"Forestry, Faculty of"@en . "DSpace"@en . "UBCV"@en . "Thomson, R. N. (Robert Newton)"@en . "2008-09-05T18:30:19Z"@en . "1991"@en . "Master of Science - MSc"@en . "University of British Columbia"@en . "Relations among bull age, sex ratio and reproductive parameters in British Columbia moose populations were investigated. Specifically, the effects that bull age-selective harvest regulations versus non-selective harvest regulations might have on pregnancy rate, conception timing, and recruitment were compared. Existing harvest, inventory, and reproductive data from four management subregions were analyzed. There was no evidence that reductions in the prime and senior-aged bull social classes in a population resulted in reduced pregnancy rates or later conception timing. No evidence was found that greatly skewed sex ratios in favour of cows resulted in reduced pregnancy rates. No relation was found between bull/100 cow ratios and calf/100 cows in the winter inventory."@en . "https://circle.library.ubc.ca/rest/handle/2429/1692?expand=metadata"@en . "2018473 bytes"@en . "application/pdf"@en . "AN ANALYSIS OF THE INFLUENCE OF MALE AGE AND SEX RATIOON REPRODUCTION IN BRITISH COLUMBIA MOOSE (Alces alces L.)POPULATIONSByROBERT N. THOMSONB.Sc., The University of Victoria, 1980A THESIS SUBMITTED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFMASTER OF SCIENCEinTHE FACULTY OF GRADUATE STUDIES(Department of Forestry)We accept this thesis as conformingto the required standardTHE UNIVERSITY OF BRITISH COLUMBIADecember 1991\u00C2\u00A9 Robert N. ThomsonIn 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.(Signature)Department of^r(7) ,e -&---,S'The University of British ColumbiaVancouver, CanadaDate ^J)e---(2: /^/ (-1/DE-6 (2/88)ABSTRACTRelations among bull age, sex ratio and reproductiveparameters in British Columbia moose populations wereinvestigated. Specifically, the effects that bull age-selective harvest regulations versus non-selective harvestregulations might have on pregnancy rate, conceptiontiming, and recruitment were compared. Existing harvest,inventory, and reproductive data from four managementsubregions were analyzed. There was no evidence thatreductions in the prime and senior-aged bull social classesin a population resulted in reduced pregnancy rates orlater conception timing. No evidence was found thatgreatly skewed sex ratios in favour of cows resulted inreduced pregnancy rates. No relation was found betweenbull/100 cow ratios and calf/100 cows in the winterinventory.iiiTABLE OF CONTENTSPageAbstract^ iiTable of Contents^ iiiList of Figures ivAcknowledgmentsIntroductionStatus and Management^ 1Male Social Behaviour 3Social Stress Theory 6Mating Systems^ 7Hypothesis 10Study Areas^ 15MethodsAge^ 17Sex Ratio and Recruitment^ 19Pregnancy Rate and Conception Timing^ 20Statistical Tests^ 22Results ^ 24Discussion 34Management Recommendations ^ 40Literature Cited ^ 42ivLIST OF FIGURESPage1. The proposed relationships between somedemographic characteristics in a moosepopulation^ 122. A comparison of the proportion of prime andsenior-aged bulls using inventory data for theOmineca subregion and tooth return data for theother subregions^ 253. Pregnancy rates for cows 2 years and older forfour regions of British Columbia, 1985-1990^ 264. Mean conception dates for cows 2 years and olderfor four regions of British Columbia, 1985-1990...275. A comparison of pregnancy rates with ratios fromthe previous winter and post-rut bull/100 cowratios^ 296. Calf/100 cow ratios compared to pregnancy ratesin the previous year^ 307. A comparison of calf/100 cow ratios withbull/100 cow ratios from the previous year in theThompson-Nicola subregion, 1964-1991^ 328. A comparison of calf/100 cow ratios withbull/100 cow ratios from the previous year in theOmineca subregion, 1973-1989^ 33ACKNOWLEDGMENTSI would like to thank my principle advisor Dr. F.L.Bunnell for his support and direction during thisproject. I would also like to thank the other members ofmy advisory committee, Dr. D. Shackleton, Dr. D. Eastmanand Dr. T. Sullivan for their insightful comments.Financial support was provided by the Faculty of Forestryand Canadian Forest Products Ltd. through graduatefellowships.Much of this study depended on information collectedby others and I would like to express my appreciation tothe members of the Fish and Wildlife Branch who providedme with their raw data and were willing to discuss ideas,in particular; Ken Child, Doug Jury, Herb Langin, RandyWright, Rob Woods, and Sean Barry. Many thanks also go toBrian Churchill for his understanding and efforts toprovide me with the necessary time to finish this work.1INTRODUCTIONStatus and Management - The moose is the largest cervid inthe world and is found globally throughout the subborealand boreal forests. Three subspecies are recognized inBritish Columbia, A. a. shirasi, A. a. andersoni and A. a.gigas (Cowan and Guiget 1965). Ecologically, BritishColumbia is very diverse and moose are found over a widerange of habitat types and climatic regimes. Spalding(1990) provides a thorough discussion of the earlycolonization and historical distribution of moose in B.C.In British Columbia, the moose is one of the mosthighly valued wildlife species for both consumptive andnon-consumptive activities. The provincial populationestimate is approximately 173,000, with the majority(135,000) in the Ministry of Environment's northern regionsof Skeena and Omineca-Peace (I. Hatter pers. commun.). TheMinistry of Environment's draft provincial moose statementdivides the province into nine moose management areas. Infour of the management areas, the population trend from1985 to 1989 is listed as stable, in four areas it islisted as stable to declining and in one area it is listedas stable to increasing. The moose management statementdiscusses three options for population objectives, rangingfrom accepting some decline in the provincial numbers toincreasing moose numbers by 20% over a 20 year period2(Hatter et al. 1990).For hunters, moose in B.C. are managed for bothtrophies and meat, with 90% of hunters interestedexclusively in meat (Hatter et al. 1990). Historically inNorth America, bull moose have been managed with openseasons and regulations in the form of bag limits andseason length. In B.C., harvest management of moose hasconsisted of seasons for either antlered or antlerlessmoose or both. Varying season lengths was another commontechnique. In 1974, however, limited entry hunting wasintroduced to provide managers more control of the numbersharvested and control of the geographic distribution of theharvest. The limited entry regulation requires thathunters be drawn from a pool of applicants for the hunt.In 1981, age-selective hunting was introduced, in whichspecific age-classes of bulls were subjected to differinghunting pressures. The primary reason for introducing anage-selective harvest strategy for bulls was a growingconcern that altered age-structures might adversely affectthe productivity and health of ungulate populations throughdisruption of social processes. The concept of consideringsocial aspects of ungulate populations in managementstrategies was introduced into North America by Bubenik(1971, 1972). Moose managers in several provinces havesince implemented age-selective harvest regulations3(Stewart 1978; Macgregor and Child 1981; Euler 1983). Therationale for the age-selective harvest strategies isdiscussed in the remainder of this chapter.Male Social Behaviour - Compared to other ungulate species,little is known about the behaviour of moose and much ofthat is anectodal. Very little quantitative analysis hasbeen done.Being primarily a forest dweller and a 'concentrateselective' herbivore requiring high quality forage (Hofmann1985), moose are less social than many open terraindwelling ungulates such as bighorn sheep (Ovis canadensisShaw) (Geist 1971), and Scottish red deer (Cervus elaphusL.) (Geist 1963), and they are the least gregarious of theNorth American cervids (de Vos et al. 1967; Peek et al.1974). Both sexes are primarily solitary during much ofthe year, however, bulls appear to be more gregarious thancows (Houston 1968; Peek et al. 1974). In a comparison ofmoose populations in Alaska, Montana and Minnesota, Peek etal. (1974) found that bull aggregations varied in size overthe year and that seasonal timing of peak aggregationsvaried geographically. On the Kenai Peninsula, Alaska, thelargest aggregations were seen in July and August beforethe rut. In the Montana and Minnesota populations, thelargest aggregations were after the rut, from mid-October4to December. In all three areas, the largest aggregationsgenerally occurred in open habitat such as alpine tundra,willow bottoms and cutover areas. Observations on moose inWells Gray Park, British Columbia, found small aggregationsof bulls in winter, spring, and during the rut (Geist1963). Many of these observations occurred in the opensub-alpine terrain and in burned-over brush-meadow valleybottoms. In relatively open habitat in the Jackson Holearea of Wyoming, Houston (1968) observed small aggregationsof bulls but found little variation in their occurrencethroughout the year.The solitary nature of the moose may be anevolutionary adaptation to transient or small habitatpatches (Houston 1968; Geist 1971). Forage supplies inriparian areas or habitat patches isolated by deep snowsmay be quickly decimated by large aggregations. Largepatches of good habitat are often created aftercatastrophic events such as fire and larger groupings ofmoose can be found in such areas, however the use of thesehabitats is temporally limited by forest succession (Geist1971). Habitat structure itself will likely affect groupformation through forage distribution patterns andavailability of openings. Peek et al. (1974) suggestedthat the solitary behaviour and dispersal of moose may bea predator-avoidance strategy. The large size and5aggressive behaviour of the moose (Mech 1970; Geist 1971)may also reduce the need to aggregate with others as ananti-predator strategy.During the breeding season, a bull may mate withseveral cows. The mating system has been referred to as alimited or conditional polygamy (Houston 1968; Markgren1969) or serial monogamy (Bubenik 1985), because throughoutmuch of the range of the moose, the bull spends timetending an individual breeding cow during her estrus periodrather than tending several in a group.Bulls become more aggressive with the onset of thebreeding season (Lent 1974) and in open-terrain such as thetundra, they establish dominance hierarchies (Peek et al.1974). Lincoln et al. (1985) define a dominance hierarchyas \"a social order of dominance sustained by aggressive orother behavioural patterns\". Clutton-Brock et al. (1982)stated that competition in ungulates may be most intensewhen the potential exists for individual males to gainexclusive access to many females for the purpose ofbreeding. In these situations, females are a limitedresource and are very valuable for male fitness. Seriousbattles between bulls are rare (Markgren 1969) andgenerally occur only between large bulls of equal status orsize. Sparring matches between young bulls of equal size6tend to be less intense (Geist 1963; Peek et al. 1986).Young bulls with little chance of breeding would not beexpected to invest much energy in fighting.Observations indicate that the presence of higher-ranking bulls in breeding groups prevent lower-rankingbulls from participating in breeding activity, althoughthey are almost certainly capable physiologically (Altmann1959; Houston 1968; Bubenik 1971). Sparring matchesbetween adult and juvenile red deer bulls have beenobserved late in the breeding season and probably occurbecause the juveniles are testing and strengthening theirown abilities (Clutton-Brock et al. 1982). McCullough(1979) stated that for young males to move up in rank in ahierarchial system, they must continuously interact andtest themselves with individuals of both lower and higherrank.Social Stress Theory - Bubenik (1971) felt that the terms'adult' and 'juvenile' were too vague for describing socialclasses of animals and suggested that the terms 'kids','pre-teens', 'teens', 'primes' and 'seniors' were moredescriptive. He argued that the presence of prime (6-10year old) bulls maintains order in a population during therutting period by suppressing the breeding activities ofyounger, less experienced bulls. Hunting practices that7target prime bulls could artificially skew the populationage-structure to younger bulls and also significantly alterthe sex ratio in favour of cows. A surplus of cows in thepopulation would increase the breeding opportunities forteen bulls. Bubenik argued that having a majority ofyounger bulls in the bull population would acceleratesexual maturation, possibly inhibiting skeletal growth.With few prime bulls in a population, the teens would lackexperience in sorting out social ranks making them behaveabnormally towards the primes. Also, they would not havedeveloped the proper courting behaviour towards cows. Theactions of the teens would significantly stress the primebulls and the cows, and this in turn would negativelyimpact the reproductive rate. Reproduction and recruitmentwould be impaired because of:a) an increase in cows being bred after their first estrus,resulting in late-born calves with reduced survival rates,and b) an overall decrease in the percentage of cows bred(Bubenik 1971, 1972; Stringham and Bubenik 1974).Mating Systems - Behavioural studies have found theenvironment to be a major factor in determining matingsystems, more so than phylogenetic heritage (Vehrencamp andBradbury 1984). The lack of visibility in much of theboreal and sub-boreal forests of B.C., together with thepatchiness of food resources, probably limits group8formation. Low densities of moose and lack of groupingwould make it difficult for a dominance hierarchy to becomeestablished in forested habitat. Peek et al. (1986)questioned whether dominance hierarchies would occur inforests to the south. Bubenik (1987) suggested thathabitat density and climate severity cause differencesbetween the mating systems of open-dwelling tundra mooseand moose in forested habitat. Clutton-Brock et al. (1982)also identified habitat and climate as major factors indetermining mating systems in cervids.For a behavioural pattern to persist in a population,it must not reduce the reproductive fitness of theperformer. Behaviour that reduces fitness should disappearthrough natural selection. Studies indicate that, innorthern tundra populations at least, cow moose preferlarge-antlered bulls (Knowles 1983; Bubenik 1987). Ifwell-developed secondary sex traits (i.e., antlers, bodysize) increase mating success in bulls, then this behaviourof the cow would be expected, assuming it is a heritabletrait. The cow should select the mate which is most likelyto pass these characteristics on to their offspring,improving the mating success of the offspring andincreasing the chance that her genes will continue to bepassed on.9Research on captive animals has found the estrusperiod for cow moose to be approximately 24 hours (Schwartz1987) and there is some evidence that the presence of abull may play a role in initiating estrus (Bubenik 1987;Schwartz pers. commun.). It also has been suggested thatcalves born later because breeding in the second or thirdestrus will have a reduced chance of survival in theirfirst winter because of their smaller size (Baker 1975;Crichton 1988; Child and Aitken 1989). Research on captivemoose has found that late-born calves do not experiencecompensatory growth and go into the winter with lower bodyweights than calves conceived in the first estrus (Schwartzpers. commun.). Given the short estrus period and thepotential importance of first estrus breeding for calfsurvival, it should be expected that in situations wherecows may have few bulls to choose from (i.e., in low-density forest-dwelling populations or populationspioneering new habitat) cows will be less selective intheir choice of mates. Courtship behaviour and secondarysexual characteristics, described by Bubenik (1987) to besignificant factors in the breeding system, should be lessimportant under these conditions. Being selective woulddecrease the cows chances of breeding in the first estrusand decrease her overall fitness. The degree to which acow is selective should depend on her knowledge of theavailability of bulls. This knowledge would be gained from10the frequency of interactions with bulls, from vocalizing,rut pits, or visual contact, shortly before and during therut period.Hypothesis Based on this review, it is my thesis that with thewide variety of habitats that moose occupy and theattraction to early seral stages encouraging them to besomewhat nomadic, the breeding behaviour of moose should beflexible enough to compensate for relatively widevariations in population density, sex ratio, and bull agestructure and therefore, reproduction should not besignificantly impaired. This is contrary to Bubenik'spredictions described earlier.Figure 1 illustrates the potential pathways by whichdensity, age structure, and sex ratio may influencereproduction in a moose population. The dotted linesindicate suspected relations while the solid lines indicaterelations which are more certain. Productivity was definedas the number of calves born. Recruitment in the model wasdefined as the number of calves surviving to approximatelyseven months of age (when most population inventory work isdone). For the purpose of data analysis and comparisonlater in the paper, the number of calves per 100 cows wasused as the measure of recruitment.11Density may impact reproduction in ungulates throughchanging nutritional levels, affecting conception timing(McCullough 1979), and pregnancy rates (Blood 1974; Boer1987). Cow age may affect conception timing throughdelayed estrus of younger cows, as has been observed in reddeer (Mitchell and Lincoln 1973) and in moose (Schwartzpers. commun.). Cow age has also been related to pregnancyrates (Pimlott 1959; Markgren 1969; Franzmann and Schwartz1985) and twinning rates (Pimlott 1959; Boer 1987). Mostof this relation is due to lower fecundity in yearlingcows. The impact of bull age on reproduction has not beendocumented for moose. Several studies on elk, however,have suggested that yearling bulls are not as effectivebreeders as branch-antlered bulls and breeding by yearlingbulls may result in later conception dates and reducedpregnancy rates (Hines and Lemos 1979; Prothero et al.1979; Smith 1980). Reduced bull/cow ratios may delayconception timing (Lent 1974) and reduce pregnancy rates(Boer 1987). As previously mentioned, timing of conceptionhas been suggested as a factor affecting calf survival andtherefore may impact on recruitment. Pregnancy rates andtwinning rates will be related to productivity in that theydetermine the maximum potential productivity. Similarly,limits to potential recruitment will be determined byproductivity.12DENSITY^AGE STRUCTURE^SEX RATIOmale femaleVCONCEPTIONTIMINGPREGNANCYRATE PRODUCTIVITYTWINNINGRATERECRUITMENTFigure 1. The proposed relationships between some demographiccharacteristics in a moose population.(dotted lines indicate suspected relationships)13External mortality factors, such as predation,disease, and weather, were omitted for simplification. Itshould be recognized however that these factors will impactquantitatively on the model, particularly betweenproductivity and recruitment. Also, some form of mortalityis implicit in the proposed relation between conceptiontiming and recruitment.Bubenik's predictions of the 'social stress' theorycould not be evaluated without a lengthy field study.Therefore, I identified some logical consequences of thetheory that could potentially be addressed with existingharvest and inventory data. My general hypothesis can berestated in the form of several testable questions:1. Does the reduction of prime and senior-aged bullsresult in reduced pregnancy rates?2. Does the reduction of prime and senior-aged bullsresult in delayed conception dates?3.^Does skewing of the sex ratio in favour of cows resultin reduced pregnancy rates?4.^Do these changes affect recruitment?14Data from several B.C. Ministry of Environmentsubregions were used in this study. The Omineca, Peace,Cariboo, and Thompson-Nicola subregions were chosen becauseof their significant moose harvests and because they havecow harvests providing reproductive tracts. There also isa difference in the harvest policies between thesubregions. The Omineca implemented an age-selective bullstrategy in 1981 while the other subregions have maintainednon-selective bull seasons.15STUDY AREASThe following descriptions of the subregions includedin the study were adapted from the ecoregionclassifications described by Demarchi et al. (1990).The Thompson-Nicola subregion represents the SouthernInterior Ecoprovince. This area is characterized by a warmdry climate. Vegetative cover ranges from extensivegrasslands and open parkland to dense coniferous forests athigher elevations.The Cariboo subregion represents the Central InteriorEcoprovince. Winters are cold, summers are warm,andprecipitation is greater than in the Thompson-Nicolasubregion. The vegetation is intermediate between the drysouthern interior and the cold boreal forests. Grasslandhabitat in the south mixes with open coniferous forests.Extensive wetlands are also present in parts of thesubregion. Denser subboreal forests in the north areprimarily coniferous but deciduous stands are common.The Omineca subregion represents the SubborealInterior Ecoprovince. Winters are cold and arctic airoutbreaks and relatively high snowfalls are common. Thedominant vegetative cover is the dense coniferous forestwith deciduous stands more numerous than in the south.16The Peace subregion represents the Boreal PlainsEcoprovince and is on the eastern side of the RockyMountains. The continental climate is relatively dry withcold winters dominated by arctic air. Extensive deciduousforests and grasslands create an aspen (Populus tremuloidesMichx.) parkland habitat. Dense coniferous forests arealso common throughout the subregion.17METHODSExisting B.C. Wildlife Branch data were used to lookfor evidence of the suspected relations in the model(Figure 1) and to address the stated questions. Lack ofavailable data on population density and on productivity,as it is defined here, precluded any analysis of relationsinvolving these parameters. It was also felt thatmanagement actions would have little impact on cow agestructure as hunters would not be selective on cow-age,therefore, no analyses were done with cow age data. Astwinning rates are likely affected primarily by habitatquality and possibly by cow age (Pimlott 1959; Franzmannand Schwartz 1985), these data were not evaluated.All data were collected by Ministry of Environmentstaff and made available for this study. The types of dataused were: age of bulls in the harvest; age of cows in theharvest; pregnancy rate and dates of conception; and winterpopulation composition data in the form of bull/cow/calfratios.Age - Teeth from harvested cows were collected along withthe reproductive tracts. Cow-age data were available onthe raw data sheets from each of the subregions. Bull agedata were extracted from the Summary Statistics Data Baseon the Ministry of Environment's VAX computer. Each year18teeth from hunter-killed bull moose are voluntarily sent inby hunters for ageing. Age was determined by grinding orsectioning teeth and counting the cementum annuli in theroot (Sergeant and Pimlott 1959).Yearling bulls were excluded from the databases asgraphical analysis of age distributions from the toothreturn database indicated that yearlings wereunderrepresented. This is likely due to hunters being ableto age yearlings in the field and therefore not beingcurious about age. Distributions of the other age classesappeared to fit that of a typical age distribution and itwas therefore assumed that they were representative of thedistribution of the harvest. It was also assumed that theproportion of ages in the tooth return was representativeof those in the actual population. Because the age-selective bull regulations in the Omineca restricted theharvest of older bulls, the tooth return database for thissubregion was biased toward younger bulls, and thereforecould not be compared to other subregions. Comparisonswere done using the social maturity class data from theOmineca winter inventories. Maturity classes weredetermined using a modified version of Oswald (1982).Based on analyses of antlers from bull harvests in theOmineca, bulls were considered to be in the prime categoryat age 5 1/2 (K. Child pers. commun.). In the tooth return19database these animals are recorded as 5 years old.Sex Ratio and Recruitment - Winter inventories wereperformed regularly by Ministry staff in the Omineca, Peaceand Thompson-Nicola subregions. The timing of theinventories varied from early December to late January.The management units surveyed varied from year to year.Generally, line transect surveys were used to obtainclassified count information, except in the Thompson-Nicolasubregion where stratified random block inventories weredone each year.Most of the inventory information was in the form ofclassification at the level of bull, cow, and calf. Bullmoose start to drop their antlers in early Decemberbeginning with the older animals. For this reason, it wasusually not possible to classify bulls by age class. Cowmoose were distinguished from antlerless bulls by thepresence of a white vulval patch on the cows (Mitchell1970) or presence of antler pedicles on the bulls. Ifconditions made identification difficult then an animal wasrecorded as an unclassified adult. Classificationinformation was standardized for comparison by convertingit to two statistics: the number of bulls per 100 cows; andthe number of calves per 100 cows.Surveys in the Omineca subregion were performed early20enough to classify the bulls into maturity classes based onantler size.Pregnancy Rate and Conception Timing - Hunters in each ofthe four subregions were requested to submit reproductivetracts from the limited entry cow hunts held in lateNovember or early December. Submission of the reproductivetracts was compulsory in the Omineca, Peace, and Thompson-Nicola subregions. In the Omineca subregion, thecollection of reproductive tracts began in 1977 and initialsample sizes were small. In the Thompson-Nicola subregion,collections began in 1986 and in the Peace subregion, theybegan in 1988. The Cariboo subregion began collecting in1983, however, submission of reproductive tracts was notcompulsory and samples were small for most years. Onlydata from 1985 and 1986 were used from the Cariboo.Pregnancy was usually determined by the presence of afoetus in the uterus. In some cases where foeti were notpresent due to late conception, pregnancy could also bedetected by the presence of embryonic threads or ablastocyst (Markgren 1969). Detailed analysis of theovaries may also provide an indication of pregnancy(Markgren 1969), however, the Omineca was the onlysubregion to carry out thorough ovarian analyses. Tostandardize the pregnancy data, cows were only considered21to be pregnant if the determination was based onexamination of the uterus. In the Omineca data set, cowsdetermined to be pregnant based on analysis of the ovarieswere recorded as not pregnant.Yearling cows were removed from the data sets becauseof their highly variable pregnancy rates which are likelyrelated to nutrition (Pimlott 1959; Franzmann and Schwartz1985). The large numbers of yearling cows could have asignificant effect on the pregnancy rate of the populationmaking it difficult to determine relationships withpopulation sex ratio and bull age structure.For cases of pregnancy, the conception dates could beestimated from the size of the foetus. During the first 90days of development, foeti can be aged by measuring thecrown-rump length (Markgren 1969). The Peace, Cariboo andThompson-Nicola subregions used the foetus length/agerelationship determined by Markgren (1969), for ageing.The Omineca subregion aged feoti with a modified version ofthe Markgren method, resulting in different ages than theother subregions for given foetal lengths. There alsoappeared to be some differences between the subregionsusing the Markgren method. For these reasons, all foetusages were recalculated and standardized using the Markgrenmethod. Date of conception was determined by subtracting22the foetus age from the date of kill.Statistical Tests Differences between the percentage of primes andseniors in the Omineca bull population and the percentageof 5+ year-olds in the bull populations from the Peace,Cariboo and Thompson-Nicola subregions were tested with theKruskal-Wallis test, a nonparametric analogue of a singleclassification analysis of variance test (Sokal and Rohlf1981).Differences in pregnancy rates between the subregionswere tested for with the Kruskal-Wallis test.Linear regression analysis (Sokal and Rohlf 1981) wasused to test for relations between pregnancy rates andbull/100 cow ratios; calf/100 cow ratios and pregnancyrates; and calf/100 cow ratios and bull/100 cow ratios.Data from individual management units were used to comparepregnancy rates and calf/100 cow ratios or bull/100 cowratios. There were few management units in each regionwhich had both inventory data and reproductive tractsamples, therefore the regions were pooled to look forrelationships. Calf/100 cow ratios were compared againstpregnancy rates from the previous year as this would havebeen when the calves were conceived. Similarly, calf/10023cow ratios were compared against bull/100 cow ratios fromthe previous winter as these should be the closest to thepopulation sex ratio when the calves were conceived.Pregnancy rates were compared against bull/100 cow ratiosfrom both the previous winter and from the winterimmediately following collection of the reproductive tractsto determine which inventory provided the bestrelationship.24RESULTSDoes the reduction of prime and senior-aged bulls result in reduced pregnancy rates? - Figure 2 compares the proportionof prime and senior-aged bulls from the Omineca wintersurveys with the proportion of bulls five years and olderin the harvests of the other subregions. Yearlings werenot included in the datasets. The graph indicates that theOmineca subregion has a greater proportion of prime andsenior-aged bulls than the other subregions.^Thedifference was found to be statistically significant(p < 0.001).Figure 3 shows pregnancy rates, by subregion,calculated for cows two years of age and older. Nosignificant differences were found between the subregions(p > 0.5).Does the reduction of prime and senior-aged bulls result in delayed conception dates? - Figure 4 shows the mean datesof conception for each subregion. Dates were quitenormally distributed in all cases with median values withinone day of the means in all but one case in which themedian was two days from the mean. Standard deviationswere similar in all cases. Confidence limits (95%)calculated about each mean indicated806040200% prime/senior or 5+ males1981^1982^1983^1984^1985Year---- Omineca^--I-- Thompson1986^1987^1988^1989 Peace^o CaribooFigure 2. A comparison of the proportion of prime and senior-aged bullsusing inventory data for the Omineca subregion and tooth returndata for the other subregions.2611^OMINECA' I PEACE THOMPSON CARIBOOi1985 76 77(n=53) (n=69)1986 80 80 83(n.103) (n=54) (n=70)1987 90 83(n=83) (n=86)1988 80 83 77(n=86) (n=193) (n=60)1989 80 80 93(n=86) (n=159) (n=96)1990 76 81 79(n=80) (n=149) (n=65)Figure 3. Pregnancy rates for cows 2 years and older for fourregions of British Columbia, 1985-1990.27H0 OMINECA^PEACE^THOMPSON^CARIBOO1 1Oct.7 Oct .71985(41,5.47) (57,6.36)Oct.9 Oct .7 Oct .81986(82,7.08) (37,2.99) (81,6.61)Oct.5 Oct.101987(79,5.82) (67,4.31)Oct.7 Oct .9 Oct .81988(68,6.27) (166,6.08) (43,4.07)Oct.6 Oct .8 Oct .91989(72,5.92) (134,6.08) (83,7.08)Oct.6 Oct .9 Oct .71990(60,4.89) (119,5.78) (51,4.49)Figure 4. Mean conception dates for cows 2 years and older forfour regions of British Columbia,^1985-1990.(with sample size and standard deviation)28that there were no significant differences in conceptiondates between subregions or between years withinsubregions.Does the skewing of the sex ratio in favour of cows result in reduced pregnancy rates? - Figure 5a compares thepregnancy rates in the Omineca, Thompson-Nicola and Peacesubregions with bull/cow ratios from the previous winterinventory. Over the range of bull/100 cow ratios tested,no significant relationship was found (p > 0.05). Figure5b compares the pregnancy rates for the same subregionswith the post-rut sex ratios. No significant relationshipwas found (p > 0.25).Do the changes affect recruitment? - Figure 6 comparescalf/100 cow ratios with the corresponding pregnancy ratesfrom individual management units in the Omineca, Peace andThompson-Nicola subregions. The relationship is notsignificant(p > 0.05).100 - 801_ ^^\u00E2\u0080\u00A2 \u00E2\u0080\u00A26040ra=0.0066n=29 p>0.250 ^2029a. Ratios from the previous winter.pregnancy rate100 ^-^80 ^+ ^60 ^\u00E2\u0080\u00A2 \u00E2\u0080\u00A240 r2=0.1087n=2120 p>0.05 00^20^40^60^80b0/100 cowb. Post-rut ratios.pregnancy rate20^40^60^80^100b0/100 cowOmineca^^ Thompson^+ PeaceFigure 5. A comparison of pregnancy rates with ratios fromthe previous winter and post-rut bull/100 cowratios.rk=0.1450n=19p>0.05E ucalf/100 cows0^20^40^60pregnancy rateOmineca^E Thompson^+ Peace70605040302010080^100Figure 6. Calf/100 cow ratios compared to pregnancy rates in the previous year.31Figure 7 compares calf/100 cow ratios with bull/100cow ratios from the previous year, using winter inventorydata taken from 1964 to 1991 in the Thompson-Nicolasubregion. The data were divided into three periods ofapproximately ten years each.^There is a significantnegative relationship (p < 0.0005).^The graph alsoindicates that recruitment has been greater after 1975 thanbefore. Bull/100 cow ratios have declined since 1975.Figure 8 compares calf/100 cow ratios with bull/100cow ratios from the previous winter for the Ominecasubregion. The data were divided by the year when the age-selective regulations were implemented. No significantrelationship was found (p > 0.25).calf/ 100 cows100806040200 r2=0.3242^ n=44^ p<0.0005o+ ++3^^^0 0+ + \u00E2\u0080\u00A2^+ .0 00^. \u00E2\u0080\u00A2 \u00E2\u0080\u00A20 ....a0^20^40^60^80^100^120^140bull/1 00 cows\u00E2\u0080\u00A2^1964-74^^^1975-84^+ 1985-91Figure 7. A comparison of calf/100 cow ratios with bull/100 cow ratios fromthe previous year in the Thompson-Nicola subregion, 1964-1991.e=0.0016n=61^p>0.25* +^\u00E2\u0096\u00AA :+^44-H+ . +\u00E2\u0080\u00A2+\u00C2\u00B1 \u00E2\u0080\u00A2++ 4_\u00E2\u0080\u00A2calf/100 cows0^20^40^60^80^100^120^140bull/1 00 cows^pre-1981^+^post-19811 00806040200Figure 8. A comparison of calf/100 cow ratios with bull/100 cow ratios fromthe previous year in the Omineca subregion, 1973-1989.34DISCUSSIONAge Structure - A difference in the age structure of thebull moose population was found between the Ominecasubregion and the three other subregions (Figure 2). TheOmineca subregion, with the age-selective policy, had asignificantly greater percentage of prime and senior-agedbulls (p < 0.001) based on the comparison of the inventoryclassification data with the tooth return data from theother subregions. There was no evidence that the lowerpercentage of prime and senior-aged bulls in the Peace,Cariboo and Thompson-Nicola subregions resulted in reducedpregnancy rates. No significant differences in pregnancyrates (p > 0.5) were found between the subregions (Figure3).There was also no evidence that the difference in agestructure between the subregions had any affect onconception timing. Mean conception dates for allsubregions and over all years sampled ranged over a 5 dayperiod in early October (Figure 4). There was nosignificant difference in the mean dates based on the 95%confidence limits about the means.The observations support the hypothesis that areduction in the proportion of prime and senior-aged bullsin the population, over the range evaluated here, has no35significant effect on reproduction. The social classes ofbulls performing the majority of the breeding cannot bedetermined from the existing data. However, it may bereasonable to assume that in the subregions with lowproportions of prime and senior-aged bulls, the youngerbulls are significantly involved in the breeding activity.Despite these findings, age structure of populationsfor all ungulates may be a management concern for geneticrather than reproductive reasons. Using modellingtechniques and data from moose and white-tailed deer(Odocoileus virginianus Zimmermann), Ryman et al. (1981)found that different hunting regimes could greatly alterthe generation interval in a population and this had severeeffects on heterozygosity. Prothero et al. (1979) statedthat the major contribution that mature males make to apopulation is a genetic one and they suggested that removalof the selective pressures for breeding may have longtermconsequences for the fitness of the population.Sex Ratio - No evidence was found for any relation betweensex ratio and pregnancy rate or recruitment. In the casesof pregnancy rate versus bull/100 cow and calf/100 cow(Figures 5a, 5b, and 6), it is known that the relation mustpass through the origin. It appears that the effect of sexratio is only evident at very low bull/cow levels, i.e.,36less than 20 bulls/100 cows (Figures 5a and 5b). Only afew data points are available below this level (for thePeace and Thompson-Nicola subregions) and no reduction inpregnancy rate was obvious although sample size was small.No significant relation (p > 0.05) was found betweenpregnancy rate and recruitment (Figure 6). One likelyreason for lack of a relationship would be the limitedsample size that was available. Given the large size ofmost management units, it was only appropriate to comparepregnancy rates and calf/cow ratios from the same unit. Ineach subregion, there were very few management units whichhad a reasonable reproductive sample and a winter inventoryfor the appropriate years. For this reason data from thesubregions were pooled and the relation illustrated inFigure 6 is based on 19 samples.The apparent negative relation between calf/100 cowand bull/100 cow in the Thompson-Nicola subregion (Figure7) is likely due to factors outside of the population. Theapparent trend is caused by low calf/100 cow ratios in the1964-74 period of inventory. During much of this period,inventories were performed on management areas which werelarger and had different boundaries than the presentmanagement units. Areas with higher snowfall and possiblygreater wolf predation levels were included in the earlyperiod but not in the later periods (D. Low and R. Ritcey37pers. commun).Sex ratios were reported to be close to 50/50 andoften in favour of bulls in Newfoundland and Ontario duringthe late 1940's and early 1950's (Peterson 1955). Based onthe literature, Bubenik (1972) stated that equal sex ratiosare normal for naturally regulated moose populations.Bubenik (1987) stated that in taiga moose populations itwas important to maintain a minimum 50/50 sex ratio toavoid a decrease in reproduction. Both Boer (1987) andCrete et al. (1981) believed that 40% bulls in the adultwintering population (67 bulls/100 cows) was the thresholdbelow which reproduction may be reduced. I found noevidence that either of these levels are critical;pregnancy rates in B.C. moose populations with adult sexratios much less than 40 bulls/100 cows are comparable tothose reported in natural populations with higher sexratios.Analysis of the available data in B.C. produced noevidence that prime-aged bulls or high bull/cow ratios arenecessary for maintaining relatively high pregnancy ratesin moose. This analysis makes the assumptions that the agestructure in the hunter sample for the Peace, Cariboo andThompson-Nicola were representative of the actualpopulation with the exception that yearling moose were38likely underrepresented as hunters would be less likely tosubmit a yearling tooth for ageing.The validity of the analysis of sex ratio andrecruitment information depends upon the winter inventorydata being an accurate representation of the populations.Any bias in the inventory data is likely to be greater inthe bull/cow data as it is possible that one or the othersex is selecting habitat which would reduce its visibility.If groups of one sex were being missed strictly because thetwo sexes were not associating together, it would beexpected that within and across inventories, this source ofbias would be removed. Calves should remain with theirmothers through the first winter and therefore, the chancesthat calf/cow ratios are inaccurate should be less than forbull/cow ratios. Another possible source of error incalf/cow ratios would be if there was a difference inhabitat selection between cows with calves and cows withoutcalves.An alternative explanation to the lack of relationshipbetween pregnancy rate (Figure 6) or bull/cow ratios(Figures 7 and 8) with recruitment is that externalmortality factors were impacting calf numbers before thewinter inventories were done. As most inventories weredone in December or early January, it is unlikely that39winter weather conditions would have had time to reducecalf numbers. Disease, predation and poor weatherconditions shortly after parturition may be reducing calfnumbers. Research in Alaska has found brown bear (Ursusarctos L.) predation to be the major mortality source formoose calves (Ballard et al. 1991). Wolf (Canis lupus L.)predation was identified as a major mortality source formoose in Quebec (Messier and Crete 1985), and the maindirect mortality agent on moose calves on Isle Royale (Mechet al. 1987).The possible effect of population density onreproductive parameters must be considered wheninterpreting these results. It was not possible toevaluate the influence of density on the variables testeddue to lack of data. Density-dependent factors may haveinfluenced the results, particularly when looking at therelation between calf/100 cow and bull/100 cow ratios(Figures 7 and 8). The majority of the high bull/cowratios (i.e., > 50 bulls/100 cows) were observed before the1980's and before reproductive tracts were being collected.It is possible that moose densities were higher when manyof the earlier inventories were performed and this may havereduced recruitment.40MANAGEMENT RECOMMENDATIONS1. The Wildlife Branch should continue to investigate therelations between bull age, sex ratio and reproduction toget a clearer picture of their significance in B.C. moosepopulations. This could be facilitated with larger samplesand by ensuring that winter inventories are performed inthe same management units where the reproductive samplesare collected.2. Trials should be performed to test whether inventorytechniques are accurate. This could possibly be done byfollowing transect inventories with a total block count.This should be attempted on several areas which may havedifferent compositions of open and forested habitat.3. Winter inventories should be performed and recorded ona management unit basis rather than a geographic area sothat the information can be more easily compared with otherindices such as reproductive data and hunter harvest data.This problem would be lessened if management unitboundaries were delineated along majorgeographical/ecological features as much as possible sothat it is possible to differentiate between populations ofanimals. Also, georeferencing all harvest and inventorydata on a computer database would greatly enhance theability to explore relations in the data.414. The Wildlife Branch is currently reviewing itscommittment to collecting and ageing tooth samples from theannual ungulate harvests.^It is recommended that thisprogram be maintained as much as possible, particularly formoose. The data set may prove valuable for future work ongenetic and reproductive changes in populations. The agedata also will likely be valuable and necessary forestimating moose population sizes through modellingexercises.5. Foetus ageing techniques should be standardized acrossthe province. The accuracy of the ageing method is lessimportant than consistancy as the primary interest is inidentifying trends or changes.6. Reproductive data should be stored in a computerdatabase similar to the harvest and inventory databasescurrently used by the Wildlife Branch.42LITERATURE CITEDAltmann, M. 1959. Group dynamics in Wyoming Moose duringthe rutting season. J. Mammal. 40:420-424.Baker, R.A. 1975. Biological implications of a bulls-onlymoose hunting regulation of Ontario. Proc. N. Am.Moose Conf. Workshop. 11:1-10.Ballard, W.B., J.S. Whitman, and D.J. Reed. 1991.Population dynamics of moose in South-Central Alaska.Wildl. Monogr. 114:1-49.Blood,^D.A.^1974. Variation in reproduction andproductivity of an enclosed herd of moose (Alcesalces). Proc. Int. Cong. Game Biol. 11:59-66.Boer, A.H. 1987. Reproductive productivity of moose in NewBrunswick. Alces 23:49-60.Bubenik, A.B. 1971. Social well-being as a special aspectof animal sociology. Intern. Conf. on the Behay. ofUngulates and Its Relation to Management. Calgary.25pp.Bubenik, A.B. 1972. North American moose management inlight of European experiences. Proc. N. Am. MooseConf. Workshop. 8:276-295.Bubenik, A.B. 1985. Reproductive strategies in cervids.Royal Soc. New Zeal. Bull. 22:367-373.Bubenik, A.B. 1987. Behaviour of moose (Alces alces sp) ofNorth America. Swed. Wildl. Res. (Suppl.) 1:333-365.Child, K.N. and D.A. Aitken. 1989. Selective harvests,hunters, and moose in central British Columbia. Alces25:81-97.Clutton-Brock, T.H., F.E. Guinness and S.D. Albon. 1982.Red Deer: the Behaviour and Ecology of Two Sexes.Univ. Chicago Press, Chicago. 378 pp.Cowan, I. McT., and C.J. Guiget. 1965. The Mammals ofBritish Columbia. B.C. Prov. Museum. Handbk No. 11.414pp.Crete, M., R.J. Taylor and P.A. Jordan. 1981. Optimizationof moose harvest in southwestern Ontario. J.Wildl.Manage. 45:598-611.43Crichton, V. 1988. In utero productivity of moose inManitoba. A1ces 24:143-149.Demarchi, D.A., R.D. Demarchi, R.D. Marsh, A.P. Harcombeand E.C. Lea. 1990. The Environment. pp.55-144 In:R.W. Campbell, N.K. Dawe, I. McTaggart-Cowan, J.M.Cooper, D.W. Kaiser and M.C.E. McNall. The Birds ofBritish Columbia. Royal B.C. Museum. 514 pp.de Vos, A., P. Brokx and V. Geist. 1967. A review of socialbehavior of the North American cervids during thereproductive period. Am. Midl. Nat. 77:390-417.Euler, D. 1983. Selective harvest, compensatory mortalityand moose in Ontario. Alces 19:148-160.Franzmann, A.W., and C.C. Schwartz. 1985. Moose twinningrates: a possible population condition assessment. J.Wildl. Manage. 49:394-396.Geist, V. 1963. On the behaviour of the North AmericanMoose (Alces alces andersoni Peterson 1950) in BritishColumbia. Behaviour. 20:377-416.Geist, V. 1971. Mountain Sheep: A Study in Behavior andEvolution. Univ. Chicago Press, Chicago. 383 pp.Hatter, I., K. Child and H. Langin. 1990. Provincial MooseStatement for British Columbia 1990-1995. B.C. Min.Environ. 54pp. Draft.Hines, W.W., and J.C. Lemos. 1979. Reproductive Performanceby Two Age-Classes of Male Roosevelt Elk inSouthwestern Oregon. Oregon Dept. Fish and Wildl.,Wildl. Res. Rep. No.8. 54pp.Hofmann, R.R. 1985. Digestive physiology of the deer -Their morphophysiological specialisation andadaptation. Royal Soc. New Zeal. Bull. 22:393-407.Houston, D.B. 1968. The Shiras moose in Jackson Hole,Wyoming. Nat'l Park Service, U.S. Dept. Int. Tech.Bull. No.l. 110pp.Knowles, W.C. 1983. An ethological analysis of the use ofantlers as social organs by rutting bull moose (A1cesalces gigas Miller). M.S. thesis. Univ. Alaska,Fairbanks. 93pp.Lent, P.C. 1974. A review of rutting behavior in moose.Nat. Can. 101:307-323.44Lincoln, R.J., G.A. Boxshall and P.F. Clark. 1985. ADictionary of Ecology, Evolution and Systematics.Cambridge Univ. Press, Cambridge. 298pp.Markgren, G. 1969. Reproduction of moose in Sweden.Viltrevy 6:127-285.McCullough, D.R. 1979. The George Reserve Deer Herd:Population Ecology of a K-Selected Species. Univ.Michigan Press, Ann Arbor. 271 pp.Macgregor, W.G. and K.N. Child. 1981. Changes in moosemanagement in British Columbia. Alces 17:64-77.Mech, L.D. 1970. The Wolf. Natural History Press, GardenCity. 384 pp.Mech, L.D., R.E. McRoberts, R.O. Peterson and R.E. Page.1987. Relationship of deer and moose populations toprevious winters' snow. J. Anim. Ecol. 56:615-627.Messier, F. and M. Crete. 1985. Moose-wolf dynamics and thenatural regulation of moose populations. Oecologia65:503-512.Mitchell, B. and G.A. Lincoln. 1973. Conception dates inrelation to age and condition in two populations ofRed deer in Scotland. J. Zool., Lond. 171:141-152.Mitchell, H.B. 1970. Rapid aerial sexing of antlerlessmoose in British Columbia. J. Wildl. Manage. 34:645-646.Oswald, K. 1982. A Manual for Aerial Observations of Moose.Ontario Min. Nat. Res. 103pp.Peek, J.M., R.E. LeResche and D.R. Stevens. 1974. Dynamicsof moose aggregations in Alaska, Minnesota, andMontana. J. Mammal. 55:126-137.Peek, J.M., V. Van Ballenberghe and D.G. Miquelle. 1986.Intensity of interactions between rutting bull moosein central Alaska. J. Mammal. 67:423-426.Peterson, R.L. 1955. North American Moose. Univ. TorontoPress. 280 pp.Pimlott, D.H. 1959. Reproduction and productivity ofNewfoundland moose. J. Wildl. Manage. 23:381-401.45Prothero, W.L., J.J. Spillett and D.F. Balph. 1979. Ruttingbehavior of yearling and mature bull elk: someimplications for open bull hunting. pp.160-165 In: M.Boyce and L. Hayden-Wing (eds.) North American Elk:Ecology, Behavior and Management. Univ. Wyoming.Ryman, N., R. Baccus, C. Reuterwall and M. H. Smith. 1981.Effective population size, generation interval, andpotential loss of genetic variability in game speciesunder different hunting regimes. Oikos 36:257-266.Schwartz, C.C. 1987. Evaluation and Testing of TechniquesFor Moose Management. Alaska Dep. Fish and Game, Fed.Aid in Wildl. Restor. Ann. Rep., Proj. W-22-6. 36pp.Sergeant, D.E. and D.H. Pimlott. 1959. Age determination inmoose from sectioned incisor teeth. J. Wildl. Manage.23:315-321.Smith, J.L. 1980. Reproductive rates, age structure andmanagement of Roosevelt elk in Washington's OlympicMountains. pp.67-111 In: W. Macgregor (ed.) Proc.Western States Elk Workshop.Sokal, R.R. and F.J. Rohlf. 1981. Biometry. W.H. Freeman.859 pp.Spalding, D.J. 1990. The Early History of Moose (Aloesaloes): Distribution and Relative Abundance in BritishColumbia. Royal B. C. Museum, Contributions to Nat.Sci. No. 11. 12pp.Stewart, R.R. 1978. Introduction of sex and age specifichunting licenses for the moose harvest inSaskatchewan. Proc. N. Am. Moose Conf. Workshop.14:194-208.Stringham, S.F. and A.B. Bubenik. 1974. Physical conditionand survival rate of Chamois (Rupicapra rupicapra L.)as a function of maturity-sex class ratios in thepopulation: implications for ungulate harvest plans.Proc. I.U.G.B. Congress. XI:147-153.Vehrencamp, S.L. and J.W. Bradbury. 1984. Mating systemsand ecology. pp. 251-278. In: Krebs, J.R. and N.B.Davies (eds.). Behavioural Ecology: An EvolutionaryApproach. Blackwell Scientific, Oxford."@en . "Thesis/Dissertation"@en . "1992-05"@en . "10.14288/1.0075118"@en . "eng"@en . "Forestry"@en . "Vancouver : University of British Columbia Library"@en . "University of British Columbia"@en . "For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use."@en . "Graduate"@en . "An analysis of the influence of male age and sex ratio on reproduction in British Columbia moose (Alces alces L.) populations"@en . "Text"@en . "http://hdl.handle.net/2429/1692"@en .