A slow life in hell or a fast life in heaven: demographic analyses of contrasting roe deer populations

A slow life in hell or a fast life in heaven: demographic analyses of contrasting roe deer populations
of 11
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  See discussions, stats, and author profiles for this publication at: A slow life in hell or a fast life in heaven:Demographic analyses of contrasting roe deerpopulations  ARTICLE   in  JOURNAL OF ANIMAL ECOLOGY · JUNE 2009 Impact Factor: 4.5 · DOI: 10.1111/j.1365-2656.2009.01523.x · Source: PubMed CITATIONS 61 READS 41 7 AUTHORS , INCLUDING:Erlend B. NilsenNorwegian Institute for Nature Research 73   PUBLICATIONS   1,197   CITATIONS   SEE PROFILE Jean-Michel GaillardClaude Bernard University Lyon 1 344   PUBLICATIONS   12,906   CITATIONS   SEE PROFILE John OddenNorwegian Institute for Nature Research 98   PUBLICATIONS   2,002   CITATIONS   SEE PROFILE John D C LinnellNorwegian Institute for Nature Research 269   PUBLICATIONS   5,368   CITATIONS   SEE PROFILE All in-text references underlined in blue are linked to publications on ResearchGate,letting you access and read them immediately.Available from: John D C LinnellRetrieved on: 04 February 2016   Journal of Animal Ecology  2009, 78  , 585–594doi: 10.1111/j.1365-2656.2009.01523.x  © 2009 The Authors. Journal compilation © 2009 British Ecological Society  Blackwell PublishingLtd  A slow life in hell or a fast life in heaven: demographic analyses of contrasting roe deer populations  Erlend B. Nilsen   1  *, Jean-Michel Gaillard   2  , Reidar Andersen   3,4  , John Odden   4  , Daniel Delorme   5  , Guy van Laere   5  and John D. C. Linnell   4   1  Faculty of Forestry and Wildlife Management, Hedmark University College, Evenstad, 2480 Koppang, Norway; 2  Biométrie et Biologie Evolutive, Unité Mixte de Recherche 5588, Bât. 711, Université Lyon 1, 43 boulevard du 11 Novembre 1918, F-69622 Villeurbanne cedex, France; 3  Museum of Natural History and Archaeology, Norwegian University of Technology and Science, NO-7491 Trondheim, Norway; 4  Norwegian Institute for Nature Research, NO-7485 Trondheim, Norway; and 5  Office National de la Chasse et de la Faune Sauvage, Centre National d’Etudes et de Recherche Appliquée Cervidés-Sangliers, 1 Place Exelmans, 55000 Bar-le-Duc, France  Summary  1.  Environmental conditions shape population growth through their impact on demographicparameters. While knowledge has accumulated concerning the effects of population density andclimatic conditions, a topical question now concerns how predation and harvest influencedemographic parameters and population growth (  λ   ).  2.  We performed a comparative demographic analysis based on projection matrix models forfemale roe deer. Population-specific matrices were parameterized based on longitudinal data fromfive intensively monitored populations in Norway and France, spanning a large variability inenvironmental characteristics such as densities of large predators, hunter harvest and seasonality.  3.  As expected for a large iteroparous vertebrate, temporal variation was invariably higher inrecruitment than in adult survival, and the elasticity of adult survival was consistently higher thanthat of recruitment. However, the relative difference in elasticity of λ   to recruitment and adultsurvival varied strongly across populations, and was closely correlated with adult survival.  4.  Different traits accounted for most of the variance in λ   in different ecological settings. Adultsurvival generally contributed more in populations with low mean adult survival and low meangrowth rate during the study period. Hunters and predators (Eurasian lynx and red foxes) occurredin two of our study populations and contributed substantially to the variance in λ   , accounting fora total of 35% and 70% in the two populations respectively.  5.  Across populations, we did not find any evidence that roe deer increased their reproductiveoutput when faced with harsh conditions, resulting in some populations having negative growth rates.  6.  Generation time, a measure of the speed of the life-history cycle, increased from less than 4 yearsin the most productive population (‘roe deer heaven’) to more than 6 years in declining populationsfacing predation from lynx, red fox and hunters (‘roe deer hell’), and was tightly and inverselycorrelated with λ   . Such a deceleration of the life cycle in declining populations might be a generalfeature in large herbivores.  7.  Our results shows that the plethora of environmental conditions faced by populations of largeherbivores also induce high intraspecific variation in their ranking along the ‘fast–slow’ continuumof life-history tactics.  Key-words:  Capreolus capreolus  , demographic variation, Eurasian lynx, harvest, life history  Introduction  To increase our mechanistic understanding of populationfluctuations and the evolution of life histories, an importantfirst step is to identify the demographic parameters thatare most important for explaining variation in populationgrowth rate. In most taxonomic groups, species’ life historiesare distributed along a ‘fast–slow’ continuum (Stearns 1983;Gaillard  et al   . 1989; Harvey, Read & Promislow 1989; Sæther& Bakke 2000; Bielby  et al   . 2007) even after accounting for  *Corresponding author. E-mail:   586  Nilsen et al.  © 2009 The Authors. Journal compilation © 2009 British Ecological Society, Journal of Animal Ecology  , 78  , 585–594  differences in body mass. Slow-living species reproducerelatively late in their life span, have low fecundity and longlife expectancy, and thereby have long generation times(defined after Leslie 1966 as the weighted mean age of mothersin a population, and typically more than 2 years in slow-livingspecies; Lebreton & Clobert 1991). As the potential impacton λ   of a given change in a demographic parameter is closelyrelated to generation time (Gaillard  et al   . 2005), such markeddifferences in generation time among species might suggestthat the functional dependence of λ   to changes in adult survivalvs. recruitment is species specific. For instance, a given changein adult survival always has a stronger impact on λ   in long-livedspecies than that of any other trait (Gaillard  et al   . 2000). Sincelife histories partly reflect adaptations to particular environ-ments (Stearns 1992) and temporal variability usually reducesthe long-term population growth rate (Tuljapurkar & Orzack1980), traits that have a large impact on λ   should be bufferedagainst temporal variation, as reported for large mammalianherbivores (Gaillard & Yoccoz 2003), birds (Sæther & Bakke2000) and plants (Pfister 1998). As a result, the observedvariation in λ   of large herbivores is often attributed to inter-annual variation in recruitment when covariation amongdemographic rates is not accounted for, despite the strongerfunctional dependence of λ   to variation in adult survival(Gaillard et al   . 2000).However, recent analyses have shown that different demo-graphic parameters might account for the observed variationin λ   in different populations of the same species (see e.g. Coulson,Gaillard & Festa-Bianchet 2005; Morrison & Hik 2007), andeven in different habitats within a population (Ezard  et al   .2008). Environmental conditions shape λ   by their effects ondemographic parameters, and empirical evidence shows thatage-specific demographic parameters and their covariationare differently affected by environmental variation (Coulson  et al   . 2001). While stressful climatic and nutritional conditionsusually affect prime-aged individuals less than pre-prime andpost-prime individuals (Gaillard  et al   . 2000; Coulson  et al   .2001), free-ranging populations might be affected by a plethoraof environmental factors acting on different demographicparameters. For instance, while some coursing predators suchas wolves (  Canis lupus  L.) are highly selective and mostlyfocus on young and old individuals within the prey population(Delgiudice  et al   . 2006), ambush predators such as cougars(  Puma concolor L.) are less selective, and kill a large proportionof prime-aged females (Festa-Bianchet  et al   . 2006). In roedeer (  Capreolus capreolus  L.), fawns are generally not exposedto significant lynx (  Lynx lynx L.) predation during their firstmonths of life (Panzacchi  et al   . 2008), while older age classesare usually selected over other prey species when present(Jedrzejewski  et al   . 1993; Schmidt 2008). Consequently, whenlynx predation varies over time, survival rates at ages otherthan juveniles are likely to show substantial temporal variation.On the other hand, red fox (  Vulpes vulpes Say 1823) predationon roe deer is mainly concentrated within the first two monthsof life (Panzacchi et al   . 2008). Thus, fluctuations in red foxabundance also cause variation in juvenile survival (Kjellander& Nordstrom 2003). Finally, in human-dominated landscapes,large herbivores are often subject to intensive harvesting(Gordon, Hester & Festa-Bianchet 2004). With harvestingregimes ranging from strongly juvenile biased to trophy harvestof prime age animals (Milner, Nilsen & Andreassen 2007), suchhuman-induced mortality is also likely to create very differentpatterns in the relationship between λ   and demographicparameters than those found in non-hunted populations.Here we examine the population dynamic responses of amedium-sized ungulate species, the roe deer, to differentenvironmental conditions. More precisely, we use matrixperturbation analysis (Caswell 2001) to investigate the sourcesof variation in λ   in five contrasting European roe deer popu-lations in a total of eight time periods. These populationsinhabit widely contrasting environments, spanning gradientsin seasonality, predation pressure, harvest levels and popula-tion densities. The purpose of this study was (i) to assess thefunctional dependence of λ   on variation in demographicparameters across a wide range of environmental conditions,(ii) to quantify the contribution of changes in demographicparameters to variation in λ   , and (iii) to quantify the contribu-tion of predation and harvesting to temporal variation in λ   .This latter is particularly important as many of the best-studiedpopulations of temperate ungulates such as red deer (  Cervuselaphus  L., Clutton-Brock, Guinness & Albon 1982), Soaysheep (  Ovis aries  L., Clutton-Brock & Pemberton 2004), roe deer(Gaillard  et al   . 2003) and pronghorn (  Antilocapra americana  Ord 1815, Byers 1997) were not subject to either predation onadults or harvesting, while Festa-Bianchet et al   . (2006) havedemonstrated how dramatic the effect of predation can bewhen it strikes a previously predator-free population.  Methods  STUDY    AREAS  We used demographic data from five populations studied in France(Trois Fontaines and Chizé) and Norway (Storfosna, Østerdalen andAkershus/Østfold). These populations span a wide range of envi-ronmental conditions (Table 1). The sampling designs that variedamong areas are described elsewhere (see e.g. Gaillard  et al   . 1993;Nilsen, Linnell & Andersen 2004).The study area (7·8 km   2  ) on Storfosna (10·8 km   2  ) is located on asmall island 2 km off the coast of central Norway (63  °  4  ′ N, 09  °  3  ′ E).The island consists of a fine-scaled mosaic of heather-dominatedmoorland, abandoned meadows, cultivated pasture, and mixedconiferous/deciduous woodland. During the study period, from1991 to 1994, the population density increased from 10·1 deer km   −   2  in spring 1991 to 34·5 deer km   −   2  in May 1994 (Andersen & Linnell2000). Demographic parameters were estimated based on radiocol-lared roe deer (  n  = 128), mostly marked as fawns (  n  = 94, see Nilsen  et al   . 2004 for further details).The study area in Østerdalen is located in south-eastern Norway(61  °  15  ′ N, 11  °  30  ′ W). The topography consists of parallel river valleysrunning from north to south at about 200–500 m above sea level withhills ranging from 600 to 900 m above sea level. Agricultural landand human settlements are scattered along most valley bottoms. Thevegetation is predominantly boreal coniferous forest (Scots pine  Pinus sylvestris  , Norwegian spruce Picea abies  and birch Betula sp.),with 72% of the study area covered with forest. Based on hunter   Roe deer demography and population dynamics  587  © 2009 The Authors. Journal compilation © 2009 British Ecological Society, Journal of Animal Ecology  , 78  , 585–594records and counts made at supplementary feeding stations, theroe deer density is believed to be extremely low (< 1 per km   2  ).Demographic parameters were estimated based on radiocollared roedeer (  n  = 62) often marked as fawns (  n  = 23, see Panzacchi et al   .2008 for further details).The Akershus/Østfold study area, situated 100 km south of Østerdalen, is also dominated by boreal forest, but includes patchesof deciduous forest represented mainly by birch. The landscapeis human modified, with the forest fragmented by cultivated landand water bodies, and the altitude is not higher than 300 m abovesea level. Roe deer density estimated from hunting records suggestthat they occur at moderate densities, and that the density declinedslightly during the study period from 2001 to 2005. Demographicparameters were estimated based on radiocollared roe deer (  n  = 116),mostly marked as fawns (  n  = 44, see Panzacchi et al   . 2008 andRatikainen  et al   . 2007 for further details).The study area at Chizé (26·6 km   2  ) in western France (46  °  05  ′  N,0  °  25  ′ W) has an oceanic climate with Mediterranean influences,with mild winters and warm often dry summers. This fenced reservemanaged by the Office National de la Chasse et de la Faune Sauvage(ONCFS) consists of a forest dominated by oak Quercus sp. andbeech Fagus sylvatica  . Overall, the forest is not highly productivebecause of poor soil quality and of frequent summer droughts. Since1977, capture–mark–recapture estimates of population size anddemographic parameters are available (see Gaillard et al   . 2003 forfurther details). High variation in population size (controlled byyearly removals) and climate among years allowed us to definethree periods of contrasting roe deer performance (see Table 1).Demographic parameters were estimated based on individuallymarked (both ear-tags and numbered collar) and known-age roedeer (  n  = 599) often marked as newborns (  n  = 322, see Gaillard et al   .2003 for further details).The study area at Trois Fontaines (13·6 km   2  ) in eastern France(48  °  43  ′ N, 54  °  10  ′ E) has a continental climate, with quite harshwinters and warm but moist summers. This fenced area managed bythe ONCFS consists of a forest dominated by oak and beech. Theforest is quite homogeneous at a large spatial scale (> 100 ha) buthighly heterogeneous at small spatial scales (< 10 ha). TroisFontaines is a rich and productive forest (Pettorelli et al   . 2006) dueto high-quality soils and generally high rainfall in spring. Variationin population size (controlled by yearly removals) allowed us todefine two periods of contrasting roe deer performance (see Table 1).Demographic parameters were estimated based on individuallymarked (both ear-tags and numbered collar) and known-age roedeer (  n  = 997) mostly marked as newborns (  n  = 907, see Gaillard  et al   . 2003 for further details).As well as differing in climate and density, these populationsdiffered with respect to predation. Hunters, Eurasian lynx and redfoxes were predators of roe deer in both Norwegian mainland sites,with both hunters and lynx killing all age classes (Andersen  et al   .2007) and foxes killing fawns during the post-natal period (Panzacchi  et al   . 2008). Although red foxes were present in the two French studysites, there is no evidence of marked predation on fawns in these areas.  ESTIMATION   OF   DEMOGRAPHIC   PARAMETERS  Due to different sampling regimes between study sites, parameterestimation techniques also varied between the study sites. Mostimportantly, data collection in the three Norwegian populations wasbased on radiotelemetry studies, whereas the French populationswere based on individually marked animals potentially recapturedonce every year.      T  a   b   l  e   1 .      T   h  e   fi  v  e  r  o  e   d  e  e  r  p  o  p  u   l  a   t   i  o  n  s  s   t  u   d   i  e   d  a  n   d   t   h  e   i  r  m  a   i  n  e  n  v   i  r  o  n  m  e  n   t  a   l  c   h  a  r  a  c   t  e  r   i  s   t   i  c  s .   L  o  w   d  e  n  s   i   t  y   i  s   d  e   fi  n  e   d  a  s   <   5   d  e  e  r   k  m    –   2    ,  m  e   d   i  u  m   d  e  n  s   i   t  y   i  s   5 –   1   0   d  e  e  r   k  m    –   2    ,  a  n   d   h   i  g   h   d  e  n  s   i   t  y   i  s   >   1   0   d  e  e  r   k  m    –   2    .   A   t   C   h   i  z   é ,  r  o  e   d  e  e  r   h  a   d  a   l  o  w   /  m  e   d   i  u  m  p  e  r   f  o  r  m  a  n  c  e   b  e   t  w  e  e  n   1   9   8   6  a  n   d   1   9   9   2   (   h   i  g   h   d  e  n  s   i   t  y  a  n   d   d  e  n  s   i   t  y  -   d  e  p  e  n   d  e  n   t  r  e  s  p  o  n  s  e  s   ) ,  a   h   i  g   h  p  e  r   f  o  r  m  a  n  c  e   b  e   t  w  e  e  n   1   9   9   4  a  n   d   2   0   0   2   (   l  o  w   d  e  n  s   i   t  y  a  n   d  g  o  o   d  c   l   i  m  a   t   i  c  c  o  n   d   i   t   i  o  n  s   d  u  r   i  n  g  s  p  r   i  n  g –  s  u  m  m  e  r   ) ,  a  n   d  a  v  e  r  y  p  o  o  r  p  e  r   f  o  r  m  a  n  c  e   b  e   t  w  e  e  n   2   0   0   3  a  n   d   2   0   0   6   (   h   i  g   h   d  e  n  s   i   t  y  a  n   d   f  r  e  q  u  e  n   t  s  p  r   i  n  g –  s  u  m  m  e  r   d  r  o  u  g   h   t  s   ) .   A   t   T  r  o   i  s   F  o  n   t  a   i  n  e  s ,   t   h  e  p  o  p  u   l  a   t   i  o  n  s   i  z  e  w  a  s  w  e  a   k   l  y  v  a  r   i  a   b   l  e   b  e   t  w  e  e  n   1   9   8   5  a  n   d   2   0   0   1 ,  w   h  e  r  e  a  s  p  o  p  u   l  a   t   i  o  n   d  e  n  s   i   t  y   i  n  c  r  e  a  s  e   d  s   t  r  o  n  g   l  y   b  e   t  w  e  e  n   2   0   0   2  a  n   d   2   0   0   6 .   R  e   d   f  o  x  e  s  w  e  r  e  a   b  s  e  n   t   f  r  o  m   S   t  o  r   f  o  s  n  a ,   b  u   t  p  r  e  s  e  n   t   i  n  a   l   l  o   t   h  e  r  p  o  p  u   l  a   t   i  o  n  s   P  o  p  u   l  a   t   i  o  n   P  r  e   d  a   t  o  r  s  o   f  a   d  u   l   t  s  p  r  e  s  e  n   t   H  a  r  v  e  s   t   i  n  g   P  r  o   d  u  c   t   i  v   i   t  y   W   i  n   t  e  r  c   l   i  m  a   t  e   (   t  e  m  p   /  s  n  o  w  c  o  v  e  r   )   *   S  u  m  m  e  r  c   l   i  m  a   t  e   (   t  e  m  p   /  p  r  e  c .   )   †   S  p  r   i  n  g  c   l   i  m  a   t  e   (   t  e  m  p   /  p  r  e  c .   )   ‡   R  o  e   d  e  e  r   d  e  n  s   i   t  y   T   i  m  e  s  p  a  n   C  o  m  m  e  n   t  s   ¶   S   t  o  r   f  o  s  n  a   N  o   N  o   H   i  g   h   1  ·   2   C   6  ·   5  c  m    –   1      1   2  ·   8   C   8   5  ·   7  m  m    –   1      9  ·   1   C   7   2  ·   7  m  m    –   1    H   i  g   h   1   9   9   1 –   9   4   I  s   l  a  n   d  p  o  p .   Ø  s   t  e  r   d  a   l  e  n   L  y  n  x   Y  e  s   L  o  w    −    6  ·   1   C   4   2  c  m  –   1    1   4  ·   7   C   7   5  ·   1  m  m  –   1    1   0  ·   3   C   8   1  ·   5  m  m  –   1    L  o  w   1   9   9   5 –   9   8   O  p  e  n  p  o  p .   A   k  e  r  s   h  u  s   /   Ø  s   t   f  o   l   d   L  y  n  x   (  a  n   d  a   f  e  w  w  o   l  v  e  s   )   Y  e  s   M  e   d   i  u  m    −    2  ·   8   C   1   3  ·   3  c  m  –   1    1   6  ·   2   C   7   4  ·   7  m  m  –   1    1   2  ·   4   C   8   0  ·   8  m  m  –   1    M  e   d   i  u  m   2   0   0   1 –   0   4   O  p  e  n  p  o  p .   T  r  o   i  s   F  o  n   t  a   i  n  e  s   N  o   N  o   §   H   i  g   h   4  ·   9   C   0  c  m  –   1    1   9  ·   6   C   7   0  ·   1  m  m  –   1    1   5  ·   8   C   7   0  ·   2  m  m  –   1    M  e   d   i  u  m   1   9   8   5 –   2   0   0   1   F  e  n  c  e   d  p  o  p .   5  ·   1   C   0  c  m  –   1    2   0  ·   0   C   6   9  ·   5  m  m  –   1    1   6  ·   7   C   5   1  ·   6  m  m  –   1    H   i  g   h   2   0   0   2 –   0   6   C   h   i  z   é   N  o   N  o   §   L  o  w   6  ·   7   C   0  c  m  –   1    2   0  ·   7   C   4   4  ·   2  m  m  –   1    1   6  ·   6   C   4   7  ·   3  m  m  –   1    H   i  g   h   1   9   8   6 –   9   2   F  e  n  c  e   d  p  o  p .   5  ·   5   C   0  c  m  –   1    2   0  ·   4   C   5   6  ·   2  m  m  –   1    1   6  ·   7   C   5   3  ·   8  m  m  –   1    M  e   d   i  u  m   1   9   9   4 –   2   0   0   2   6  ·   3   C   0  c  m  –   1    2   1  ·   0   C   5   0  ·   0  m  m  –   1    1   7  ·   5   C   3   6  ·   4  m  m  –   1    H   i  g   h   2   0   0   3 –   0   6   *   M  e  a  n  v  a   l  u  e  s   f  o  r  w   i  n   t  e  r ,   d  e   fi  n  e   d  a  s   J  a  n  u  a  r  y –   M  a  r  c   h .   †   M  e  a  n  v  a   l  u  e  s   f  o  r  s  u  m  m  e  r ,   d  e   fi  n  e   d  a  s   J  u   l  y –   A  u  g  u  s   t .   P  r  e  c   i  p   i   t  a   t   i  o  n   i  s  m  e  a  n  v  a   l  u  e   /  m  o  n   t   h .   ‡   M  e  a  n  v  a   l  u  e  s   f  o  r  s  p  r   i  n  g ,   d  e   fi  n  e   d  a  s   M  a  y –   J  u  n  e .   P  r  e  c   i  p   i   t  a   t   i  o  n   i  s  m  e  a  n  v  a   l  u  e   /  m  o  n   t   h .   §   T   h  e  s   i  z  e  o   f   t   h  e  s  e  e  n  c   l  o  s  e   d  p  o  p  u   l  a   t   i  o  n  s  w  a  s  c  o  n   t  r  o   l   l  e   d   b  y  r  e  m  o  v  a   l  s   d  u  r   i  n  g  w   i  n   t  e  r ,   b  u   t  e  s   t   i  m  a   t  e  s  o   f   d  e  m  o  g  r  a  p   h   i  c  p  a  r  a  m  e   t  e  r  s  a  r  e  c  o  r  r  e  c   t  e   d   f  o  r   t   h   i  s  m  a  n  a  g  e  m  e  n   t ,  m  e  a  n   i  n  g   t   h  a   t   t   h  e   d  e  m  o  g  r  a  p   h  y  r  e   fl  e  c   t  s   t   h  e   d  e  m  o  g  r  a  p   h  y  o   f  u  n  e  x  p   l  o   i   t  e   d  p  o  p  u   l  a   t   i  o  n  s .   ¶   O  p  e  n  p  o  p  u   l  a   t   i  o  n   i  n   d   i  c  a   t  e  s   t   h  a   t   t   h  e  p  o  p  u   l  a   t   i  o  n   i  s  n  o   t  c   l  e  a  r   l  y   d  e   fi  n  e   d  s  p  a   t   i  a   l   l  y ,  a  s   i  s   t   h  e  c  a  s  e  w   i   t   h   t   h  e   i  s   l  a  n   d  p  o  p  u   l  a   t   i  o  n   (   S   t  o  r   f  o  s  n  a   )  a  n   d   t   h  e   f  e  n  c  e   d  p  o  p  u   l  a   t   i  o  n  s   (   T  r  o   i  s   F  o  n   t  a   i  n  e  s  a  n   d   C   h   i  z   é   ) .  588 Nilsen et al. © 2009 The Authors. Journal compilation © 2009 British Ecological Society, Journal of Animal Ecology , 78 , 585–594In the Norwegian populations, survival rates were estimated byknown-fate capture–mark–recapture (CMR) models well suited tosurvival analyses of radiotracked animals (White & Burnham 1999).We censored animals with unknown fate (e.g. due to radiocollarfailure). Litter size and proportion of females that gave birth wereestimated based on direct observations of radiocollared does in thespring close to the birth period (see also Nilsen et al  . 2004 for furtherdetails).In the French populations, survival rates were estimated usingstandard CMR methods (Lebreton  et al  . 1992) to account forcapture rates being less than 1 (about 0·5 at both Chizé and TroisFontaines for roe deer older than 1 year of age; Gaillard et al  . 1993).Age- and sex-specific survival probabilities and their SE wereobtained using the software   7·1 (Choquet  et al  . 2004). Littersize and proportion of females that gave birth were estimated fromultrasounds performed during winter captures since 1988 and fromprogesterone assays in the same period before 1988 at Chizé (seeGaillard et al  . 1992, 2003 for further details). At Trois Fontaines, weassumed that all females older than 1 year of age produced twins, assupported by the empirical evidence (see Gaillard  et al  . 1998). Thisassumption is likely to lead to a slight overestimation of thereproductive output of females during the high-density period. ROE   DEER   LIFE   CYCLE To be able to quantify the relative contribution from differentdemographic parameters to variation in λ  , a life cycle has to bedefined. We based our analysis on a pre-breeding life cycle (Fig. 1),assuming that the populations are censused just before the breedingseason each year. This implies that the youngest age class present inthe population at the census time is the 1-year-olds and that juvenilesurvival (between birth and 1 year of age) is included in the recruitmentrate (see Caswell 2001 for further details). Consequently, we consideredthe following demographic parameters in our analyses:•Recruitment rate (F): number of female offspring that enter thepopulation at time t  + 1 (just before the birth season) per female aliveat the beginning of year t (just before the birth season). The recruit-ment is thus given by the product of the proportion of females ( ≥  2years old) that give birth in year t , the mean number of femalesproduced in year t  (litter size divided by 2, i.e. assuming balanced sexratio at birth), juvenile summer survival (from birth to weaning) and juvenile winter survival (from weaning to 1 year of age).•Yearling survival (YS): survival probability through the secondyear of life (i.e. from 1 to 2 years of age).•Adult survival (AS): annual survival probability for adult females(i.e. aged 2 years or more).We based our further analysis on perturbation analysis of a transitionmatrix describing female roe deer population dynamics in five differentareas. Although demographic rates of large herbivores show strongage dependency (see e.g. Gaillard et al  . 1993 for the case of roe deersurvival patterns), we pooled all the adult age classes into one adultstage. The reason for this is twofold; first, in our Norwegian populationsthe time span of the studies were < 5 years, and ages of the individualswere known accurately only for individuals marked during their firstyear of life. Consequently, an age-structured model would havecontained very few individuals in the older age classes. Second,although both survival and reproduction do vary with age in roedeer, it is usually relatively independent of age for prime-agedfemales (2–8 years) (Gaillard  et al  . 1993; Andersen & Linnell 2000;Festa-Bianchet, Gaillard & Côté 2003). Few individuals would reachsenescence, but we are aware that this might cause a slight bias in theestimates of demographic rates and their variance. This happensbecause the survival of senescent females is lower and more variablethan that of prime-aged individuals (Gaillard et al  . 2000; Festa-Bianchet et al  . 2003). To assess the robustness of this assumption, wealso performed the calculations based on a fully age-structuredmodel for Chizé (see Table S1) where the proportion of old femaleswas the highest. The general patterns did not differ between the twomodels, but the contribution from adult survival generally decreasedslightly when accounting for senescence.Perturbation analysis of a projection matrix is based on theassumption that the stable age structure and reproductive values aregiven by the left and right eigenvectors of the projection matrix (seebelow). However, both age structure and reproductive values mightvary for populations in different phases of development. To fullyaccount for this, using a structured demographic account (SDA)would be required, but this method requires complete historicalknowledge of all individuals in the population, and in most casesthese two methods will give comparable results when covariationbetween demographic rates is accounted for (Coulson et al  . 2005).As pointed out by several authors (see e.g. Sæther & Bakke 2000; vanTienderen 2000; Coulson  et al  . 2005), it is important to control forthe effect of covariation between demographic rates. In our study,with very short time series from some of the populations and periods(see Table 1), we did not consider covariation between traits. However,the contribution of covariation among parameters accounted foronly ~15% of the variation in λ   in populations for which we had > 10years of data (see Table S2). MATRIX    ANALYSIS Projection model and analysis – functional relationships The demographic rates described above were used to construct a pre-breeding Leslie matrix for the female portion of our roe deer populations.Consider the following population model; n ( t  + 1) = An ( t ), where n ( t ) is a vector giving the abundances of different stages, and A  is aprojection matrix (Caswell 2001), whose ij   entry gives the contribu-tion of an individual in stage  j   to stage i   over one time-step (here, thetime-step is taken to be 1 year). The transition matrix ( A ) is given aseqn 1 Fig. 1. A schematic representation of roe deer life cycle as defined inthis study. The stages are yearling (1) and adult (2) respectively,whereas the transitions are given as yearling survival (YS) and adultsurvival (AS), respectively. Only adult females reproduce (F) in ourmodel and we assume no age dependence in survival andreproduction in the adult stage. See text for justification. A   = 0FYSAS
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks