Factors Influencing Summer Habitat Use of Black-tailedDeer on South-Central Vancouver IslandByJEFFREY ALLEN MORGANB.Sc., Simon Fraser University. 1988A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTSFOR THE DEGREE OF MASTER OF SCIENCEinTHE FACULTY OF GRADUATE STUDIES(Departmeni of Forest Sciences)We accept this thesis as conformingto the required standardTHE UNIVERSITY OF BRITISH COLUMBIAAPRIL 1994©Jeffrey Allen MorganIn presenting this thesis in partial fulfilment of the requirements for an advanceddegree at the University of British Columbia, I agree that the library shall make itfreely available for reference and study. I further agree that permission for extensivecopying of this thesis for scholarly purposes may be granted by the head of mydepartment or by his or her representatives. It is understood that copying orpublication of this thesis for financial gain shall not be allowed without my writtenpermission.Department of__________________The University of British ColumbiaVancouver, CanadaDate_______DE-6 (2/88)IIAbstractUse of seral stage, biogeoclfrnatic variant, aspect and elevation by 13radio collared black-tailed deer (Odocoileus hemionus columbianus Richardson)was examined at the ‘within home range’ level of habitat selection. Pooleddata sets were evaluated for patterns of habitat use in relation to timeperiods within the summer of 1991. The July/August period of peak clearcutuse coincided with the period of peak fireweed (Epilobium angustfolium L.) useidentified by other research. The percentage of locations within clearcutswas 67.9% in April/May/June, 78.2% in July/August and 69.2% inSeptember/October. Fireweed, an early successional forb, has been shownby other research to be the most important black-tailed deer summer foragespecies. Similar trends in fireweed use and the percent of locations withinclearcuts therefore, were predictable.The use of Montane variants of biogeoclimatic subzones was higher inthe latter portions of the summer. The percentage of locations withinMontane variants was 14.5% in Aril/May, 22.6% in June/July and 25.8% inAugust/September. Use of Montane habitat was lowest in the spring whenforage species within the Submontane were more phenologically advancedand more nutritious. Shifts in elevation and changes in the use of aspectwere not detected.Parturient deer did not change their use of seral stage within thefawning period. Home range size (minimum convex polygon) was greater inthe fawning period than in any other period. Increases in home range sizeduring the fawning period were caused by increases in the magnitudeand/or rate of movements to the peripheries of the home ranges. Thisfinding supports the idea that sites on the periphery of the dam’s home rangeare used for fawning, and is consistent with the ‘hider strategy’ model. Hidersliiare known to remove their young from the birth site to prevent predatorsfrom using olfactory cues to detect their young. If black-tailed deer were togive birth within their core areas of use, they would be forced to remove theirfawns from the habitats they most IDrefer.Home range size was smaller in the immediate post-fawning periodthan in all other periods except the pre-natal period. This behavior is againconsistent with the hider strategy. Immediately after fawning, the dam mustremain close to her fawns, which are sedentary. Habitat use in the earlysummer was more concentrated in Submontane habitats where green-upoccured first. Home range size in the pre-natal period therefore, should alsohave been small.ivTable of ContentsAbstractList of Tables viList of Figures viiAcknowledgments viiiChapter 1: General Introduction 1Rationale and Objectives 1Forage quality 1Habitat use during fawning season 3Study Area 4General Methods and Materials 8Description of Study Animals 11Chapter 2:Changes In Habitat Use In Relation toForage Quality 14Introduction 14Seral stage 14Delayed phenology 17Methods and Materials 18Results 21Seral stage 22Delayed phenology 24Potential confounding variables 27Discussion 30Seral stage 30Delayed phenology 32Chapter 3: Changes in Habitat Use by FemaleBlack-tailed Deer During the Fawnihg Season 34Introduction 34Methods and Materials 37Results 39Habitat use 39Fawning site selection 39Home range size 46Discussion 50Habitat use 50Areaof use 51VChapter 4: Summary and Conclusions...........54Changes in Habitat Use in Relaticn to Forage Quality 54Changes in Habitat Use During the Fawning season 55Recommendations and Implications 56Literature Cited 58List of TablesTable 1.1. General information on study animals. 1989-199 1 ...12Table 2.1. Data collection periods for collared deer, 1991 19Table 2.2. Number of locations per deer per month, 1991 22Table 2.3. Number of locations within clearcut andforested habitats and percent of locations withinclearcut habitats,1991 23Table 2.4. Number of locations within Montane andSubmontane habitats and percent of locations withinMontane habitats, 1991 25Table 2.5. Percent clearcut habitat within the Montaneand Submontane variants within the 95% MCP homeranges of study animals 17701 and 18181 29Table 2.6. Percent use comparison of Montane andclearcut habitats for study animal 17701 29Table 3.1. Home range size (hectares) by time period 47Table 3.2. Paired difference test comparison of homerange size during the fawning period versus allother periods 49Table 3.3. Paired difference test comparison of homerange size during the immediate post-fawningperiod versus all other periods 50vilList of FiguresFigure 1.1. Location and major subdrainages of theCaycuse Study Area 6Figure 2.1. Monthly pattern of use by black-tailed deerof forage species in forested and cutover areas(Rochelle 1980. p.64) 16Figure 2.2. Percent of pooled locations withinclearcuts, 1991 24Figure 2.3. Percent of pooled locations withinMontane, 1991 26Figure 3.1. The 1989 fawning movements of study animal18181 41Figure 3.2. The 1990 fawning movements of study animal18181 42Figure 3.3. The 1990 fawning movements of study animal13201 43Figure 3.4. The 1991 fawning movements of study animal17991 44Figure 3.5. The 1991 fawning movements of study animal12801 45Figure 3.6. Mean summer home rapge size and standard errorfor 10 female deer in five periods during the summerof 1991 48viiiAcknowledgmentsI am most grateful to my wire, Beth who was foolish enough to marryme shortly after I had begun my graduate studies. Her emotional andfinancial support saw me through challenging times. My mother and father,Art and Lois, supported me likewise. Thanks also to my sister Shelley who, bydemonstrating that university bould be an enjoyable experience,encouraged me to advance. Other positive influences were Dave Dunbar,Adam Lewis and Ken Budd. Many other friends and family members havesupported me at different levels over the past few years. Their help andpatience was appreciated.Dr. F.L. Bunnell supervised my post graduate studies and research. Hiscommittment to science and ecucation ensured that I was challenged.Thanks also to my committee members, Dr. M. Pitt, Dr. T. Sullivan and Dr. A.Derocher who helped fashion the study design and/or provided valuablecriticism. My ‘operational supeMsor, Scott McNay, arranged logistic,financial, and technical assistance and was a constant sounding board.Technicians who assisted in this research were Janet Windborne, Dan Bate,Yves Giguere, Line Giguere and Wayne Wall. Other members of theVancouver Island, wildlife biology community that helped with this researchwere Marvin Eng, Dr. R. Page, Joan Voller, Myke Chutter, Don Doyle, KimBrunt and Doug Janz. Les Peterson deserves special thanks. Were it not forhis altruistic nature and analytical abilities, I might still be writing up. Under histuition I learned many of the skills required to complete this thesis. DerekWilkinson provided timely assistance with the the figures. Jackie Johnson,office manager of the unmanageable Header House, kept my academiccareer in order, proof read manuscripts and was a friend. Thanks also to theHeader House folks for their encouragement, assistance, criticism andfriendship.This project was funded by a B.C. Ministry of Forests/Forestry Canada-South Moresby Forest Replacement Agreement grant to the ResearchBranch of the B.C. Ministry of Forests. Fletcher Challenge Canada Ltd.,Integrated Resource Analysis Section, contributed finances, equipment andlogistical support. The B.C. Ministry of Environment Lands and Parks and TheB.C. Ministry of Forests (Integratec Wildlife Intensive Forestry Research) alsocontributed equipment and logistical support. My personal funding wasprovided by the Science Councii of British Columbia through a G.R.E.A.T.award to Fred Bunnell. Bob Willington, of Fletcher Challenge Canada Ltd.,graciously cooperated in the arrangement of this funding.1Chapter One: ‘General IntroductionRationale and ObjectivesForage QualityIn the Temperate and Arctic zones of the Northern Hemisphere,availability of high quality winter range is often considered to be the mostimportant factor limiting conditioh and size of ungulate herds (Connolly1981). Consequently, the habitat ecology of ungulates during other periodsof the year has received relatively little attention. When habitat use andrange characteristics during the summer are documented, it is often doneonly to provide a comparison between winter and summer ranges.In many instances winter range quality is indeed the most importantfactor limiting herd productivity; there are, however, pitfalls in viewing thisconstraint in isolation of others. Although mortalities resulting frommalnutrition usually occur in late winter or early spring, they are a function ofboth winter forage availability and pre-winter conditioning (Short 1981). Thepre-winter conditioning that occurs while the animals are on their summerranges is often taken for granted. In many carrying capacity models, (e.g.,Errington 1945; Potvin and Huot 1 9S3; Bartmann et al. 1992), the quality of thesummer range is not considered. It is assumed that during the summer,ungulates have access to surplus quantities of forage. It follows, that theyenter winter in peak condition and that, with respect to over-winter mortality,the quality of the summer range is inconsequential.There is evidence that suggests this is not always the case. In aridregions, mule deer (Odocoileus hemionus hemionus) populations have beenfound to be limited by the quality of their summer ranges (Longhurst et al.1952; Russo 1964; Robinette et al. 1977; Smith and LeCount 1979).2In moister regions gradual migrations of ungulates suggest individualsèonstantly seek out fresh, high quality summer pastures to improve theircondition. Where such migrations occur, the spatial arrangement of differenthabitat types over the landscape, and the strategies that ungulates use toexploit them, are probably important factors affecting carrying capacity.Carrying capacity is defined as “the number of animals that a habitatmaintains in a healthy and vigorous condition” (Dasmann 1981: 151).Large herbivores such as mule deer, Rocky Mountain elk (Cervus elaphusnelsoni) and moose (Alces alces) have been observed to engage in gradualelevational migrations during summer months (Altmann 1952; Edwards andRitcey 1956; leResche 1974). Some authors have found mule deer toundertake such movements in response to plant phenology (Russell 1932;Leopold et al. 1951). Bunnell (1990: 61) observed that “migratory deer thatmove to alpine and subalpine areas as the season progresses are able toprofit from the higher quality forage found there during the late summer andfall”.Because most of the black-tailed deer (Odocoileus hemionus columbianusRichardson) research in the Pacific Northwest has focused on winter range,relatively little is known of the species’ summer habitat ecology. In themountainous habitats of this region, black-tailed deer populations usuallyhave a component of elevational migrators (Harestad 1979; Loft et al. 1984;Schoen and Kirchhoff 1985; McNay and Doyle 1987). Although differences inelevation, aspect, and vegetation types between summer and winter rangesare often presented, the data sets are seldom scrutinized for gradualchanges in habitat use that occur within the season. Seasonal home rangesare depicted as static entities Id which constant foraging strategies areemployed. Our knowledge of broad seasonal habitat requirements is3extensive, but potential changes in habitat use in response to changes inforage quality within a season are not well understood. During the growingseason, black-tailed deer probably make constant adjustments to theirforaging strategies to maintain quality in their diet. Appropriate foragingstrategies could entail adjustments to the position of the home range inresponse to changes in vegetation quality (e.g., plant phenology).Alternatively deer might adjust their habitat use patterns within static homeranges.The primary objective of Chapter Two is to report the use of habitattypes within the summer range. A secondary objective, when appropriate, isto relate the changes in habitat use to gradual elevational migrations orrange expansions.Habitat Use During Fawning SeasonOne aspect of ungulate summer habitat use that has receivedconsiderable attention is birth site selection. Ungulate mortality rates areusually highest during the first few weeks after birth (Cook et al. 1971; Jacksonet al. 1972; Smith and LeCount 1979; Steigers and Fllnders 1980; Ozoga andVerme 1986). Neonate mortality from predation is sometimes a major limitingfactor. Hatter (1988) estimated fawn mortality, within a northern VancouverIsland black-tailed deer herd, to be 63%, and concluded that wolf (Canislupus) predation was the primary factor limiting deer recruitment.Often, specific habitat types that confer anti-predator advantages tothe neonate and the mother are used during calving or fawning seasons.Steep, broken and rocky sites are often selected as natal sites by bighornsheep (Ovis canadensis) (Festa-Bianchet 1988). Rocky Mountain elk neonatesused sites with dense brush cover (Johnson 1951; Altmann 1952). In Utah,mule deer avoided giving birth ir sites with dry exposures and poor cover4(Robinette et al. 1977). Moose, caribou (Rangifer tarandus), and mule deerhave been found using small islands as birthing sites apparently to reducethe risk of predation (Steigers and 9inders 1980; Edwards 1983; Bergerud andPage 1987).Other authors have associated increased intraspecific avoidance andantagonistic behaviours with calving and fawning seasons. Prior to calving,parturient mountain caribou (R. t. caribou) dispersed to high elevation sites toevade predation (Cichowski 1989). Bergerud et al. (1984) suggest that byspacing themselves out, caribou Increase the predator searching time forcalves and make it difficult for predators to subsist on a diet of cariboucalves. Similarly, mule deer seek isolation from conspecifics before givingbirth (Robinette et al. 1977; Riley and Dood 1984) and pregnant black-taileddeer are known to be highly antagonistic during the fawning period(Dasmann and Taber 1956; Miller 1974).Despite the knowledge that has been gained on the topic in general,the nature of fawning habitat requirements and birth site selection in black-tailed deer is poorly understood. The objectives of Chapter Three are:1) to compare the use of seral stages during the fawning period versus therest of the summer, and2) to compare home range size during the fawning period with home rangesizes in other summer time periods. Minimum convex polygon technique(Mohr 1947) is used as an index of peripheral movements.Study AreaThis study took place in the upper Caycuse Watershed, above andincluding Wilson Creek (Figure 1.1). This southwestern Vancouver Island5system flows westward into the head of Nitnat Lake, which in turn emptiesinto the Pacific Ocean midway between the town sites of Bamfield and PortRenfrew. With its center at appro 500 stems/ha) with a commerciallydesirable (coniferous) species. The date of establishment was determined bythe age of the tree stock. Clearcut habitats were less than 16 years postestablishment in 1990. Second-growth habitats were at least 16 years post-20establishment in 1990. Old-growth habitats were at least 150 years old andhad never been logged. Because both second-growth and old-growthforests contained low quantities of herbaceous forage, they wereamalgamated to form the ‘forested’ habitat type.Most habitat use studies compare the use of particular habitat typeswith their availability, to develop preference ratings (Neu et al. 1974, Johnson1980, Alldredge and Ratti 1986). Because the intent of this study is todocument changes in the use of various habitat types through time, thesecalculations are not necessary. Instead the summer was divided into timeperiods which were, with one exception, two months long. Percent use ([no.locations within a habitat type/total no. locations] x 100) of habitat types orclasses for each period are then compared.Time periods developed to test for shifts in clearcut use wereApril/May/June, July/August, and September/October. These periods werebased on changes in Rochelle’s (1980) percent importance value (lV%) andnutritive values of fireweed. Fireweed’s IV% were relatively low during April,May and June (see Figure 2.1). The percent importance value rose sharply toa peak in July and remained high through to October. Because the nutritivevalues (digestible dry matter and caloric content) of fireweed rapidlydeclined after August (Rochelle 1980), the data were separated further intoJuly/August and September/October time periods.Time periods developed to test for increased use of cooler sites wereApril/May (spring), June/July (summer) and August/September (early fall).October’s data were not included in this test because: 1 )they were notrequired to test the hypotheses, 2) the possibility of a downward elevationshift in October in response to cooler weather existed and 3) by late August,any upward elevational migrations should have ceased.21Mohr’s (1947) minimum convex polygon technique (MCP) was used toestimate the home range boundaries of the study animals. The 95% MCPhome range, defined as the smallest convex polygon containing 95% of thesample locations (Akerman, Leban, Samuel and Garton 1989). was used.The G-test (Sokal and Rohlf 1981) was used to test for differences in theuse of the habitat types between established time periods. A one-tailedpaired difference test (Sokal and Rohlf 1981) was used to test for changes inthe elevation of home range centers and the average location distancefrom the home range center during different time periods. A type one errorprobability (p-values) less than 0.05 was considered to be ‘strong evidenceagainst the null hypothesis’ (Ho). A probability between 0.05 and 0.10 wasconsidered to be ‘moderate evidence against the null hypothesis’. Aprobability greater than 0.10 was considered to be ‘no evidence against thenull hypothesis’.ResultsIn the summer of 1991, 812 deer locations were recorded. Eleven deerwere monitored during this period; however, one (15301) was killed by wolvesin mid-July. Table 2.2 presents the number of locations per collared deer permonth.In the summer of 1990, 11 deer were monitored and 542 locationswere recorded.22Table 2.2. Number of locations per deer per month, 1991.Deer April May June July Aug. Sept. Oct. Total13201 11 14 13 11 14 9 12 8416781 10 14 12 10 14 9 12 8117581 8 14 12 10 14 9 12 7913891 8 14 10 10 13 10 12 7717991 9 13 12 11 14 10 12 8118181 8 14 13 11 14 10 12 8217701 4 14 12 11 14 11 12 7817912 4 14 13 11 14 10 12 7815301 3 12 8 7 na na na 3012801 4 14 9 11 14 10 12 7415181 3 14 6 11 13 10 11 68TOTAL 72 151 120 114 138 98 119 812MEAN 6.5 13.7 10.9 10.4 13.8 9.8 11.9 78.2Seral StageIn 1991 the percent use of clearcut habitats was highest in the July!August period (Table 2.3). Pooled location data, for the nine deer which hadclearcuts in their home range (708 locations), reveal that the percent use ofclearcuts was 67.9% in April/May/June, 78.2% in July/August and 69.2% inSeptember/October (Table 2.3, Fig 2.2). The G-test (G=7.37, p0.025)provided strong evidence against the null hypothesis which predicted thatthe percent use of clearcuts would not change as the summer progressed.Over the entire summer, less than 3% of the locations were within oldgrowth.23Table 2.3. Number of locations withn clearcut and forested habitats andpercent of locations within clearcut habitats, 1991.*number of locations within clearcut and forested habitatsAiril/Ma /June Julv/Auciust Set.IOct. Aoril-Octoberdeer Id. cc:for* cc cc:for cc cc:for cc cc:for cc%** % % %13201 38:0 100.00 25:0 100.00 15:6 71.43 78:6 92.8513891 17:15 53.13 21:2 91.30 12:10 54.55 50:27 64.9315181 0:23 0.00 0:24 0.00 2:19 9.52 1:67 1.4716781 33:3 91.67 21:3 87.50 17:4 80.95 71:10 87.6517581 22:12 64.71 23:1 95.83 21:0 100.00 66:13 83.5417701 28:2 93.33 25:0 100.00 22:1 95.65 75:3 96.1517912 31:0 100.00 22:3 88.00 20:2 90.91 73:5 93.5817991 20:14 58.82 21:4 84.00 15:7 68.18 56:25 69.1318181 10:25 28.57 14:11 56.00 11:11 50.00 35:47 42.68pooled 199:94 E 67.9 j 172:48 78.2 135:60 69.2 506:202 71.5**=percent of locations within clearcut habitats24807876C)74.$272.! 7068::62, IApril-June July/Aug. Se pt/OctTime PeriodsFigure 2.2. Percent of pooled locations within clearcuts, 1991.These data compare favorably with the 1990 data (464 locations)where the percent use of clearcuts for the July/August andSeptember/October time periods was 78.2% and 73.7% respectively.Delayed PhenologyFor the eight deer which had Montane habitat within their home range487 locations) percent use (pooled data) of Montane habitat was 14.5% inApril/May, 22.6% in June/July and 25.8% in August/September (Table 2.4, Fig.2.3). The G-test (G=6.67, p=0.036) provided strong evidence against the nullhypothesis which predicted that the percent use of Montane habitat wouldnot change as the summer progressed.25Table 2.4. Number of locations within Montane and Submontane habitats andpercent of locations within Montcine habitats, 1991.August/Sept.(early fall)*=number of locations within Montane and Submontane habitats**=percent of locations within Montane habitatsApril/May(sDring)June/July(summer)April-Septemberdeerid. Mon:Sub Mon. Mon:Sub Mon. Mon:Sub Mon. Mon:Sub Mon.* % 70 7012801 0:18 0.00 0:20 0.00 3:21 12.50 3:59 4.8415181 6:11 35.29 11:6 64.71 14:9 60.87 31:26 54.3915301 0:15 0.00 5:10 33.33 no na 5:25 16.6717581 1:21 4.55 0:22 0.00 3:20 13.04 4:63 5.9717701 0:18 0.00 4:19 17.39 6:19 24.00 10:56 15.1517912 4:14 22.22 10:14 41.67 7:17 29.17 21:45 31.8217991 0:22 0.00 0:23 0.00 1:23 4.17 1:68 1.4518181 11:11 50.00 8:16 33.33 9:15 37.50 28:42 40.00pooled 22:130 14.5 f 38:130 22.6 43:124 25.8 J 103:384 j 21.22630 -to25-.S2015=oSpring Summer EarlyFeIlTime PeriodsFigure 2.3. Percent of pooled locations within Montane habitats, 1991.These results compare favorably with the 1990 data set which hadpercent use of Montane habitat values of 19.8% to 24.3% for June/July andAugust/September, respectively.The percent use of north and east (cooler) aspects was not higher inthe latter portion of the summer. Instead the 1991 pooled data set for all ofthe 1991 study animals (693 locations) revealed that the percent use of theseaspects combined was 26.5% for April/May, 26.1% for June/July and 24.2% forSeptember/August. When only location data from the eight deer which hadMontane habitats within their summer range (487 locations) were used, thepercent use of north and east aspects was not higher in the latter portion ofthe summer. The percent use of north and east aspects combined was36.2% for April/May, 35.7% for June/July and 32.3% for September/August.The distribution of Montane variants over the landscape is a function of0-27elevation and aspect. The higher use of Montane habitats in the latterportions of the summer without a similar trend in the use of east and northaspects suggested that the study animals were using higher elevations as thesummer progressed. The elevations of the home range centers of the deerwhich had Montane habitats within their home ranges were averaged andcompared. For the April/May, June/July and August/September time periodsthe mean elevations were 660.0 m (SE=33, n=7), 666 m (SE=51, n=7) and 674meters (SE=43, n=7) respectively (1530 excluded). The mean elevation gainper deer between the home range centers of April/May andAugust/September was 14 m (SE=11, n=7). Although the mean elevationchange was positive, a paired difference test provided no evidence againstthe null hypothesis (t=0.22, p=0.83) which predicted no change in elevation.An alternate hypothesis could be formulated to account for theincreased use of Montane habitats. If the majority of its home range waslocated below Montane variants, a deer could increase its use of Montanehabitats, without shifting its home range center (and therefore its averageelevation) by expanding its home range equally along the axisperpendicular to the Montane/Submontane ecotone. Such an expansiondid not occur. The average distance between the locations and theirrespective home range centers for the spring and early fall periods were 214m, and 205 m respectively.Potential Confounding VariablesThe pattern of logging in many Vancouver Island watersheds makes itdifficult to distinguish between increases in the use of Montane habitats andincreases in the use of clearcut habitats. Montane variants, were often thelast to be logged and as a result they usually contain a high proportion ofrecently clearcut habitats. Conversely, Submontane variants usually contain28a high proportion of second-growth habitats. Before any generalizations onchanges in the use of seral stage and or BEC variants can be made, thepotential for these variables to be confounding must be assessed.Animals 12801 and 15301 were not included in seral stage analysisbecause their summer ranges consisted entirely of one seral stage (second-growth). Similarly, animals 13201, 13891, and 16781 were not included in theBEC variant statistical analysis because their home ranges consisted entirelyof Submontane variants. The 6 study animals used in both analyses wereanimals 15181, 17581, 17701, 17912, 17991 and 18181. The Montane portionsof the home ranges belonging to animals 15181. 17581 and 17991 consistedentirely of second growth habitats; therefore, increases in the percent use ofMontane and clearcut habitats could not have been correlated. Similarly,the Submontane portions of the home range occupied by animal 17912consisted entirely of clearcuts. Again, increases in the percent use ofMontane and clearcut habitats could not have been correlated. A breakdown of seral stage habitat types within the Montane and Submontaneportions of the 95% MCP home ranges of the remaining two study animals ispresented on Table 2.5.The percentage of clearcut habitats within the home range of animal18181, was lower in the Montane variants than in the Submontane variants(Table 2.5). Thus, there was not a potential for increases in the percent use ofMontane and clearcut habitats to be correlated. Within the home range ofanimal 17701 the percentage of clearcut habitats was higher in theMontane variants than in the Submontane variants.29Table 2.5. Percent clearcut habitat within the Montane and Submontanevariants within the 95% MCP home ranges of study anImals 17701and 18181.Deer Id. Percent Clearcut Habitat Percent Clearcut Habitatwithin Montane within Submontane17701 100.0 90.918181 11.8 38.2The location data were separated by month and inspected forcoincidental increases in the use of Montane and clearcut habitats. Thesedata are summarized in Table 2.6.Table 2.6. Percent use of Montane and clearcut habitats for study animal17701.Month April May June July Aug. Sept. Oct.% Clearcut 100.0 85.7 100.0 100.0 100.0 100.0 91.7% Montane 0.0 0.0 25.0 9.1 21.4 27.3 0.0As Table 2.6 demonstrates, animal 17701 used clearcuts regardless ofwhether they occurred in Montane or Submontane variants.30DscussionSeral StageAn increase in use of clearcuts by deer during July and August is oftennoted by biologists who conduct census work on Vancouver Island withinthat period. The deer hunting sedson, which on Vancouver Island starts inearly September, may force deer into areas of greater visual screening andout of clearcuts. This however, would not explain the relatively low use ofclearcuts during the early summer.The observed trend in cledrcut use, for the Caycuse Study Area in1991, was similar to the trend documented by Rochelle (1980) for fireweeduse (IV%) in the Nimpkish Valley. Clearcut use and lV% were lowest duringApril/May/June, highest during July/August and of intermediate values inSeptember/October.Old-growth and second-growth habitats within the summer ranges ofthe nine study animals, supported very little herbaceous forage. In theadjacent clearcut areas howevei, large quantities of fireweed and otherpioneering herbs grew vigorously. It is not surprising therefore, that theincreased fireweed use (IV%) coincides with an increase in the use ofclearcut habitats.It is difficult to determine whether the increased IV% of fireweed andthe increased use of clearcuts are a function of forage availability or foragequality. Rochelle (1980) noted th9 when availability is high for a variety offorage species, deer appear to e selecting the forage with the highestnutritive value. Klein (1965) suggested that unless forage quality in thesummer months is maintained, deer cannot consume enough to meet theiroptimum growth requirements.31Hanley (1984) found that it, the western Cascades of WashingtonState, the availability (measured as biomass) for all forage species peaked inJuly. The biomass of fireweed increases until after its flowering stage inAugust however, only the most digestible, top portion of the plant isconsumed (Moen 1981, pers. obs.). I believe that, between emergence anddie-back, the availability of fireweed tops at any given site is constant.Fireweed availability, therefore, does explain the percent use of clearcuts orthe lV% for fireweed which were highest in July and August.The nutritive quality (dry natter digestibility and caloric value) offireweed is at its peak in August. The percent dry matter digestibility (DDM) offireweed increased steadily from 60% in May to 80% in August then backdown slightly to 78% in October (lochelle 1980). Similarly, the caloric valueof fireweed rises sharply from 3 kccil/0.8g in May to 6 kcal/0.8g in August thenback to less than 3 kcal/O .8g in October (Rochelle 1980).A variety of non-herbaceous forage species do persist beneath forestcanopies. Rochelle (1980) found that, within forested habitats, Vacciniumalaskense and Vaccinium parvfolium followed trends similar to that of fireweed.However, peak DDM values, of approximately 60%, occurred much earlier inMay. Caloric values rose sharply to a plateau of approximately 5 kcal/0.8gwhich lasted from April to July. Given their similar phenologies (Haeussler etal. 1990), it is probable that other important shrub forage species, such as theRubus spp. followed similar nutritive cycles.In the early summer therefore, the nutritive values (DDM and Kcal/g) ofseveral alternate forage species, found within forested habitats, arecomparable to that of fireweed. As summer progresses however, fireweedexceeds the alternate forage species in nutritive value and it dominates thediet (lV%). The trends in the percent use of clearcut habitats and the lV% of32fireweed are similar.Clearcut habitats possess fireweed in its highest density and vigor(Haeussler et al. 1990). By selectihg clearcuts as foraging sites in July andAugust, deer maximize their foragihg efficiency. In doing so, they are able tomaintain quality in their diet and remain inactive for a greater proportion ofthe day. Inactivity allows deer to be more vigilant and it decreases theprobability of detection by predators.Delayed PhenologyDuring the spring months, forage phenology within the coolerMontane habitat is delayed (Geiger 1961, Aitken 1974). At this time theforage within the warmer Submontane habitat is more phenologicallyadvanced and more nutritious. Not surprisingly, the percent use ofSubmontane habitat is highest in the April/May period. As summer progressesthe more profitable phenological stages become available in the Montanevariants and the percent use of this habitat increases.Mackie (1970) found that mule deer used drier habitat types less oncethe forage plants became desiccated. Movements away from sites withovermature forage and into sites with new growth enable deer to maintainquality in their diet (Klein, 1965). In Caycuse however, the percent use ofMontane habitat increased between spring and early summer, well beforedesiccation would have occurred within Submontane habitats. Black-taileddeer summer forage species attain their maximum nutritive values in June,July and August (Rochelle 1980). The increased use of Montane habitatstherefore, was not in response to lower forage quality within Submontanehabitats. Instead, deer use of Montane habitat was minimal when theabundance of highly nutritious forage within it was low.33The distribution of Montane variants over the landscape is primarily afunction of elevation and aspect. It is surprising therefore, that gains inelevation and/or the use of easf and north aspects were not detectedcoincidentally with the increase In the percent use of Montane habitat.Failure to detect changes in elevation may have been a function of samplesize and/or the size of the elevational movements. The 1991 summer ranges(MCP) of 10 female study animals averaged only 55 ha (Table 3.1). Thus,movements into Montane habitats usually required only small elevationalgains.34Chapter Three: Changes in Habitat Use by FemaleBlack-Tailed Deer During the Fawning SeasonIntroductionHerding cervids such as caribou and elk often move to specific,1well-defined calving habitats (Altmann 1952; de Vos 1960). The natal habitats ofthe more solitary members of the deer family such as moose, white-taileddeer (Odocoileus virginianus) and black-tailed deer are more difficult tocharacterize (Miller 1974; Stringham 1974). Parturient deer should seekhabitats that confer anti-preddtor advantages during this period ofvulnerability. Livezey (1991) and Mackie (1970) found that black-tailed deerand mule deer respectively, used the same areas during fawning each year.They did not, however, identify favored habitat attributes in such areas.To prevent predators from detecting their newborn fawns, black-taileddeer employ the ‘hider strategy’. When very young, ‘hiders’ have little scent,are cryptically colored, and crouch when threatened (Lent 1974; Geist 1981).During their first weeks of life hiders do not follow their mothers. Instead theyremain hidden, often for many hours, while their dams forage or bed in otherareas. Red deer dams infrequently visit their offspring to nurse them and tostimulate them to defecate by licking their anal region (Clutton-Brock et al.1982). Feces and urine are ingested by the dam to remove evidence thatwould betray the presence of the fawns. Geist (1981) noted that the hiderstrategy often includes ingesting the placenta and birth fluid soakedsubstrate, early removal of young from birth sites, conspecific segregationduring fawning and movements where possible, into escape terrain.Neonate hiders have been found to use habitats with dense groundcover (Johnson 1951, Altmann 1952, Huegel et al. 1986). Although forests35provide screening cover, they often lack the dense ground cover found inmore open sites. Johnson (1951) found that new-born Rocky Mountain elkcalves were likely to be found in sagebrush (Atemisia spp.) habitats and thattheir use of forested habitats decreased as the distance from the forest edgeincreased. Calving sites were associated with the dense ground cover foundin earlier successional stages, not beneath coniferous canopies. Huegel etal. (1986) found shrub and sapling densities to be higher at or around white-tailed deer fawn bedsites and Robinette et al. (1977) found that areas withdenser cover were used by mule deer for fawning. On Vancouver IslandGates (1968) found that “shrubs and young deciduous trees dominate thelow stratum of vegetation from the third year through to at least the fifteenthye&’. If black-tailed deer prefer to fawn in areas with dense ground cover,then their use of clearcut habitats in the fawning season should increase.In 1991, all clearcut habitat polygons used by collared deer, exceptone, were between three and 16 years of age and supported dense groundcover (herbaceous, shrub and seedling percent cover 2 m from theground). The adjacent second- and old-growth stands supported sparserground cover (pers. obs.). In examining the use of clearcut habitats duringthe fawning period versus other summer periods, I proposed a null and analternate hypothesis.Ho The percent use of clearcuts by maternal black-tailed deer is notdifferent during the fawning season than in the remainder of thesummer.Hi The percent use of clearcuts by maternal black-tailed deer is higherduring the fawning season than in the remainder of the summer.36During the fawning period, adult females appeared to move outsidetheir typical area of use with greater frequency. Often only a short time wasspent outside the typical area of use before the doe returned. Theseobservations led me to postulate that black-tailed deer give birth outside oftheir core areas of use to avoid contaminating their preferred sites withscents associated with parturition. After a few days, they lead their fawnsback into their core area and away from scents that might attract predators.If these postulates are true, then we should expect to measure an increase inhome range size during the fawning period. In examining this relationship, Iproposed a null and an alternate hypothesis.Ho Home range size of maternal black-tailed deer is not differentduring the fawning period than in other summer periods.H2 Home range size of maternal black-tailed deer is larger during thefawning period than in other summer periods.I also suspected that maternal females would reduce their area of usein the period immediately after fawning because of their tendency to remainnear their sedentary fawns. I again proposed a null and an alternatehypothesis.Ho Home range size of maternal black-tailed deer is not differentduring the period immediately after fawning than in other summerperiods.37H3 Home range size of maternal black-tailed deer is smaller during theperiod immediately after fawning than in other summer periods.Method and MaterialsThe summer of 1991 was divided into the following 29-day periods: thepre-natal period (PrNP), May 8 to June 5; the fawning period (FP), June 6 toJuly 4; and three post-natal periods, (PoNP1) July 5 to August 2, (PoNP2)August 3 to September 1, and (P0NP3) September 2 to September 30.Because the average number of locations per deer was lower in the PoNP3than in the FP, an adjusted P0NP3 was also analyzed. To produce the‘PoN P3-adjusted’ data set, locations from early October were added in theorder of their chronology to the PoNP3 data set. The locations were added(or subtracted) until the number of locations for each deer was the same inthe P0NP3-adjusted as in the FP. In the case of 15181 and 12801, locationswere removed. The number of locations per period and per deer aresummarized on Table 3.1.During the summers of 1988, 1989, 1990 and 1991, an affiliatedresearch project captured a totdl of 246 neonate fawns in the CaycuseStudy Area and in the South Fork of the Nanaimo River. The Nanaimo RiverStudy Area is 30 km north of the Caycuse Study Area. The earliest and latestfawn capture dates within any summer were June 6th and July 4th,respectively. These dates were used to define the 29 day range of thefawning period (FP). Like white-tailed deer fawns (Heugel et al. 1985), blacktailed deer fawns, up to approximately 10 days old, could be captured by aperson on foot. The FP therefore, extends further into the summer than itwould have if it had been based on the time of parturition. The median for38parturition. The median for the FP fell between the 19th and 20th of June.Thomas (1970) estimated the mean date of birth for a central VancouverIsland watershed to be 4.5 days edrlier, June 14th.The fawning period for this study was designed to produce informationon natal sites and habitat selection. Because black-tailed deer neonatesare not immediately mobile (Miller 1965) and because I could not be sure oflocating the study animals at the exact time when they gave birth, the lag inthe Caycuse FP was appropriate for this analysis.As in Chapter Two, forest age determined stand class. Forestedhabitats were old-growth or at least 16 years post stocking in 1990. Siteslogged and restocked more recently were classified as clearcuts.The G-test (Sokal and RohIf 1981) was used to test for interactionsbetween the percent use of the stand class variable and the establishedtime periods.The second section of this chapter examines home range size duringthe time periods. The minimum convex polygon technique (Mohr, 1947) wasused to estimate home range size for each study animal in all time periods. Aone-tailed paired difference test (Sokal and Rohlf 1981) was used tocompare the home range sizes of different time periods. Type one errorprobabilities (p-values) less than 0.05 were considered to be ‘strong evidenceagainst the null hypothesis’ (HO). Probabilities between 0.05 and 0.10 wereconsidered to be ‘moderate evidence against the null hypothesis’.Probabilities greater than 0.10 were considered to be ‘no evidence againstthe null hypothesis’.Anecdotal information depicting the fawning movements of somestudy animals is also provided. The data for this section were collected in thesummers of 1989, 1990 and 1991. Many of the locations used for the39‘Fawning Site Selection’ section were not separated in time by a 24 hourminimum. Many of the locations were separated by only a few hours whileothers were separated by a few days. In this section, the post-natal homerange was not subdivided into three post-natal home ranges. The 95% homeranges (Hartigan 1987) were calculated by program Home Range(Ackerman et al. 1989). Pre- and post-natal home ranges were separated intime by the ‘fawning movements’. Fawning movements were those thatoccurred between June 6th to July 4th and that were known, or weresuspected, to be movements to, or from, the natal site.ResultsDuring the summer of 1991, ten adult (>1.5 years old) females weremonitored through five time periods producing a total of 563 locations.All of the study animals except 12801 and 17581 were observedinteracting with fawns that appeared to be theirs. Animals 12801 and 17581may have fawned, but because they occupied home ranges with heavyvisual screening, their reproductive status was not determined.Habitat UsePercent use of clearcuts did not increase during the FP. Pooledlocation data (563 locations) revealed that 67.9% and 65.6% of locationswere within clearcuts during the FP and during the remainder of the summer(PrNP, PoNP1, PoNP2 and PoNP3 combined) respectively. The G-testprovided no evidence against the null hypothesis (G=0.20, p=0.66).Fawning Site SelectionExact fawning sites were determined in only two instances. In June of1989 and 1990 study animal 18181 was located once every three hours for athree-day monitoring period. Her natal sites were found in both years. They40were within 70 m of one another and both were beyond the animal’s typicalarea of use. The 1989 and 1990 natal sites were 250 m and 600 mrespectively outside of the 95% pre-natal home ranges (Figs. 3.1 and 3.2).Similarly, the 1989 and 1990 natal sites were 28 m inside and 35 mrespectively outside their 95% post-natal home ranges. The ‘fawningmovements’ shown on Figs. 3.1 and 3.2 were from all locations taken duringthe three-day monitoring periods.In 1991, 13201 made several movements beyond the northwesternperiphery of her 95% home range (Fig. 3.3). Although the natal site wasnever detected, the fawn was first observed on the northern periphery of the95% post-natal home range. The ‘fawning movements’ shown on Fig. 3.3were developed from all locations collected within the FP.In other instances, large erratic movements were made during the FP.In 1991, animals 17991 and 12801 were found 1.6 and 3.5 km respectivelyoutside of their 95% pre- and post-natal summer ranges (Figures 3.4 and 3.5).Since both movements were single events involving only one peripherallocation each, the fawning movements are the departures from, and returnsto, the 95 % home ranges.540860n395000413958005407500< Pre-natal 95% MCP Fawning movements • Birth location• Post-natal locations1:7,000Post-natal 95% MCP* Pre-natal locationsFig. 3.1. The 1989 fawning movements of study animal 18181.Two locations that were out side the typical home range area that occurredin July are not shown. Their UTM coordinates are (393320, 5406590) and (394620, 5408850).., •• •- • ••• •• ••• ••.- •• •..F •• ••-•• • •••SSSt•_42...I\.\.\ I• . \ ••• .... . ‘•.• I •.. .. ..1 •\\*•.•I•\• . \ .._••• • •1.. . .•.—•).•—--- •_,-. • —.-‘ I•_• — )•_•/•‘ —•—•—.,•‘“• Brthbcation•- Pre natal 95% MCP Fawning movements< .I• Post-natal locahons! ! Post-natal 95% MCP* Pre-natal locations1:7,000396000•407600•••I••I•• ••• •I••••540B60(I394800Fig. 3.2. The 1990 fawning movements of study animal 18181.435405300.3922005406500390400< Pie-natal 95% MCP Fawning• First lawn obseivatloni. -. • Post-natal locations1:10,000 L.. - .! Post-natal 95% MCP * Pie-natal locationsFig. 3.3. The 1990 fawning movements of study animal 13201.4454108003904• 8-.—..• f • •1••‘i a?. ft• II ;-I.!.-.•*38930C’_______________________________5408400I J Pre-natal 95% MCP Fawning movements FHSt fawn observationi-.• Post-natal locationsI • P t- ta195°/ MCPLU2,000 L. —.0* Pre-natal locationsFig. 3.4. The 1991 fawning movements of study animal 17991.45‘392400.5410700--.-.- .‘,..•‘ •..4•_,I4.•* *3I00n5406300Pre-natal 95% MCP Fawning movements • Post-natal locations1 Post-natal 95% MCP * Pre-natal locations1:22,000Fig. 3.5. The 1991 fawning movements of study animal 12801.46Home Range SizeHome range sizes (100%) for each study animal within all time periodsare summarized in Table 3.1. The size of each deers total summer range(occupied from May 8 to September 30) is also given. The mean homerange size for each period is also presented graphically in Fig. 3.6.Table3.1.Homerangesize(hectcires) bytimeperiod.DeerTotalnPrNPiiFPnPoNPIitPoNP2itPoNP3itPoNP3itIDSummeradjustedHomeRange12801140.88556.081267.28911.931215.781220.601015.8191320177.505721.641125.531318.171216.871222.31951.46131389132.32547.541214.64105.851114.28119.27109.27101518173.54514.741216.11616.531216.521134.851010.9361678134.925517.161112.341216.101113.821217.33917.46121758119.36573.08137.14124.801110.61126.5296.52121770128.425912.651212.53128.44127.751210.761110.76121791230.335911.731212.06136.371212.151213.281013.28131799186.585716.911149.82125.471210.471213,221015.11121818128.025912.211217.191313.251210.25123.89104.4113mean55.1956.311.3711.823.4611.210.6911.712.8511.815.139.815.5011.2SE12.231.916.171.620.972.814.20==48302520015x I.F50 I I I IPrNP FP PoNP1 PoNP2 PoNP3Time PeriodFigure 3.6. Mean summer home range size and standard error for 10 femaledeer In five periods during the summer of 1991.The distribution of the home range size data set (PrNP, FP, PoNP1,PoNP2 and P0NP3 combined) was not normal (Sha,piro-Wilk normality test,W=0.71, p0.001 [SAS 1988]). A natural log (In x) transformation improved thevalue of the W-statistic to 0.97 (p=0.32).Home range sizes during the FP were compared to those of the othertime periods with a paired difference test. The test results of the transformed(In x) data are summarized in Table 3.2.49Table 3.2. Paired difference test comparison of home range size duringthe fawning period versus all other periods.Comparison FP-PrNP FP-PoNP1 FP-P0NP2 FP-P0NP3 FP-PoNP3adjustedMean Difference 12.1 12.8 10.6 8.3 8.0in Hectarest-value 2.54 2.71 1.89 1.53 1.70Probability 0.016 0.012 0.046 0.080 0.06 1(p-value)Home range sTze was greater in the FP than in any other period. Whenthe FP was compared to the PrNP, the P0NP1 and the P0NP2 the t-testprovided strong evidence against the null hypothesis. When the FP wascompared to the PoNP3 and the P0NP3-adjusted the f-test providedmoderate evidence against the null hypothesis.A migration in the FP away from an early-spring range might explainthe large home ranges in this period. The mean distance between the PrNPand the PoNP1 home range centers however, was determined to be only162 m (SE=30). This small shift was less than the 190 m mean difference(SE=30) between the home range centers of the PoNP1 and PoNP3. Shifts inhome range location therefore, were not larger in the early portion of thesummer than in the late portion and they can not be used to explain thelarger home ranges during the FP.Home range sizes in the immediate post-fawning period (PoNP1) werecompared to those of other time periods with a paired difference test. Theresults are summarized in Table 3.3.50Table 3.3. Paired difference test comparison of home range size duringthe immediate post-fawning period versus all other summer periods.Comparison PrNP-P0NP 1 FP-PoNP 1 P0NP2-PoNP 1 PoNP3-PoNP1Mean Difference 0.68 12.8 2.2 4.4in Hectarest value 0.088 2.71 1.95 1.58Probability 0.466 0.012 0.042 0.075(p-value)Home range size was smaller in the PoNP1 than in all other time periodsexcept the PrNP. When the PoNP1 was compared to the FP and the P0NP2the t-test provided strong evidence against the null hypothesis. When thePoNP1 was compared to the PoNP3 the t-test provided moderate evidenceagainst the null hypothesis.DiscussionHabitat UseThe hider strategy predicts that mother and offspring are together foronly brief periods. A fawning-related increase in the use of any habitat typetherefore, would be difficult to detect because the relationship between thesites occupied by the dams and those of their offspring is weak.It is probable however, that a relationship between clearcuts andfawning habitats did not exist. Forest canopy and shrub layer heterogeneitycould have allowed the parturint study animals to meet their coverrequirements within forested habitats. Although ground-cover was usually51less dense in the forested habitats, many pockets of dense ground coverwere present along creeks, slides, wet areas, and in other areas of low standdensity. Depending on the site conditions, shrubs, such as red huckleberry,devil’s club (Oplopanox horridus (Smith) Miq.) and salal (Gauliheria shallon Pursh),often thrive within second- and old-growth forests.Area of UseBecause the locations of fawning sites were rarely determined, theanecdotal evidence presented serves only to illustrate the development ofthe alternate hypothesis (H2); ‘home range size of maternal black-tailed deeris larger during the fawning period than in other summer periods’.Average home range size was larger in the fawning period than in allother periods. Increases in home range size during the FP were caused byincreases in the magnitude and/or rate of movements to the peripheries ofthe home ranges. Similar peripheral movements of parturient black-taileddeer and mule deerwere found by Livezey (1991) and Riley and Dood (1984)respectively. Livezey reported that “three deer (of 16) fawned in areas thatwere at least 150 m (range 150-450) higher in elevation than the maximumelevation of their ranges during the rest of the year”. Riley and Dood statedthat “no relationship was evident between the point of capture and a fawnssummer home range”. They found that 37% of the fawns were neverreobserved in their area of capture and that an additional 22% werecaptured on the periphery of their summer range. My findings, and those ofthe others, support the idea that sites on the periphery of the dam’s homerange are used for fawning. This behavior conforms to the ‘hider strategy’model. Hiders are known to ingest the placenta and to remove their youngfrom the birth site to prevent predators from using olfactory cues to detecttheir young. If black-tailed deer were to give birth within their core areas of52use, they would be forced to remdve their fawns from the habitats they mostprefer. By giving birth on the periphery of their home ranges, female black-tailed deer avoid contaminating favorite areas and are able to move backinto them immediately after parturition.It could be argued that core areas are avoided during fawningbecause they are already laden with the scent of deer. This is not a likelyexplanation for two reasons. First, on Vancouver Island, black-tailed deer areubiquitous and their densities are usually high. By moving outside their coreareas to give birth, it would be difficult not to enter another deers core area.The known birth sites of animal 18181 were within an area that was heavilyused by animals 17912 and 177O1. Thus the dam’s movements did notremove her from areas with deer scent. Second, the study animals oftencontinued to frequent their core areas during the fawning period and theyusually brought their fawns into their core areas a few days after theirperipheral movement(s). It is doubtful therefore, that scent dissipation withinthe core area would have occurred before the return of the dam andoffspring.Average home range size was lower in the immediate post fawningperiod (PoNP1) than in all other periods except the pre-natal period (PrNP).Miller (1974) also found that “maternal black-tailed deer reduce the sizes oftheir home ranges during the post fawning period”. This behavior is alsoconsistent with the hider strateg’. During the period immediately afterfawning, the mother must remain close to her fawns which are sedentary.Large distances between foraging sites and the fawn’s hiding place wouldincrease the dam’s travel costs and risk of predation. By minimizing thisdistance, energy savings may be transferred to the fawns or used by thefemale. A reduction in travel would also allow the dam to remain motionless53and vigilant for greater periods and thereby reduce predation risks.Chapter Two suggested that in the earlier portions of the summer,habitat use was more concentrated in Submontane habitats where greenup occurred first. It is predictable therefore, that home range size should alsobe small during the pre-natal period (PrNP).54Chapter Four: Summary and ConclusionsChanges in Habitat Use In Relation to Forage QualityThe habitat use patterns of 13 radio collared black-tailed deer wereinvestigated in the summers of 1990 (June 1-October 31) and 1991 (April 1-October 31). Because the location gathering in 1990 starled late, these datawere not statistically tested. They were, however, used to substantiate the1991 findings. The summers were subdivided into time periods and thepercent use of various habitat types were compared. The use of seral stage,biogeoclimatic ecosystem varianls, elevation and aspect within the timeperiods was examined.The data sets of the nine study animals that had clearcut habitatswithin their summer ranges were pooled. Their use of clearcuts was highest inthe same period that Rochelle (1980) found fireweed to receive its highestuse. Percent use of clearcuts was 67.9% in April/May/June, 78.2% inJuly/August and 69.2% in September/October. Given that fireweed is anearly successional species and that it is the black-tailed deer’s mostimporlant summer forage species (Rochelle 1980), the relationship is notsurprising. I suggest that the availability of fireweed was not higher in theJuly/August period than in the other periods. Increased clearcut usetherefore, was probably in response to the nutritional values (digestibility andcaloric) of fireweed which peak in August.The use of Montane variants (BEC) was higher in the Later portions ofthe summer. The pooled data set for the eight deer that had Montanehabitats within their summer range revealed that the percent use ofMontane was 14.5% in April/May, 22.6% in June/July and 25.8%August/September. The use of Montane habitats therefore, was lowest when55the forage species within Submontane were more phenologically advancedand of greater nutritional value. As the summer progressed and as foragephenology within the Montane habitats advanced, the use of Montaneincreased.Although the distribution of Montane variants over the landscape isprimarily a function of elevation and aspect, changes in the use of the lattertwo habitat variables were not detected. With respect to elevation, the lackof a significant difference may have been an arlifact of sample size.Changes in Habitat use During the Fawning SeasonDuring the summer of 1991, ten adult females were monitored throughfive periods producing a total of 563 locations. The summer was divided intothe following 29-day periods: the pre-natal period (PrNP), May 8 to June 5;the fawning period (Fr), June 6 to July 4; and three post-natal periods,(PoNP1) July 5 to August 2, (PoNP2) August 3 to September 1, and (PoNP3)September 2 to September 30.Unlike other researchers who have found relationships between natalsites and early seral stages, the use of clearcut habitats did not changeduring the fawning period. Forest canopy and shrub layer heterogeneitycould have allowed the parturiént study animals to meet their coverrequirements within forested habitats. Many pockets of dense ground coverwere present in areas of low stand density and many species of shrubs, suchas red huckleberry, devil’s club and salal, often thrive in the second and oldgrowth forests of coastal British Columbia.Home range size was greater in the fawning period than in any otherperiod. Increases in home range size during the fawning period were causedby increases in the magnitude and/or rate of movements to the peripheriesof the home ranges. This finding supports the idea that sites on the periphery56of the dam’s home range are used for fawning and is consistent with the‘hider strategy’ model. Hiders are known to remove their young from thebirth site to prevent predators from using olfactory cues to detect theiryoung. If black-tailed deer were to give birth within their core areas of use,they would be forced to remove their fawns from the habitats they mostprefer.Home range size was lower in the immediate post fawning period(P0NP1) than in all other periods except the pre-natal period. This behavior isalso consistent with the hider strategy. During the period immediately afterfawning, the mother must remain close to her fawns which are sedentary.Large distances between foraging sites and the fawn’s hiding place wouldincrease the dam’s travel costs and risk of predation.In the early portion of the summer, habitat use was moreconcentrated in the Submontane where green-up occurs first. The homerange sizes in the pre-natal period therefore, should also be small.Recommendations and Implicatiofls1) In the logged watersheds of coastal British Columbia fireweed andother early successional species öre important summer forage species forblack-tailed deer. To ensure that black-tailed deer have access to highquality forage throughout the summer, shifts in habitat use that occur withinthe summer must be recognized by forest managers. A proportion of earlysuccessional habitats within both the Submontane and Montane variantsshould be maintained throughout the rotation of the forest.2) My findings, and those of others, suggest that black-tailed deer preferto fawn on the periphery of, or outside of, their typical home ranges.Researchers should consider this when interpreting dispersal results that arebased on neonate capture and adult return locations only. Such results57could be biased in favor of dispersal if the scale of fawning movements is nottaken into account.58Literature CitedAckerman, B.B., F.A. Leban, M.D. Sdmuel and E.O. Garton. 1989. UserAsmanual for program home range: second edition. For., Wild!, andRange Exp. Stn. Tech. Rep. 15, Contrib. No. 259. Univ. of Idaho,Moscow.Alidredge, J.R. and J.T. Ratti 1986. Comparison of some statisticaltechniques for analysis of resource selection. J. Wild!. Manage.50:157-165.Altmann, M. 1952. Social behavior of elk, Cervus Canadensis Nelsoni, in theJackson Hole Area of Wyoming. Behavior 4: 65-71.Aitken, Y. 1974. Flowering time, climate and genotype. Melbourne UniversityPress, Carlton, Australia. 193 pp.Bartmann, R.M., G.C. White and L}J. Carpenter. 1992. Compensatorymortality in a Colorado mule deer population. Wildl. Monogr. 121.39pp.Bergerud, A.T., H.E. Butler and D.R. Miller. 1984. Antipredator tacticts ofcalving caribou: dispersion in mountains. Can. J. Zool. 62:1566-1575.Bergerud, A.T. and R.E. Page. 1987. Displacement and dispersion ofparturient caribou at calving as antipredator tactics. Can. J. Zool.65:1597-1606.Bunnell, F.L. 1990, Ecology of black-tailed deer. pages 31-63 in Nyberg,J.B., and D.W. Janz eds. Deer and elk habitats in coastal forests ofsouthern British Columbia. B.C. Mm. For, and B.C. Mm Environ., Res. Br.Special series 5. Victoria, B.C. 310 pp.British Columbia Ministry of Environment. 1988. Ecoregions of BritishColumbia, map.British Columbia Ministry of Forests. J 988. Biogeoclimatic Zones of BritishColumbia, map.Cichowski, D.B. 1989. Seasonal movements, habitat use, and winterfeeding ecology of woodland caribou in west-central BritishColumbia. M.Sc. Thesis, Univ. B.C., Vancouver. 114 pp.59Clover, M. 1956 Single-gate deer trap. Calif. Fish and Game 42:199-201.Clutton-Brock, T.H., F.E. Guinness and S.D. Albon. 1982. Red deer: theecology of two sexes. Univ. Chicago Press, Chicago. 378 pp.Connolly, G.E. 1981. Limiting factors and population regulation. Pages245-286 in O.C. Walmo, ed. Mule deer and black-tailed deer ofNorth America. Univ. Nebraska Press, Lincoln. 605 pp.Cook, R.S., M. White, D.C. Trainer and W.C. Glazener. 1971. Mortality ofyoung white-tailed deer fawns in south Texas. J. Wildl. Manage.35:47-56.Dasmann, R.F. 1981. Wildlife Biology, 2nd ed. John Wiley and Sons, NewYork. 212 pp.Dasmann, R.F. and R.D. Taber. 1956. Behavior of black-tailed deer withreference to population ecology. J. Mammal. 37:143-164.Daubenmire, R.F. 1974. Plants and environment: a textbook of plantautecology, 3rd ed. John Wiley and Sons, New York.de Vos, A. 1960. Behavior of barren ground caribou on their calvinggrounds. J. Wildi. Manage. 24: 250-258.Dyrness, C.T. 1973. Early stages of plant succession following logging andburning in the western Cascades of Oregon. Ecology 54:57-69.Edwards, J. 1983. Diet shifts in moose due to predator avoidance.Oecologia 60:185-189.Edwards, R.Y. and R.W. Ritcey. 1956. The migrations of a moose herd.J. Mammal. 37:486-494.Errington, P.L. 1945. Some contributions of a fifteen-year study of thenorthern bobwhite to the knowledge of population phenomena.Ecol. Monogr. 15: 1-34.Festa-Bianchet, M. 1988. Seasonal range selection in bighorn sheep:conflicts between forage quality, and predator avoidance.Oecologia 75: 580-586.60Gates, B.R. 1968. Deer food production in certain seral stages of the coastforest. M.Sc. Thesis, Univ. B.C., Vancouver. 105 pp.Geiger, R. 1961. The climate near the ground. Friedrich Vieweg and Sohn,Brunswick, Germany.Geist, V. 1981. Behavior: adaptive strategies in mule deer. Pages 157-224 inO.C. Walmo, ed. Mule deer and black-tailed deer of North America.Univ. Nebraska Press, Lincoln. 605 pp.Haeussler, S., D. Coates and J. Mather. 1990. Autecology of CommonPlants in British Columbia: A LIterature review. FRDA report 158.For. Can. and B.C. Mm. For., Victoria, B.C. 272 pp.Hanley, T.A. 1984. Habitat patches and their selection by wapiti andblack-tailed deer in a coastal montane coniferous forest. J. Appl.Ecol. 21: 423-436.Harestad, A.S. 1979. Seasonal movements of black-tailed deer on northernVancouver Island. Ph.D. Thesis, Univ. B.C., Vancouver.184 pp.Hatter, I.W. 1988. Effects of wolf predation on recruitment of black-taileddeer on northeastern Vancouver Island. B.C. Mm. Environ., Victoria,B.C., Wildi. Rep. R-23. 82 pp.Hellgren, E.C., M.R. Vaughan and D.F. Stauffer. 1991. Macrohabitatuse by black bears in a southeastern wetland. J. WildI. Manage.55:442-448.Huegel, G.N., R.B. Dahigren and H.L. Gladfelter. 1985. Use of doe behavior tocapture white-tailed deer fawns. Wild I. Soc. Bull. 13: 287-289.Huegel, G.N., R.B. Dahigren and H.L. Gladfelter. 1986. Bedsite selection bywhite-tailed deer fawns Iowa. J. WildI. Manage. 50:474-480.Jackson, R.M., M. White and F.F. Knowlton. 1972. Activity patterns ofyoung white-tailed deer fawns in south Texas. Ecology: 53: 262-270.Johnson, D.E. 1951. Biology of the elk calf, Cervus canadensis nelsoni. J. Wildl.Manage. 15:396-410.61Johnson, D.H. 1980. The comparison of usage and availabilitymeasurements for evaluating resource preference.Ecology. 61: 65-71.Jones, G.W. 1975. Aspects of winter ecology of black-tailed deer(Odecoileus hemionus columbianus Richardson) on norlhern VancouverIsland. M.Sc. Thesis. Univ. B.C. Vancouver. 78 pp.Klein, D.R. 1965. Ecology of deer rahge in Alaska. Ecol. Monogr. 35: 259-284.Klinka, K., R.N. Green, P.J. Courtin and F.C. Nuszdorfer. 1984. Site diagnosis,tree species selection .and slash burning guidelines for theVancouver Forest Region. B.C. Mm. For., Victoria, B.C. Land Manage.Rep. 25. 180 pp.Krebs, J.R. and R.H. McCleery. 1984. Optimization in behavioural ecology.Pages 91-121 in Krebs, J.R. and N.B. Davies, eds. Behavioural ecology:an evolutionary approach, 2nd ed. Blackwell Scientific Publications,Oxford. 493 pp.Lent, P.C. 1974. Mother-infant relationships in ungulates. Pages 14-53 inV. Geist and F. Waither, eds. The Behaviour of ungulates and itsrelationship to management. Vol. 1. nt. Union Conserv. Nat. NewSeries PubI. 24. Morges, Switzerland.Leopold, A.S., T. Riney, R. McCain and L. Tevis. 1951. The Jawbone deerherd. Calif. Div. Fish and Game, Game Bull. No.4, Berkley. 139 pp.leResche, R.E. 1974. Moose migrations in North America. Nat. Can.101:39-415.Lewis, T. 1988. Ecosystems of block 2 (Caycuse) and compartment 41(Doobah) of tree farm licence 46. Unpubl. rept. to FletcherChallenge Canada Ltd., P0 Box 130, Crofton, B.C., VOR 1 RO. 7 pp.Livezey, K.B. 1991. Home range habitat use, disturbance, and mortality ofColumbian black-tailed deer in Mendocino National Forest. Calif.Fish and Game 77: 201-209.62Loft, E.R., J.W. Menke and T.S. Burton. 1984. Seasonal movements andsummer habitats of female black-tailed deer. J. Wildi. Mange.48:1317-1325.Longhurst, W.M., A.S. Leopold and R.F. Dasmann. 1952. A survey ofCalifornia deer herds. Calif. Dep. Fish and Game, Bull. No.6, Berkeley.136 pp.Mackie, R.J. 1970. Range ecology and relations of mule deer, elk andcattle in the Missouri River Breaks, Montana. Wildl. Monogr. 20. 79 pp.McNay, R.S. and D.D. Doyle. 1987. Winter habitat selection by black-taileddeer on Vancouver Island: a job completion report. Research, B.C.Mm. Environment and Parks and B.C. Mm. Forests and Lands. IWIFR-34.Victoria, B.C. 90 pp.Mealey, S.P. 1975. The natural food habits of free ranging grizzly bears inYellowstone National Park. M.Sc. Thesis, Montana State Univ.,Bozeman. 158 pp.Meidinger, D. and J. Pojar. 1991 Ecosystems of British Columbia. B.C. Mm.For., Special Rep. Series 6, Victoria, B.C. 330 pp.Miller, F.L. 1965. Birth of a black-tailed deer. J. Wildl. Manage. 29: 629-631.Miller, F.L. 1968. Observed use of forage and plant communities by black-tailed deer. J. WildI. Manage. 32:142-148.Miller, F.L. 1974. Four types of territoriality observed in a herd of black-tailed deer. Pages 644-660 in V. Geist and F. Walther, eds. TheBehaviour of ungulates and its relationship to management. Vol. 2. Int.Union Conserv. Nat. New SerIes PubI. 24. Morges, Switzerland.Moen, A.R. 1981. The biology and management of wild ruminants, part IV,forage characteristics and range relationships of wild ruminants.Cornerbrook Press. Lansing, New York.Mohr, C.O. 1947. Table of equivalent populations of North American smallmammals. Am. Midl. Nat. 37: 223-249.63Morgan, J, S. McNay and J. VoIler. 1990. Habitat assessment andplanning: understanding the implications of deer movement patternsiterim data report. Unpubl. Rept. to B.C. Mi For.,Res. Br., Victoria B.C. 15 pp.Neu, C.W., C.R. Byers and J.M. Peek. 1974. A technique for the analysis ofutilization-availability data. J. Wildi. Manage. 38: 541-545.Nyberg, J.B., F.L. Bunnell, D.W. Janz and R.M. Ellis 1986. Managing youngforests as black-tailed deer winter ranges. B.C. Mm. For. Land Manage.Rep. No. 37. Victoria, B.C. 49. pp.Ozoga, J.J. and L.J. Verme. 1986. Relation of maternal age to fawn-rearingsuccess in white-tailed deer. J. Wildl. Manage. 50: 480-486.Potvin, F. and J. Huot. 1983. Estimating carrying capacity of a white-taileddeer wintering area in Quebec. J. Wildi. Manage. 47:463-475.Riley, S.J., and A.R. Dood. 1984. Summer movements, home range, habitatuse and behavior of mule deer fawns. J. Wildl. Manage.48:1302-1310.Rochelle, J.A. 1980. Mature forests, litterfall and patterns of forage quality asfactors in the nutrition of black-tailed deer on northern VancouverIsland. Ph. D. Thesis, Univ. of . C., Vancouver. 295 pp.Robinette, W.L.. N.V. Hancock and D.A. Jones. 1977. The Oak Creek muledeer herd in Utah. Utah State Division of Wildlife Resources, Pub. No. 77-15, Salt Lake City. 148 pp.Russell, C.P. 1932. Seasonal migration of mule deer. Ecol. Mono. 2:1-41.Russo, J.P. 1964. The Kiabab North deer herd-its history, problems andmanagement. Arizona Game and Fish Dep., WildI. Bull. No.7, Phoenb.163 pp.SAS. 1988. SAS procedures guide: release 6.03 edition. SAS Inst. Inc. Cary, N.C.Schoen, J.W. and M.D. Kirchhoff. 1985. Seasonal distribution and homerange patterns of Sitka black-tailed deer on Admiralty Island,southeast Alaska. J.Wildl. Manage. 49: 96-103.64Short, H.L. 1981. Nutrition and MetaDollsm. Pages 99-127 in O.C. Walmo, ed.Mule deer and black-tailed deer of North America. Univ. NebraskaPress, Lincoln. 605 pp.Smith, R.H. and A. LeCount. 1979. Some factors affecting survival of desertmule deer fawns. J. Wildl. Mdnage. 43: 657-665.Stringham, S.F. 1974. Mother-infant relations in moose. Nat. Can.101:325-369.Sokal, R.R. and F.J. Rohlf. 1981. Biometry: the principles and practice ofstatistics in biological research, 2nd ed. W.H. Freeman andCompany, New York. 859 pp.Steigers, W.D. and J.T. Flinders. 1980. Mortality and movements of muledeer fawns in Washington. J. Wildl. Manage. 44:381-388.Thomas, D.C., 1970. The ovary, reproduction and productivity of femaleColumbian black-tailed dee. Ph.D. Thesis, Univ. B. C., Vancouver.211 pp.White, G.C. and R.A. Garrot. 1991. Analysis of wildlife radio tracking data.Academic Press Inc., San Diego. 383 pp.