer saccharum), whitebirch(Betula papyrifera), and basswood( Tilia americana) . Thereis often a dense hazel (Corylus cornuta, C. americana) understory.Small farms and marginal croplandsurroundthe area. There were some cultivatedfields of rye (Secale cereale), alfalfa (Medicago sativa), corn (Zea ma?ys),and soybeans(Glycine max). Manyfields were not cultivated,and most fieldswereseparatedby rowsof treeswith understoriesof wild grape (Vitis spp.), Rubus species, and chokecherry( Prunus vtrgtnta) . .
*
.
\
The lowland areas have muck or peat soils usually poorly drained. These are normallydominatedby white cedar,tamarack ( Larix laricina), and alder (Alnus rugosa), withsuchassociatedspeciesas red osierdogwood(Cornusstolonifera), willow
(Salixspp.), white birch,and aspen ( Populus tremuloides). The generalclimatological characteristics of the CedarCreekNaturalHistoryArea have been recordedin Baker and Strub ( 1965) and Pierce ( 1954) . The general seasonal and phenologicalcharacteristics occurringduringthis study are illustrated in Fig. 1. Phenology Early Spring. Owls are nesting, occasional snow stormsoccur, killing frost is common,ice and snow are melting, the groundis bare, grassesare beginningto grow, tree buds expandand leaves grow to about one-halfsize, frogs start calling, and migratorybirdsreturn. Late Spring. Youngowls are about to
USE * Nicholls and Warner BARRED OWLHABITAT
215
leaveor have left the nests,tree leavesare fully grown, herbs and shrubs are fully developed,and mosquitoesand wood ticks (Dermacentorvariabilis) are numerous. Summer. Owls are in family groups, adult owls start moltingnear the end of summer, mosquitoesare present, wood ticks disappearabout mid-July,deer flies are ( Chrysopssp.) areactive,grasshoppers numerous,tadpoleshatch,somebirdsstart berries migration in AugustandSeptember, Fig. 2. Left: owl transmitteras it appeared before back and nuts maturein late summer,and July strap wos attachedand acrylicapplied. Right: the 70-gram temperatures rangefrom43 to 88 F. transmitter-harness used on barred owls. The transmitter Early Fall. Adult owls completemolt, operated 175 to 200 doys with a continuoussignal and had range of up to 3 miles when a MalloryZM12batterywas killingfrosts begin, mosquitoesand deer aused for power. flies disappear,leaves begin to turn color (most leaves are fallen by mid-October exceptfor some dead leaves persistingon with the backstrapon the midlineof the the oaks), and most of the Passerinebirds backbetweenthewings. Theharnessesand transmitters on all birds were completely leavethe area. an advantagein cold covered by feathers Late Fall. Waterfowlmigratethrough were kept weather because transmitters thearea,watersbeginto freeze,snowstorms warm. start,and saw-whetowls migratethrough the area. AutomaticRadio-trackingSystem Winter. Owls increase their daytime The Universityof Minnesota'sCedar hunting,owls begin courtshipactivitiesin System(Cochranet Creek Radio-tracking Februaryand March,most of the water rotatingdirectional al. 1965) utilizes two areasare frozen,snowstormsare common, Each radiotransmitted receiving antennas. andsubsnowcoveris normallycontinuous, When the radio at a different frequency. are commonin January zero temperatures the direction of an owl antennas pointed in andFebruary. with a transmitter,the radio frequency energy emittedfrom the transmitterwas MATERIAL AND METHODS detectedby receiversat a centrallaboratory. Owl Transmitter Permanentrecordsof receivedsignals,as was de- displayedby indicatorlights on receivers, A successfultransmitter-harness veloped( Nichollsand Warner1968) from were made by photography.The data on a solid strandof No. 12 or No. 14 cop- the filmwerereadandtabulatedto conform per electricalwire formedinto a harness to a samplingintervalthat normallywas with a neck loop and body loop to which one fix every 15 minutesduringdarkness transmitterparts were attached(Fig. 2). and one fix every30 minutesduringday"Transmitter partsandbatterywereencap- time,orasnearlyas possibleto thisinterval. du- Degreebearingsobtainedfromtheantennas sulatedin a waterproof,cold-resistant, rableacrylic... hardenoughto withstand permittedowls to be locatedon mapsby code The date,identification biting and clawing."The transmitterwas triangulation. locatednearthe frontof the owl'ssternum, foreachowl,time,anddegreebearingfrom
216
Journal of Wildlife lManagement,Vol. 36, No. 2, April 1972
if both an openfield and an each tower were punched on computer For example7 cardsand analyzedby a digitalcomputer oak woods appearedin the same 1.6-acre that calculated movement and location square and the computerlocation map and drewa mapof movements showedmostlocationsoccurringin the oak parameters for each animal (Siniff and Tester 1965> woods, the squarewas coded as an oak for woodshabitat.The habitatinformation Siniff1966). systemusing tri- eachsquare,alongwith all locationsoccurIn any animal-tracking be- ring in each squarefor each barredowl) angulationtheremay be discrepancies tween the true locationand the locations were programmedfor computeranalysis. Importantsources Theoutputfromtheprogramwaspresented found by triangulation. of errorsand limitationsof the automatic as the numberand percentageof locations systemhave been discussed occurringin each habitat type for any radio-tracking by HeezenandTester( 1967)andSargeant desiredunit of time et al. ( 1965). Cochranet al. ( 1965) deter- Habitats were categorizedinto seven mined that combinedsystemand reading types fouruplandtypes (oak woodsnoakerrorswere no greaterthan +0.3° with savannaopen field, and mixedhardwoods strongsignalson a calm day, and +0.5° and conifers) and three lowland types (white cedar swamp, alder swamp, and with strongsignalson a windyday. marsh). Brief descriptionsof these types HabitatAnalysis follow. The oakwoodstype usuallyconsistedof TheCedarCreekareawasgriddedintoa of two or threespeciesof oak ln mixture a (0.002S 1.6-acre 2,080, of systemconsisting high densities.Treesvariedin to moderate numwere squaremile) squares.Squares to 65 feet. Few shrubsand 20 from height beredfrom 1 to 2,080for computeridengrew underthese dellse plants herbaceous tification.Choiceof squaresize was arbihabitatconsisted oak-savanna The stands. trary,but 1.6-acresquareswereconvenient densitieswere where trees oak scattered of in termsof accuracyof the radio-tracking otherherand grasses that so enough low systemand alsopermitteddetailedclassifidomactual the were vegetation baceous cation of habitattypes. Squaresize was hardmixed largeenoughso thateachsquarecontained inantsof the community.The inter- woodandconifertype,dependinguponthe at least one possibledegree-bearing area, had variouscombinationsof sugar section. basswood red oak, white birch, maplen type habitat a assigned was Eachsquare pin oak,and white)red, and jack northern that was determinedfrom aerial photographs,vegetationmaps7andfield observa- pine. Few shrubsand herbaceousplants morethanone habitat grewunderthemixedhardwoodandconifer tions. Occasionally, type was presentin a 1.6-acresquare. To canopy. Open-fieldhabitatsconsistedof determinewhichhabitattype was used by pastures,cultivatedfields,andprairieswith clumpsof trees. Mostopenfields the owl in such cases, a computer-drawnoccasional by woodedareas. bordered were mapwas madeof all locationsfor eachowl andplacedovera vegetationmapof Cedar Whitecedarwas the dominantlowland Creekthat also had the 1.6-acregrid. The foresttree. In someareaswhitecedaroversquarewas coded for the most important lapped and mixed with tamarack.White in latesuccessional habitatin termsof the owl'spreference,as cedarreplacestamarack indicatedby the computermapof locations stages at Cedar Creek. The more dense
USE * Nicholls and Warner 0WL HABITAT BARRED
217
stands of white-cedarwere effective in type by the totalnumberof observedfixes reducingthe amountof light reachingthe obtainedfor each owl. of owl fixeswith forestfloor,andonlya few shrubsandherbs Dataon the distribution toleratedthese low light conditions.The respect to different types of available alder swamp habitat consistedof dense habitatwere tested for significantdifferstands of alder in moist lowland areas. ences by chi-squaremethodsof analysis. Somealderstandswereso densethatit was Theprobabilitylevel of 0.05percentor less difficultto walk throughthem. An occa- was used to indicate significance. The testwasstated sionaltamarackwas found in this habitat hypothesisforthe chi-square type. Marshareaswere characterizedby as follows: If an owl entered different open wetlandswith cattails,sedges, and habitatsby chance alone, the numberof grasses,generallywithouttrees but some- radiofixes in each habitatwill be proportimeshavingscatteredclumpsof red osier tional to availabilityof the habitattypes within the home range of the individual dogwood,alder,andwillowspecies. A determinationof home range was owl. To reduceerrorsin location,fixes were necessaryin evaluatinghabitatuse. The termhomerange,as usedin thisstudy,was taken only when owls were not flying, defined as that area used, traversed,or becausethe two antennasseldomreceived by an owl duringits signalsat exactlythe sametime. Whenthe regularlysurrounded mating, samplingintervalcouldnot be maintained normalactivitiesof food-gathering, nesting,caringforyoung,andseekingshel- because of random uncontrollabledata ter. The areaof eachowl'shomerangewas gaps,we consideredthat the percentageof determinedby multiplyingby 1.6 (acres) time unknownwas unbiasedtoward any thenumberof squaresin whichoneormore specifichabitattype. locations were obtained. Home range boundarieswere determinedby drawing RESULTS AND DISCUSSION linesaroundthe outermost1.6-acresquares Barredowlsadaptedquicklyto the transwithfixesin them. Thosefew squareswithand it did not appear to out fixes,but insidethe homerangebound- mitter-harness ary,wereconsideredpartof thehomerange influencetheir naturalactivitiesor cause by injuriesevenafterprolongeduse. Onlyone becausethey wereregularlysurrounded barredowl died duringthe study and its squareswith fixes. death was believedto be due to natural owl's home of each acreage After the as Severalowlscarriedtransmitters causes. the totalnumberof rangewas determined, acres of each of the seven habitattypes long as 180 days before the transmitters presentwithin the home range was also failed. determined.Habitatpreferenceand inten- Ten barredowlsweresuccessfullyradiosity of use were determinedby comparing trackedfor a total of 1,182days. During the observednumberof fixes occurringin this time, 28,338 locations (fixes) were eachhabitattype andthe expectednumber sampledfromsome two millionowl locarecordedby the radioof fixes that would have oucurredif owls tions automatically had enteredthe differenthabitattypes by trackingsystem(Table1) . Sufficientdata chancealone.Theexpectednumberof fixes forhabitatanalysiswereobtainedfrom9 of was obtaindby multiplyingthe percentage the 10 owls and comprisedsome 26,841 of the totalacreagepresentforeachhabitat radio locations.The averagehome range
218
Journal of Wildlife Management, Vol. 36, No. 2, April 1972
Table 1. Summaryof tronsmitterdays, dota doys, and numberof fixes obtained from burred owls during 1965-66 on the Cedor CreekNaturol HistoryArea. BARRED OWL
DATE RELEASED WITH TRANSMrrTER
END OF TRACKING PERIOD
TRANSMITTER DAYSa
DATA DAYSb
APPROX:[MATE NUMBEROF FIXESSAMPLED
701 702 7()3 704 707 709 710 711
S19-65 3-22-66 7-156S 11-12-65 2-15-66 2-23-66 3- 8-66 3-14-66
8- 9-65 S-19-66 9-20 65 2-13-66 W15-66 9-11-66 9-11-66 5-17-66
84 58 67 94 181 200 186 64
77 56 65 92 65 197 185 14
2,251 1,027 2,479 4,653 2,122 6,431 5,043 400
712 714
3-14-66 3-22-66
5- S-66 10- W66
52 196
51 99
2,587 1,345
1,182
901
28,338
Total a b
NOTES
Stoppedtransmitting Stoppedtransmitting Stoppedtransmitting Owl died Stoppedtransmitting Stoppedtransmitting Stoppedtransmitting Owl often out of receiver range Owl left area Stoppedtransnaitting
Transmitter days = the number of days the transmitter was functional. Data days _ the number of days in which usable locations were obtained.
size for the nineowlswas 565 acres. Home of fixes, dle particularwoodsthe owl was rangesize variedfrom213 to 912 acres. using. For example,the heavilyused oak Computermaps of owl locationsand woodsin the northwestcornerof Fig. 3 was movementsweremadefor all owls. Fig. 3 filledin withfixes. A pictureof thatwoodis an exampleof such a map for a 45-day lot can be seen at the top of Fig. 4 for period from August24 to September20, comparisorl.The open fields north and 1965,for barredowl No. 703. Each plus southof this oakwoodsreceivedlittle use. markon the mapdenotesone or moreradio The large oak woods to the south of the fixesobtainedfor the owl in thatparticular small open field in the middle of Fig. 4 location.Linesbetweenplusmarksindicate was intensivelyused, as can be seen by movementsbetweensuccessivelocations. the numerousplus markson the computer Computermaps,whenplacedovervege- mapin Fig. 3. tation maps of the study area ( Fig. 3 ), Comparison of mapsin Figs.3 and5 can illustratehow intensivelythe owl used dif- be madeby keepingin mindthe reference ferent areas within its home range. The pointsof the radio-tracking towers)Cedar woodedareaswere almostcompletelyout- Creek,CedarBog Lake, and the various lined with radio fixes, whereasthe open characteristicpatterns of the deciduous fields and marshon eitherside of Cedar woodlots.Barredowl 703 had a 258-acre Creekhad few fixes. In some cases the homerangefor the 6S-dayperiodbetween map areas representingdeciduouswoods July 1S and September20, 1965 (Fig. 5). were completelycoveredwith overlapping Eachsquareis 1.6 acresin size, and numplus symbolsdenotingowl locations.The bers indicatethe total radio fixes falling lines crossingthe open areasshowedthat within each square. A large numberof owls frequentlyflew back and forth be- radiofixesoccurredin the deciduouswoods tween differentwoodlotswithinthe home comparedwith the few fixes in the white range. By knowingthe characteristic out- areas,whichdenoteopenfieldsandmarshes line of each woods, it was often easy to (Fig. 5). Numerousradiofixes were obdetexmine, fromthe computermappattern tainedin someedge squares.Forexample,
s-l
El
at
+:f:f:i:t00
^"
...... ...
.E
E
.
S;Dt7
....
E
E
-
it:
E.:
£
'
sw
D
.; E
:
s
.:
:L:
-Li....-v: N? b
.
-::
::i:: :
-::
, :
j
E
i
*
;
7
f
f
1 2 2 1S
7
0
219
USE * Nichollsand Warnet OWLHABITAT BARRED
|
y .
ffic
,4
.,_
Lake 1b<M9
- B
M
, 1§ 1 2
2 2 3 22
1 1 101 }t1
6 128
a
map of locationsand movements Fig. 3. A computer-drawn of barredowl 703 fromAugust24 to September20, 1965. A total of 1,055 fixes were used in compilingthe map. White Fig. 5. The home ronge of barredowl 703 fromJuly 15 to areas on map are wooded, and shoded areas are open fields September20, 1965, was 258 acres. Eachsquareis 1.6 acres or marshes. Note the heavy use of wooded areas and the in size; the numbersin squaresindicatethe total numberof little use of open areas and marshes. radio fixes folling within eoch square. The white areas on the mop indicateopen fields or marshes. Note how little use was made of these areas comparedwith the intensivelyused the largestdeciduouswoods (woodlot E, deciduouswoods. >
Fig. 5) in the home rangehad both field andoakwoodshabitatsin someof the same squareson its northernedge. The question to be answeredwas whetherthe fixes in thosesquareswerelocatedin the openfield or in the oakwoods. Computermapsmade at 10-dayintervalsduringeachowl'stracking periodansweredthis questionfor they showedexactlywherethe owl was. In the examplecited,morethan98 percentof the fixes in the edge squaresoccurredin the oak woods. This explainedthe straight
g
mule
l/4
1
borderoutlinedby fixes along the open field and the northernedge of woodlotE (Fig.3). An aerialview of the samewoodsfield edge is seen in Fig. 4. Less use was made of the white cedar partof the home swampin the northeastern range. A mixed hardwoodand conifer ..
:
...
:
.. cy
{
Si.d
iSE:-E
i
7
:f.i.E:f:E,
0
Sv
'
&0
,;:S
:4::
:
:
:
--
E
*6^ia
A .
:::::
':i:
:::::LE.i:,
j , §8
Tower \
,
> § }/4
(
-t . '::0;
0:
_
.
;:
;:
:
A
:
1 0"^
-0-'-
, m!le l, I >,c89
'.;%egi;
S*iM3 ; fj
Je5ai3 :0
Fig. 6. Computermop of the fixes and movementsof barred owl 709 fromApril9 to April 19, 1966, based on 480 fixes. Fig. 4. Aerial view of open-field and oak woods hobitats. The intensivelyused area to the west of the loke is a mixed The objectsin the center of the open field are experimental hardwoodand coniferhabitat that is surroundedby a littleusedwhite cedor swamphabitat. duckpens.
220
Journal of Wildlife AIanagement, Vol. 36, No. 2, April 1972
habitat(woodlotC, Fig. 5) in the middle of a lowlandcedarswampreceivedintensive use. By comparison,squareshaving only white cedar in them had few fixes. Computermaps showed that most fixes occurredin themixedhardwoodandconifer habitatin those edge squaresalso having the whitecedarhabitat.Thesamerelationship was true for the deciduouswoods (woodlot F, Fig. 5) to the south of the white cedarswamp. Few or no fixeswere obtainedin the openmarsharealocatedon eitherside of CedarCreek. A differentbarredowl intensivelyused a mixedhardwoodand coniferhabitat( Fig. 6). A 10-daycomputermap of the movementsof No. 709 fromApril9 to April19, 1966,is shown. The numerousfixesto the west of CedarBog Lakeoutlinedan almost circularisland of mixed hardwoodsand conifersthatwascompletelysurrounded by a lowlandwhite cedar swamp. An aerial photographof the same area showedthe circularislandto the upperleft of the lake, surroundedby the cedar swamp, which receivedlittle use (Fig. 7). Barredowls showeddefiniteand significant (P < 0.05) preferenceor avoidanceof differenthabitattypes. An exampleof how habitatpreferencewas obtainedfor each
Fig. 7. Aerial photographof Cedar Bog Lakeshowingthe circulormixed hardwoodand coniferisland, adjacentto the lake, which was completelyoutlined by fixes, as seen in Fig. 6.
individualowl,by usingthe chi-squaretest to determine significantdifferencesbetween observedand expectednumberof radio fixesforeachhabitattype,is shownin Table 2. The resultsof these tests, summarized in Table3, showedwhichavailablehabitats were preferredby individualowls. The resultswere consistentfor all owls, with the exceptionof No. 707. This owl made intensiveuse of the oak-savanna habitatin contrastto five other owls that had this habitatin their home rangesbut did not makeintensiveuse of it. A total of 52 habitatcomparisons were
Table 2. Order of hobitat preferenceas indicated by the chi-square test for significant differences between observed and expected numberof radio fixes in each hobitat type for barred owl 709 {sex unknown)from February23 to September 11, 1966. HABITAT TYPE
Mixedhardwoodsand conifers Oak woods Wkte cedar swamp Adderswamp Marsh Open field Total
PERCENTAGE OF ACREAGE TOTALACRES AVAILABLE AVAILABLE
NUMBEROF RADIOFIXES ObSerVeda
EXPeCtedb
X2 VALUE
1,786.90*** 644.03*** 116.96*** 331.64*** 683.99*** 734.01***
126 279 238 42 97 130
14 30 26 5 11 14
2,081 2,869 1,296 4 52 148
903 1,93S 1,677 323 709 903
912
100
6,450
6,450
***P
