quantitative traits Data from selective harvests underestimate temporal trends in

Human harvests can select against pheno-types favoured by natural selection, and naturalresource managers should evaluate possible arti-ficial selection on wild populations. Because therequired genetic data are extremely difficult togather, however, managers typically rely on har-vested animals to document temporal trends.It is usually unknown whether these data areunbiased. We explore our ability to detecta decline in horn size of bighorn sheep (Oviscanadensis) by comparing harvested males withall males in a population where evolutionarychanges owing to trophy hunting were previouslyreported. Hunting records underestimated thetemporal decline, partly because of an increasingproportion of rams that could not be harvestedbecause their horns were smaller than thethreshold set by hunting regulations. If har-vests are selective, temporal trends measuredfrom harvest records will underestimate themagnitude of changes in wild populations.Keywords: artificial selection; sport hunting;time series; ungulates1. INTRODUCTIONRecently, it has become evident that exploitation canlead to artificial selection [1–4]. As humans oftenprefer sex-age classes or morphological traits associatedwithhighnaturalsurvival,harvestmortalitydiffersfromnatural mortality [5]. Thus, harvests may lead to evol-utionary responses in life histories and morphology[6], outpacing other selective agents [7,8]. There havebeen calls to consider potentially undesirable artificialselection in management and conservation [9,10].A first step to avoid artificial selection is evaluatingwhich practices have undesired consequences. Asdetailed, individual monitoring is rarely available,attempts to quantify artificial selection in wild popu-lations usually rely on morphological, life history anddemographic data collected from harvested animals[6,11]. Harvest data, however, will not reflect popu-lation values if harvest is selective and varies inintensity over time, as may occur with trophy huntingand size-selective fisheries. For example, the agedistribution of shot red grouse (Lagopus lagopus scoti-cus) was biased by harvest intensity [12]. Similarbiases may affect temporal trends in morphologicaltraits estimated from harvest data. If harvest prob-ability is affected by regulations for minimum size,gear selectivity [13] or by cultural preferences, theaverage size of harvested animals should be greaterthan the population average. These biases have beenacknowledged [14], but it remains unknown howthey affect our ability to detect temporal trends inphenotypic traits.To test whether data from selective harvests can beused to detect temporal trends in phenotype, we ana-lysed nearly four decades of detailed individualmonitoring of bighorn sheep (Ovis canadensis)inapopulation where unlimited trophy hunting favouredrams with slow-growing horns [2,15]. We comparedharvested animals with the whole population. In mostof the province of Alberta, harvest of bighorn rams isbased on minimum horn curl (figure 1). This manage-ment strategy protects sub-adults but leads to theharvest of rams with rapidly growing horns aged 4–6years [5,16], before they obtain the high reproductivesuccess associated with large horns at ages 7–11[17,18]. Similar regulations for both bighorn and thin-horn (Ovis dalli) sheep, combining curl restrictionsand unlimited resident permits, exist in most ofCanada and in Alaska. The ‘legal’ definition of a har-vestable ram is thus comparable to antler pointrestrictions often used for cervids [19].2. MATERIAL AND METHODS

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.