- Email: [email protected]
Diel variation in movement patterns and habitat use by the Iberian endemic Cabrera vole: Implications for conservation and monitoring
Mammalian Biology 83 (2017) 21–26
Contents lists available at ScienceDirect
Mammalian Biology journal homepage: www.elsevier.com/locate/mambio
Original investigation
Diel variation in movement patterns and habitat use by the Iberian endemic Cabrera vole: Implications for conservation and monitoring Ana Rita Grácio a , António Mira b , Pedro Beja c,d , Ricardo Pita b,d,∗ a
Departamento de Biologia, Universidade de Évora, Núcleo da Mitra, Apartado 94, 7002-554 Évora, Portugal Unidade de Biologia da Conservac¸ão, CIBIO/InBio-UE, Centro de Investigac¸ão em Biodiversidade e Recursos Genéticos, Pólo de Évora, Universidade de Évora, Núcleo da Mitra, Apartado 94, 7002-554 Évora, Portugal c EDP Biodiversity Chair CIBIO/InBio-UP, Centro de Investigac¸ão em Biodiversidade e Recursos Genéticos, Universidade do Porto, Campus Agrário de Vairão, 4485-661 Vairão, Portugal d CEABN/InBIO, Centro de Ecologia Aplicada “Professor Baeta Neves”, Instituto Superior de Agronomia, Universidade de Lisboa, Tapada da Ajuda, 1349-017 Lisboa, Portugal b
a r t i c l e
i n f o
Article history: Received 9 June 2016 Accepted 15 November 2016 Handled by Adriano Martinoli Available online 16 November 2016 Keywords: Circadian cycles Movement behaviour Habitat selection Microtus cabrerae Radiotelemetry
a b s t r a c t Understanding variations in animal movement and habitat selection behaviour over fine spatial and temporal scales remains a particularly challenging goal in ecology and conservation. Here we document for the first time the diel variations in movement patterns and habitat use by wild-ranging Cabrera voles in fragmented Mediterranean farmland, based on radiotracking data (2006–2008) of 25 adult individuals occupying stable home-ranges in vegetation mosaics dominated by wet grasses and shrubs. Results indicated that the proportion of time animals spent moving, the distance moved, and the selection strength of main vegetation types were closely linked behavioural traits, which varied considerably across different periods of the 24-h cycle. In general, voles moved more frequently and over larger distances during daytime (between 06 h15–22 h00), which was when wet grasses were also used more intensively. These patterns were generally consistent across seasons, though during the dry season there was some tendency for a decrease in movement activity during the hottest hours of the day (between 10 h15–14 h00), with peaks around crepuscular hours (06 h15–10 h00 and 18 h15–22 h00). Overall, our study provides evidence that Cabrera voles may show notable shifts in habitat use and movement patterns on a finer scale than previously considered. This supports the idea that knowledge of the diel variations in species movement-habitat relationships should strongly contribute to improving local habitat management, as well as effective sampling and monitoring programs targeting the species. ¨ Saugetierkunde. ¨ © 2016 Deutsche Gesellschaft fur Published by Elsevier GmbH. All rights reserved.
Introduction The survival and reproduction of an animal depends on whether it can satisfy its basic needs, such as food, water, shelter, and breeding sites (Manly et al., 2002). Usually, no single habitat can fulfil all these needs, and thus animals often need to move among habitat types, with habitat selection reflecting a trade-off between energy intake (growth) and mortality risk (Moe et al., 2007; Railsback and Harvey, 2002). This trade-off is determined by animals’ responses to a variety of conflicting demands related to physiological needs, food acquisition, forage phenology, photoperiod, predation risk, thermal
∗ Corresponding author at: Unidade de Biologia da Conservac¸ão, CIBIO/InBioUE, Centro de Investigac¸ão em Biodiversidade e Recursos Genéticos, Pólo de Évora, Universidade de Évora, Núcleo da Mitra, Apartado 94, 7002-554 Évora, Portugal. E-mail address: [email protected] (R. Pita).
comfort, social associations, and disturbances (Lima and Dill, 1990; Verdolin, 2006). In the management of a species, it is thus essential to know which habitats are used and why and when animals select some habitats over others (Manly et al., 2002). Identifying the patterns of animal movement and the suite of conditions that defines species habitat requirements is therefore a primary goal in ecology and conservation (Mayor et al., 2009). To understand the habitat requirements of an animal species it is generally important to have information on how individuals move and shift among habitat types at fine spatio-temporal scales, i.e. along their routine activities within their home-ranges (Ager et al., 2003; Godvik et al., 2009), which in turn implies examining the foraging dynamics at the patch level, while integrating diel cycles in movement behaviour (Moe et al., 2007; Morris, 2014; Onorato et al., 2011; Di Stefano et al., 2009). Most studies regarding species habitat selection, however, are based on a collection of point locations where animals were observed over relatively long
http://dx.doi.org/10.1016/j.mambio.2016.11.008 ¨ Saugetierkunde. ¨ Published by Elsevier GmbH. All rights reserved. 1616-5047/© 2016 Deutsche Gesellschaft fur
22
A.R. Grácio et al. / Mammalian Biology 83 (2017) 21–26
(annual, seasonal) time periods (Ager et al., 2003; Godvik et al., 2009), thus ignoring the diel variations in individuals’ movement and habitat use. Also, species movement-habitat relationships are often disregarded, despite the fact that these behavioural traits together should strongly determine the scale and types of interactions between individuals and their environment (Railsback and Harvey, 2002). Understanding the short-term changes in routine movement patterns and habitat use by species along the 24-h cycle is thus fundamental to strengthen the predictions about species local abundance and persistence under environmental change (Bjørneraas et al., 2011; Ordiz et al., 2014; Tigas et al., 2002), as well as to shed light on many ecological processes operating at the population, community and ecosystem levels (Horton, 2001; Kronfeld-Schor and Dayan, 2003). While it is now widely recognized that habitat selection is a hierarchical process by which individuals choose what habitat to use at different scales, the significance of the fine-scale circadian variations in animals’ habitat use and associated movement patterns is still rarely discussed in the literature (Bestley et al., 2012; Godvik et al., 2009). This study addresses this issue, assessing movements and habitat use by the ‘near-threatened’ and Iberian endemic Cabrera vole (Microtus cabrerae) along the circadian cycle, in the dry and wet seasons. Our main hypothesis is that Cabrera voles show diel movement and habitat use patterns that are consistent with trade-offs between foraging needs and predator avoidance at fine spatial and temporal scales. To address this general hypothesis, we tested the following specific predictions: i) According to previous information on the species daily activity patterns (Pita et al., 2011a), Cabrera voles should spend more time moving, and move over larger distances in daytime, especially during the wet season; ii) Higher movement rates and distances should be associated with an increased use of open grass swards that provide the preferred foods for voles (e.g., Rosário et al., 2008), while periods of resting should be associated with habitats dominated by shrubs and tussockforming grasses that offer more cover against predators; and iii) According to the predominantly monogamous mating system of the species (e.g., Pita et al., 2014), the circadian variations in habitat use and movement patterns should not differ between sexes. Support for such behavioural patterns along the 24-h cycle may have important implications for local habitat management, particularly regarding the relative amount, and spatial arrangement of main vegetation types favouring the persistence of resident individuals. In addition, the patterns of movement and habitat use at fine spatio-temporal scales may provide useful behavioural indicators to infer species population status, and to monitor and predict their responses to environmental change (e.g., Caro, 2007; Sutherland, 1998).
out in summer, whereas permanent water bodies are scarce and mostly associated with irrigation infrastructure such as concrete channels and reservoirs (Ferreira and Beja, 2013; Pita et al., 2010). Over the past two decades, agricultural practices have intensified considerably, with largely negative consequences for farmland biodiversity (e.g., Beja and Alcazar, 2003; Beja et al., 2014; Ferreira and Beja, 2013; Pita et al., 2009). Study animals We used data from 25 radio-tracked adult Cabrera voles (mean ± se weight = 50.5 ± 1.39 g) exhibiting stable home-ranges (465.90 ± 69.03 m2 , based on 95% Kernel) encompassing mosaics of wet grassy and shrubby vegetation (see Pita et al., 2010, 2011a,b). Animals were surveyed within 11 colonies in two farmland areas in south-west Portugal (37◦ 58 –37◦ 54 N, 08◦ 48 –08◦ 44 W and 37◦ 38–37◦ 35 N, 08◦ 49 –08◦ 46 W), between April-2006 and April2008. Overall, 10 voles (7 females and 3 males) were tracked during the dry season (May-September, 4 colonies), and 15 voles (9 females and 6 males) were tracked during the wet season (October–April, 7 colonies). Details regarding capture and handling of animals are provided in Pita et al. (2010, 2011a,b). In brief, we used Sherman live-traps baited with apple and supplied with hay and hydrophobic cotton for bedding, with a mean ± se traps × days per sampled colony of 403.7 ± 84.2. At capture, voles were weighed, sexed, and fitted with collar radio-transmitters (Wildlife Materials, Inc., Murphysboro, IL, USA), adding no more than 5% of the animals weigh (Gannon and Sikes, 2007). All animals were lightly sedated with a subcutaneous injection of Dormitor® (0.2 mg/kg) to reduce handling stress. After transmitter attachment, voles were induced out of anaesthesia using an equivalent dose of Antisedam® . Before release, animals were kept under observation for at least 2 h to ensure they were suffering no ill effects or loss of mobility. Voles were released at their point of capture and radiotracking started at least 4 h after release, to avoid sampling atypical behaviours while animals became familiar with collars and recovered from stress. A total of 48% of collars were recovered from re-trapping by the end of surveys, with no signs of injuries on voles ever detected. In addition, 8% of collars were found within home ranges, presumably because voles managed to remove them. Another 16% of collars were found outside home ranges and showed damaged coats or cable ties, presumably because of predation. Direct evidence of predation was only observed in the case of one female consumed by a ladder snake, Elaphe scalaris, 7 days after collar attachment. The remaining collars (24%) were lost because of vole dispersal, battery discharge or untraced predation (Pita et al., 2011a). Sampling design and variables
Methods Study area The study was conducted in southwest Portugal, which is included in the thermo-Mediterranean bioclimatic zone and is characterized by strong seasonality in environmental conditions. Mean monthly temperature range between 9 and 23 ◦ C and average annual rainfall is around 680 mm, of which about 80% occurs between October and April (Sistema Nacional de Informac¸ão de Recursos Hídricos, http://snirh.pt). The region is predominantly flat and dominated by agriculture, with almost half of the land devoted to irrigated annual crops. Cattle production is also important, resulting in large areas occupied by pastures, fodder crops and silage corn or sorghum. Wood cover is restricted to arboreal windbreaks and a few woodlots. Surface waters are mostly associated with temporary ponds which flood during the rainy season and dry
Routine movement patterns and habitat use by voles within their home ranges were sampled in six different 4h-periods of the day covering the whole 24 h cycle, and thus reflecting different sunlight, climate and biotic conditions: 02 h15–06 h00 (late night/early morning); 06 h15–10 h00 (morning); 10 h15–14 h00 (noon); 14 h15–18 h00 (afternoon); 18 h15–22 h00 (evening); and 22 h15–02 h00 (night). Each animal was sampled in a different period every day along a mean ± se of 11.8 ± 1.6 days per vole (range: 3–23 days). During each sampling period, locations were carried out at 15-min intervals, totalling 16 locations per sampling period. Voles were located using a TRX-100S receiver and an external 3-element Yagi directional antenna (Wildlife Materials, Inc.). Locations were made by homing and by multiple triangulations when the tracker was close to the animals (about 50 cm). Voles were frequently seen during tracking, and appeared little affected by the presence of the observer. At each location, the GPS coordinates were recorded (±4 m error) along with the dominant vegetation
A.R. Grácio et al. / Mammalian Biology 83 (2017) 21–26
23
type within 50 cm-radius plots (Pita et al., 2010, 2011a,b). For this, two main vegetation types were considered: i) herbaceous swards dominated by wet grasses and forbs, usually >30 cm tall (hereafter grasses); and ii) shrubs (e.g., bramble Rubus thickets) or tussockforming herbs (e.g., sedges, rushes, or reeds), usually >60 cm tall (hereafter shrubs). From these data we extracted, for each sampling period and animal, the following dependent variables: i) Proportion of time spent moving (PM), given by the number of times the animal moved in each period divided by the total number of observations; ii) Overall movement length (OM), given by summing up the distances travelled between successive point locations; and iii) Grass selectivity index (GS), given as the frequency of point locations (habitat used) dominated by grassy vegetation relative to respective overall frequency (habitat available), recorded for each animal throughout overall sampling time. The latter corresponds to the Ivlev selec˜ et al., 2010), and measures the strength tivity index (e.g., Magana of third order selection of grasses relative to availability (Johnson, 1980).
Table 1 Ranking of the models for each season, specifically the ‘null’ models and the ‘circadian’ models (including the period of the day as a fixed effect).
Data analyses
Results
To investigate whether the period of day affected the three variables describing vole movements and habitat use in each season, we used mixed-effects multivariate analysis of variance (MANOVAs), which is a generalization of analysis of variance (ANOVA) that allows the analysis of multiple dependent variables, while accounting for possible correlation among them (Bray and Maxwell, 1985). Survey years, colonies, and animals were included as random intercept effects, i.e. grouping variables with random variations around an overall population mean (see e.g. Gillies et al., 2006). To reduce the influence of extreme values and enhance normality of model residuals, the scaled dependent variables were arcsin-transformed in the case of the proportion of time spent moving, and logtransformed in the case of the overall movement length (e.g. Wilson et al., 2010). This also resulted in improved model mixing and convergence. In model building, the intercept was suppressed as suggested by Hadfield (2016). Analyses were conducted for each season separately, because previous studies have shown considerable seasonal differences in both circadian activity rhythms and habitat selection by Cabrera voles (Pita et al., 2011a,b). For each season we first built a ‘circadian’ model, in which the period of the day was included as a fixed effect. In order to assess the support of the ‘circadian’ models, we then built concurrent ‘null’ models (i.e. including only the random effects), and checked whether the ‘circadian’ models yielded deviance information criterion (DIC) at least 5 units lower than that of the corresponding ‘null’ models (Spiegelhalter et al., 2002). To test the importance of the sex in the circadian variation of habitat use and movement patterns, we built an additional model for each season including the interaction between the period of the day and the sex of tracked voles (‘circadian*sex’ model). We also assessed eventual differences in mean circadian movement patterns and habitat use between males and females, by running a further model including the main effect of sex (‘circadian + sex’ model). Support for any of these models was assessed by checking whether the corresponding DIC values were at least 5 units lower than that of the simple ‘circadian’ models. Finally, model fit of the most supported models was estimated using pseudo-R2 (Johnson, 2014), and the respective posterior distributions were used to estimate correlations among the three dependent variables in each season (e.g., Hadfield, 2010; Wilson et al., 2010), with significant correlations determined by the 95% credible intervals not overlapping with zero. All models were run with Bayesian Markov Chain Monte Carlo simulation using the package ‘MCMCglmm’ version 2.22 (Hadfield, 2010) and the program R 3.1.3 (R Core Team, 2016), keeping
A total of 3,664 location records (2,080 in the wet season and 1,584 in the dry season) were obtained for the 25 voles radio tracked, with a mean ± se of 146.56 ± 14.68 locations per animal. This corresponded to a total of 130 and 99 experimental units (4-h periods) sampled in the wet and dry seasons respectively. Despite the variability observed in individual responses (see Fig. SM1, Supplementary material), there were some differences in the mean values of the dependent variables estimated among the six different 4-h sampling periods considered, for both seasons. In general, voles tended to spend more time moving and moved over larger distances in daytime, which is also when they used grassy areas more often (Table SM1, Supplementary material). This was confirmed by the mixed-effects MANOVA, which showed that the ‘circadian’ models for each season were strongly supported when compared to the respective ‘null’ models (Table 1). In particular, using the period between 02 h15–06 h00 as baseline, posterior estimates of model coefficients regarding the proportion of time animals spent moving during the wet season were significantly higher between 06 h15 and 22 h00, with the highest values between 10 h15 and 18 h00 (noon and afternoon) (Table SM2, Supplementary material), resulting in higher predicted estimates during these periods (Fig. 1a). These results were generally also observed during the dry season, though the highest posterior estimates of model coefficients were recorded between 06 h15–10 h00 (morning) (Table SM2, Supplementary material; Fig. 1b). Very similar patterns were found for both the overall movement length (Fig. 1c–d) and the strength of selection of grassy areas over shrubs, with significantly higher values between 06 h15 and 22 h00 during both seasons (Table SM2, Supplementary material; Fig. 1e–f). Overall movement length was higher between 14 h15–18 h00 during the wet season (afternoon), and between 06 h15–10 h00 (morning) during the dry season (Table SM2, Supplementary material; Fig. 1c–d). In general, the selection of grasses was particularly strong between 06 h15–10 h00 during both seasons (Table SM1, Supplementary material; Fig. 1e–f). Despite the general consistency in diel movement patterns and habitat use by voles in each season, the three random effects considered in the models, while not being explanatory, also introduced variation influencing response variables. In general, posterior mean estimates of explained variation suggested a greater importance of inter-annual variation, followed by inter-colony and interindividual variation (see Table SM2, Supplementary material). The ‘circadian*sex’ models yielded much higher DIC values than the simple ‘circadian’ model for both the wet (DIC = 22.73) and
Season
Model
DIC
DIC
Adjusted pseudo R2
Wet
‘null’ model ‘circadian’ model
666.49 634.66
31.83
– 0.39
Dry
‘null’ model ‘circadian’ model
661.55 609.67
51.88
– 0.51
>1000 posterior samples (Hadfield, 2016). Models were run using default uninformative flat priors, considering acceptable levels of first order autocorrelation (generally <0.01 for successive iterations, Hadfield, 2010; Plummer et al., 2006) and until they reached convergence, as assessed computationally using Geweke’s convergence diagnostic (Plummer et al., 2006). This implied runs with 20,000 iterations, burn-in size of 2000 and thinning interval of 10 iterations. Adjusted pseudo-R2 values were estimated with ´ 2016). ‘MuMin’ (Barton,
24
A.R. Grácio et al. / Mammalian Biology 83 (2017) 21–26
Fig. 1. Posterior estimates and 95% confidence intervals of predicted responses in each of the six 4-h periods of the day considered, according to the ‘circadian’ models for the wet and dry seasons: Proportion of time spent moving (a and b); overall movement length (c and d); and grass selectivity index (e and f).
Table 2 Correlations among the proportion of time spending moving (PM), overall movement length (OM) and grass selectivity index (GS) extracted from the ‘circadian’ models for each season. Season
Variables
Correlation(posterior mode)
95% CI
Wet
PM vs. OM PM vs. GS OM vs. GS
0.73 0.48 0.39
0.65–0.84 0.29–0.62 0.23–0.58
Dry
PM vs. OM PM vs. GS OM vs. GS
0.78 0.31 0.41
0.67–0.85 0.13–0.56 0.20–0.56
dry (DIC = 14.97) seasons, indicating that sex had little effect on circadian movement and habitat use patterns (see Table SM3 in Supplementary material). The ‘circadian + sex’ models also yielded higher DIC values than the ‘circadian’ models, though the respective DIC were lower than 5 in both the wet (DIC = 2.97) and dry (DIC = 0.47) seasons. These models suggested some tendency for sex differences in movement activity and habitat use, though these were never significant for either season (see Table SM4 in Supplementary material). There were significant pairwise correlations between all dependent variables estimated using the posterior distributions of the ‘circadian’ models for both seasons (Table 2). Discussion Patterns of habitat use and movement are key topics in ecological and conservation studies, because they affect animals’ distribution at different temporal and spatial scales, and can strongly influence their fitness and survival (e.g., Morales et al., 2010; Russell et al., 2003). These patterns are usually suggested to result from trade-offs between food acquisition efficiency and mor-
tality risk, which are expected to vary across the 24-h cycle for most taxa, according to endogenous biological rhythms and external environmental factors (Railsback et al., 2005). Although seasonal variation in resource selection and circadian activity rhythms of Cabrera voles have been reported previously (Pita et al., 2011a,b; Santos et al., 2005), our study provides the first integrated approach showing the link between habitat use and movement patterns within individuals home-ranges, and how joint decisions of what habitat to occupy and how to move within it may reflect circadian variations in foraging versus hiding/resting activities across seasons (Pépin et al., 2009; Tigas et al., 2002). Overall, results were consistent with our first prediction that, despite the likely influence of inter-annual, inter-patch and interindividual variations, Cabrera voles should spend more time moving and move over larger distances in daytime than at night. This unimodal (diurnal) movement pattern was particularly evident during the wet season, and it is in line with previous studies on the circadian activity pattern of this species (Pita et al., 2011a). Also in line with previous research, the movement during the dry season was largely bimodal (crepuscular), peaking in the morning (06 h15–10 h00), declining at mid-day (10 h15–14 h00), and then increasing again in the evening (18 h15–22 h00), while it was minimal at night (22 h15–06 h00). The mid-day decline in movement during the dry season probably reflects a behavioural response to minimize exposure to high temperatures, helping individuals to cope with physiological and metabolic needs during the hottest hours (Mathias et al., 2003; Pita et al., 2011b). Our second prediction that diel movement and habitat use patterns are linked and cyclic was also confirmed, as results showed that increased movements during the day were associated with a higher use of grassy areas, where individuals were probably foraging, while reduced movements were associated with the use of shrubby areas, where
A.R. Grácio et al. / Mammalian Biology 83 (2017) 21–26
individuals were probably resting. Considering that individuals select the combination of habitat and movement that maximizes their fitness (e.g., Rhodes et al., 2005), it is likely that these temporal shifts were related to a higher forage abundance and quality of grassy areas to fulfil food intake requirements while foraging, and a higher value of shrubby areas for reducing predation risk (LuqueLarena and López, 2007; Pita et al., 2006; Rosário et al., 2008). Resting in shrubby areas should be particularly favourable, because shrubs and tussock-forming herbs may provide concealment and some protection to voles in periods when vigilance is presumably low and thus vulnerability might be high (Ganskoop and Johnson, 2007; Semeniuk and Dill, 2005). Finally, there was support for our prediction that male and female Cabrera voles should have similar circadian patterns of movement and habitat use. This is probably a consequence of the species’ predominantly monogamous mating system and biparental care (e.g., Fernández-Salvador et al., 2001; Pita et al., 2014), which implies that both sexes likely share much the same time intervals to search for feeding areas and to secure resting and nesting. Taken together, results from this study support the view that the way animals use their environment depends on the characteristics of the resources available at fine scales, and that circadian variations in environmental conditions may impose differences in individuals behavioural choices related to movement patterns and habitat use (e.g., Bennitt et al., 2015). In particular, our study showed that within their home-ranges, Cabrera voles may distinguish between particular vegetation types based on their structural and functional value, which in turn affect their suitability for particular behaviours along the 24 h cycle. While we acknowledge that replicating surveys across a larger number of animals, patches and years would have probably increased the robustness and consistency of our findings, we believe that the relationships uncovered here provide useful information for a more complete understanding of movement activity and habitat use by Cabrera voles. This highlights the need to include individual behavioural traits when analysing resource use by the species at the population level, as some vegetation types may be more suited for particular behaviours than others (e.g., Bjørneraas et al., 2011; Dzialak et al., 2012). Implications for conservation and monitoring Although we cannot exclude the influence of other sources of variation not considered here (e.g., geographic variation, and occurrence of predators and intra- and interspecific abundance at local scales) (e.g., Ager et al., 2003), our study clearly shows that the period of the day is an important driver of changes in movement and habitat use of Cabrera voles. Therefore, based on the patterns found, a set of recommendations may be posited for improving local conservation management and monitoring programs targeting the species. In particular, our findings suggest the need to maintain or create potential habitat patches composed of a mosaic of wet grasses and shrubs or tussock-forming herbs, likely to guarantee high food intake, while minimizing foraging space and travel time to safer sites during voles’ routine activities within their homeranges (e.g., Ferron and Ouellet, 1992). This may require occasional management of local habitat patches, such as for instance shrub control to prevent encroachment into grassy, food-rich areas (Pita et al., 2007). It should be noted, however, that possible management actions should take into account the dynamic nature of habitat selection and movement patterns by the species across the diel cycle. Another important contribution of our study is that, by analysing the fine-scaled shifts in diel movement and habitat selection of adult individuals with stable home-ranges, it provides baseline information for future research aiming to assess individual fitness after disturbances, or translocation programs, as well as to identify
25
deviations from the expected behaviours at particular life-history stages such as dispersal or colonization (Caro, 2007; Sutherland, 1998). Such behavioural indicators of individual conditions may in turn be used to infer the overall population status at early stages, i.e. before demographic indicators become noticed (e.g., Haroldson and Fritzell, 1984; Onorato et al., 2011; Di Stefano et al., 2009). The circadian changes in Cabrera voles’ movement and habitat use shown here may also help refining the sampling and monitoring programs focusing on particular behaviours or ecological processes, namely in what concerns the choice of the temporal scale and optimal allocation of resources for sampling. For instance, in telemetrybased studies focusing on foraging voles and their interactions with predators, it may be preferred to sample more often during daytime or more seldom during night time, according to the species diel behavioural patterns (e.g., Moe et al., 2007). Overall, therefore, our study illustrates the practical utility of incorporating fine-scaled shifts in animal behaviour across the diel cycle as part of spatial prediction of species persistence ability and overall state of populations, thereby contributing to improving species conservation and monitoring programs (e.g., Berger-Tal et al., 2011; Kotler et al., 2016). Acknowledgements This study was funded by the European Regional Development Fund (ERDF) through the Operational Program for Competitiveness Factors (COMPETE), and National Funds through the Foundation for Science and Technology (FCT), under the projects ‘NetPersist’ (PTDC/AAG-MAA/3227/2012) and ‘MateFrag’ (PTDC/BIA-BIC/6582/2014). RP was supported by the grants SFRH/BPD/73478/2010 and SFRH/BPD/109235/2015. PB was supported by the EDP Biodiversity Chair. Capture and handling of voles were carried-out under the permission of the Portuguese nature conservation authority (ICNF) and were supervised by Dr. Manuel Mestre from ZooMédica Veterinary Clinic. All procedures conformed to the guidelines approved by the American Society of Mammalogists for the use of wild mammals in research. We thank R. Brito and M. Duarte for help during field surveys. We also thank Julian Di Stefano and two anonymous reviewers for helpful comments to improve the paper. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.mambio.2016.11. 008. References Ager, A.A., Johnson, B.K., Kern, J.W., Kie, J.G., 2003. Daily and seasonal movements and habitat use by female rocky mountain elk and mule deer. J. Mammal. 84, 1076–1088. ´ K., 2016. Multi-Model Inference. Package MuMIn. https://cran.r-project. Barton, org/web/packages/MuMIn/index.html. Beja, P., Alcazar, R., 2003. Conservation of Mediterranean temporary ponds under agricultural intensification: an evaluation using amphibians. Biol. Conserv. 114, 317–326. Beja, P., Schindler, S., Santana, J., Porto, M., Morgado, R., Moreira, F., Pita, R., Mira, A., Reino, L., 2014. Predators and livestock reduce bird nest survival in intensive Mediterranean farmland. Eur. J. Wildl. Res. 60, 249–258. Bennitt, E., Bonyongo, M.C., Harris, S., 2015. Behaviour-related scalar habitat use by cope buffalo (Syncerus caffer caffer). PLoS One 10, e0145145. Berger-Tal, O., Polak, T., Oron, A., Lubin, T., Kotler, B.P., Saltz, D., 2011. Integrating animal behavior and conservation biology: a conceptual framework. Behav. Ecol. 22, 236–239. Bestley, S., Jonsen, I.D., Hindell, M.A., Guinet, C., Charrassin, J.B., 2012. Integrative modelling of animal movement: incorporating in situ habitat and behavioural information for a migratory marine predator. Proc. R. Soc. B: Biol. Sci. 280, 20122262. Bjørneraas, K., Solberg, E.J., Herfindal, I., Moorter, B.V., Rolandsen, C.M., Tremblay, J.-P., Skarpe, C., Sæther, B.-E., Eriksen, R., Astrup, R., 2011. Moose Alces Alces
26
A.R. Grácio et al. / Mammalian Biology 83 (2017) 21–26
habitat use at multiple temporal scales in a human-altered landscape. Wildl. Biol. 17, 44–54. Bray, J.H., Maxwell, S.E., 1985. Multivariate Analysis of Variance. SAGE Publications California, Newbury Park, CA. Caro, T., 2007. Behavior and conservation: a bridge too far? Trends Ecol. Evol. 22, 394–400. Di Stefano, J., York, A., Swan, M., Greenfield, A., Coulson, G., 2009. Habitat selection by the swamp wallaby (Wallabia bicolour) in relation to diel period, food and shelter. Austral. Ecol. 34, 143–155. Dzialak, M.R., Olson, C.V., Harju, S.M., Webb, S.L., Winstead, J.B., 2012. Temporal and hierarchical spatial components of animal occurrence: conserving seasonal habitat for greater sage-grouse. Ecosphere 3, 30. Fernández-Salvador, R., García-Perea, R., Ventura, J., 2001. Reproduction and postnatal growth of the Cabrera vole, Microtus cabrerae, in captivity. Can. J. Zool. 79, 2080–2085. Ferreira, M., Beja, P., 2013. Mediterranean amphibians and the loss of temporary ponds: are there alternative breeding habitats? Biol. Conserv. 165, 179–186. Ferron, J., Ouellet, J.-P., 1992. Daily partitioning of summer habitat and use of space by the snowshoe hare in southern boreal forest. Can. J. Zool. 70, 2178–2183. Gannon, W.L., Sikes, R.S., 2007. Animal care and use committee of the american society of mammalogists guidelines of the american society of mammalogists for the use of wild mammals in research. J. Mammal. 88, 809–823. Ganskoop, D.C., Johnson, D.D., 2007. GPS error in studies addressing animal movements and activities. Rangel. Ecol. Manag. 60, 350–358. Gillies, C.S., Hebblewhite, M., Nielsen, S.E., Krawchuk, M.A., Aldridge, C.L., Frair, J.L., Saher, D.J., Stevens, C.E., Jerde, C.L., 2006. Application of random effects to the study of resource selection by animals. J. Anim. Ecol. 75, 887–898. Godvik, I.M.R., Loe, L.E., Vik, J.O., Veiberg, V., Langvatn, R., Mysterud, A., 2009. Temporal scales, trade-offs, and functional responses in red deer habitat selection. Ecology 90, 699–710. Hadfield, J.D., 2010. MCMC methods for multi–response generalised linear mixed models: the MCMCglmm R package. J. Stat. Softw. 33, 1–22. Hadfield, J., 2016. MCMC Generalised Linear Mixed Models. Package MCMCglmm (Accessed 9 June 2016) https://cran.r-project.org/web/packages/MCMCglmm/ index.html. Haroldson, K.J., Fritzell, E.K., 1984. Home ranges, activity, and habitat use by gray foxes in an oak-hickory forest. J. Wildl. Manag. 48, 222–227. Horton, T.H., 2001. Conceptual issues in the ecology and evolution of circadian rhythms. In: Takahashi, J.S., Turek, F.W., Moore, R.Y. (Eds.), Handbook of Neurobiology: Circadian Clocks 12. Kluwer Academic/Plenum Publishers, New York, pp. 45–57. Johnson, D.H., 1980. The comparison of usage and availability measurements for evaluating resource preference. Ecology 61, 65–71. Johnson, P.C.D., 2014. Extension of Nakagawa & Schielzeth’s R2 GLMM to random slopes models. Method Ecol. Evol. 5, 44–946. Kotler, B.P., Morris, D.W., Brown, J.S., 2016. Direct behavioral indicators as a cconservation and management tool. In: Berger-Tal, O., Saltz, D. (Eds.), Conservation Behaviour: Applying Behavioural Ecology to Wildlife Conservation and Management. Cambridge University Press, Cambridge, pp. 307–353. Kronfeld-Schor, N., Dayan, T., 2003. Partitioning of time as an ecological resource. Annu. Rev. Ecol. Syst. 34, 153–181. Lima, S.L., Dill, L.M., 1990. Behavioural decisions made under the risk of predation. Can. J. Zool. 68, 619–640. Luque-Larena, J.J., López, P., 2007. Microhabitat use by wild-ranging Cabrera voles Microtus cabrerae as revealed by live trapping. Eur. J. Wildl. Res. 53, 221–225. ˜ M., Alonso, J.C., Martin, C.A., Bautista, L.M., Martin, B., 2010. Nest-site Magana, selection by Great Bustards Otis tarda suggests a trade-off between concealment and visibility. Ibis 152, 77–89. Manly, B., McDonald, L., Thomas, D., MacDonald, T., Erickson, W., 2002. Resource Selection by Animals. Statistical Design and Analysis for Field Studies. Kluwer Academic Publisher, Springer, Neaderlands. Mathias, M.L., Klunder, M., Santos, S.M., 2003. Metabolism and thermoregulation in the Cabrera vole (Rodentia: Microtus cabrerae). Comp. Biochem. Phys. A 136, 441–446. Mayor, S.J., Schneider, D.C., Schaefer, J.A., Mahoney, S.P., 2009. Habitat selection at multiple scales. Écoscience 16, 238–247. Moe, T.F., Kindberg, J., Jansson, I., Swenson, J.E., 2007. Importance of diel behaviour when studying habitat selection: examples from female Scandinavian brown bears (Ursus arctos). Can. J. Zool. 85, 518–525.
Morales, J.M., Moorcroft, P.R., Matthiopoulos, J., Frair, J.L., Kie, J.G., Powell, R.A., Merrill, E.H., Haydon, D.T., 2010. Building the bridge between animal movement and population dynamics. Philos. Trans. R. Soc. B 365, 2289–2301. Morris, D.W., 2014. Can foraging behaviour reveal the eco-evolutionary dynamics of habitat selection? Evol. Ecol. Res. 16, 1–18. Onorato, D.P., Criffield, M., Lotz, M., Cunningham, M., McBride, R., Leone, E.H., Bass Jr, O.L., Hellgren, E.C., 2011. Habitat selection by critically endangered Florida panthers across the diel period: implications for land management and conservation. Anim. Conserv. 14, 196–205. Ordiz, A., Kindberg, J., Sæbø, S., Swenson, J.E., Støen, O.-Gunnar, 2014. Brown bear circadian behaviour reveals human environmental encroachment. Biol. Conserv. 173, 1–9 (Elsevier). Pépin, D., Morellet, N., Goulard, M., 2009. Seasonal and daily walking activity patterns of free-ranging adult red deer (Cervus elaphus) at the individual level. Eur. J. Wildl. Res. 55, 479–486. Pita, R., Mira, A., Beja, P., 2006. Conserving the Cabrera vole, Microtus cabrerae, in intensively used Mediterranean landscapes. Agric. Ecosyst. Environ. 115, 1–5. Pita, R., Beja, P., Mira, A., 2007. Spatial population structure of the Cabrera vole in Mediterranean farmland: the relative role of patch and matrix effects. Biol. Conserv. 134, 383–392. Pita, R., Mira, A., Moreira, F., Morgado, R., Beja, P., 2009. Influence of landscape characteristics on carnivore diversity and abundance in Mediterranean farmland. Agric. Ecosyst. Environ. 132, 57–65. Pita, R., Mira, A., Beja, P., 2010. Spatial segregation of two vole species (Arvicola sapidus and Microtus cabrerae) within habitat patches in a highly fragmented farmland landscape. Eur. J. Wildl. Res. 56, 651–662. Pita, R., Mira, A., Beja, P., 2011a. Circadian activity rhythms in relation to season, sex and interspecific interactions in two Mediterranean voles. Anim. Behav. 81, 1023–1030. Pita, R., Mira, A., Beja, P., 2011b. Assessing habitat differentiation between coexisting species: the role of spatial scale. Acta Oecol. 37, 124–132. Pita, R., Mira, A., Beja, P., 2014. Microtus cabrerae (Rodentia: Cricetidae). Mamm. Species 46, 48–70. Plummer, M., Best, N., Cowles, K., Vines, K., 2006. CODA: convergence diagnosis and output analysis for MCMC. Newsl. R. Proj. 6, 7–11. R Core Team, 2016. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria (Accessed 9 June 16) http://www.R-project.org/. Railsback, S.F., Harvey, B.C., 2002. Analysis of habitat-selection rules using an individual-based model. Ecology 83, 1817–1830. Railsback, S.F., Harvey, B.C., Hayse, J.W., LaGory, K.E., 2005. Tests of theory for diel variation in salmonid feeding activity and habitat use. Ecology 86, 947–959. Rhodes, J.R., McAlpine, C.A., Lunney, D., Possingham, H.P., 2005. A spatially explicit habitat selection model incorporating home range behavior. Ecology 86, 1199–1205. Rosário, I.T., Cardoso, P.E., Mathias, M.L., 2008. Is habitat selection by the Cabrera vole (Microtus cabrerae) related to food preferences? Mamm. Biol. 73, 423–429. Russell, R.E., Swihart, R.K., Feng, Z., 2003. Population consequences of movement decisions in a patchy landscape. Oikos 103, 142–152. Santos, S.M., Rosário, I.T., Mathias, M.L., 2005. Microhabitat preference of the Cabrera vole in a Mediterranean cork oak woodland of southern Portugal. Vie et Milieu 55, 53–59. Semeniuk, C.A.D., Dill, L.M., 2005. Cost/benefit analysis of group and solitary resting in the cowtail stingray, Pastinachus sephen. Behav. Ecol. 16, 417–426. Spiegelhalter, D.J., Best, N.G., Carlin, B.P., Van Der Linde, A., 2002. Bayesian measures of model complexity and fit. J. R. Stat. Soc. B 64, 583–639. Sutherland, W.J., 1998. The importance of behavioural studies in conservation biology. Anim. Behav. 56, 801–809. Tigas, L.A., Vuren, D.H.V., Sauvajot, R.M., 2002. Behavioral respondes of bobcats and coyotes to habitat fragmentation and corridors in an urban environmental. Biol. Conserv. 108, 299–306. Verdolin, J.L., 2006. Meta-analysis of foraging and predation risk trade-offs in terrestrial systems. Behav. Ecol. Sociobiol. 60, 457–464. Wilson, A.J., Réale, D., Clements, M.N., Morrissey, M.M., Postma, E., Walling, C.A., Kruuk, L.E.B., Nussey, D.H., 2010. An ecologist’s guide to the animal model. J. Anim. Ecol. 79, 13–26.
Contents lists available at ScienceDirect
Mammalian Biology journal homepage: www.elsevier.com/locate/mambio
Original investigation
Diel variation in movement patterns and habitat use by the Iberian endemic Cabrera vole: Implications for conservation and monitoring Ana Rita Grácio a , António Mira b , Pedro Beja c,d , Ricardo Pita b,d,∗ a
Departamento de Biologia, Universidade de Évora, Núcleo da Mitra, Apartado 94, 7002-554 Évora, Portugal Unidade de Biologia da Conservac¸ão, CIBIO/InBio-UE, Centro de Investigac¸ão em Biodiversidade e Recursos Genéticos, Pólo de Évora, Universidade de Évora, Núcleo da Mitra, Apartado 94, 7002-554 Évora, Portugal c EDP Biodiversity Chair CIBIO/InBio-UP, Centro de Investigac¸ão em Biodiversidade e Recursos Genéticos, Universidade do Porto, Campus Agrário de Vairão, 4485-661 Vairão, Portugal d CEABN/InBIO, Centro de Ecologia Aplicada “Professor Baeta Neves”, Instituto Superior de Agronomia, Universidade de Lisboa, Tapada da Ajuda, 1349-017 Lisboa, Portugal b
a r t i c l e
i n f o
Article history: Received 9 June 2016 Accepted 15 November 2016 Handled by Adriano Martinoli Available online 16 November 2016 Keywords: Circadian cycles Movement behaviour Habitat selection Microtus cabrerae Radiotelemetry
a b s t r a c t Understanding variations in animal movement and habitat selection behaviour over fine spatial and temporal scales remains a particularly challenging goal in ecology and conservation. Here we document for the first time the diel variations in movement patterns and habitat use by wild-ranging Cabrera voles in fragmented Mediterranean farmland, based on radiotracking data (2006–2008) of 25 adult individuals occupying stable home-ranges in vegetation mosaics dominated by wet grasses and shrubs. Results indicated that the proportion of time animals spent moving, the distance moved, and the selection strength of main vegetation types were closely linked behavioural traits, which varied considerably across different periods of the 24-h cycle. In general, voles moved more frequently and over larger distances during daytime (between 06 h15–22 h00), which was when wet grasses were also used more intensively. These patterns were generally consistent across seasons, though during the dry season there was some tendency for a decrease in movement activity during the hottest hours of the day (between 10 h15–14 h00), with peaks around crepuscular hours (06 h15–10 h00 and 18 h15–22 h00). Overall, our study provides evidence that Cabrera voles may show notable shifts in habitat use and movement patterns on a finer scale than previously considered. This supports the idea that knowledge of the diel variations in species movement-habitat relationships should strongly contribute to improving local habitat management, as well as effective sampling and monitoring programs targeting the species. ¨ Saugetierkunde. ¨ © 2016 Deutsche Gesellschaft fur Published by Elsevier GmbH. All rights reserved.
Introduction The survival and reproduction of an animal depends on whether it can satisfy its basic needs, such as food, water, shelter, and breeding sites (Manly et al., 2002). Usually, no single habitat can fulfil all these needs, and thus animals often need to move among habitat types, with habitat selection reflecting a trade-off between energy intake (growth) and mortality risk (Moe et al., 2007; Railsback and Harvey, 2002). This trade-off is determined by animals’ responses to a variety of conflicting demands related to physiological needs, food acquisition, forage phenology, photoperiod, predation risk, thermal
∗ Corresponding author at: Unidade de Biologia da Conservac¸ão, CIBIO/InBioUE, Centro de Investigac¸ão em Biodiversidade e Recursos Genéticos, Pólo de Évora, Universidade de Évora, Núcleo da Mitra, Apartado 94, 7002-554 Évora, Portugal. E-mail address: [email protected] (R. Pita).
comfort, social associations, and disturbances (Lima and Dill, 1990; Verdolin, 2006). In the management of a species, it is thus essential to know which habitats are used and why and when animals select some habitats over others (Manly et al., 2002). Identifying the patterns of animal movement and the suite of conditions that defines species habitat requirements is therefore a primary goal in ecology and conservation (Mayor et al., 2009). To understand the habitat requirements of an animal species it is generally important to have information on how individuals move and shift among habitat types at fine spatio-temporal scales, i.e. along their routine activities within their home-ranges (Ager et al., 2003; Godvik et al., 2009), which in turn implies examining the foraging dynamics at the patch level, while integrating diel cycles in movement behaviour (Moe et al., 2007; Morris, 2014; Onorato et al., 2011; Di Stefano et al., 2009). Most studies regarding species habitat selection, however, are based on a collection of point locations where animals were observed over relatively long
http://dx.doi.org/10.1016/j.mambio.2016.11.008 ¨ Saugetierkunde. ¨ Published by Elsevier GmbH. All rights reserved. 1616-5047/© 2016 Deutsche Gesellschaft fur
22
A.R. Grácio et al. / Mammalian Biology 83 (2017) 21–26
(annual, seasonal) time periods (Ager et al., 2003; Godvik et al., 2009), thus ignoring the diel variations in individuals’ movement and habitat use. Also, species movement-habitat relationships are often disregarded, despite the fact that these behavioural traits together should strongly determine the scale and types of interactions between individuals and their environment (Railsback and Harvey, 2002). Understanding the short-term changes in routine movement patterns and habitat use by species along the 24-h cycle is thus fundamental to strengthen the predictions about species local abundance and persistence under environmental change (Bjørneraas et al., 2011; Ordiz et al., 2014; Tigas et al., 2002), as well as to shed light on many ecological processes operating at the population, community and ecosystem levels (Horton, 2001; Kronfeld-Schor and Dayan, 2003). While it is now widely recognized that habitat selection is a hierarchical process by which individuals choose what habitat to use at different scales, the significance of the fine-scale circadian variations in animals’ habitat use and associated movement patterns is still rarely discussed in the literature (Bestley et al., 2012; Godvik et al., 2009). This study addresses this issue, assessing movements and habitat use by the ‘near-threatened’ and Iberian endemic Cabrera vole (Microtus cabrerae) along the circadian cycle, in the dry and wet seasons. Our main hypothesis is that Cabrera voles show diel movement and habitat use patterns that are consistent with trade-offs between foraging needs and predator avoidance at fine spatial and temporal scales. To address this general hypothesis, we tested the following specific predictions: i) According to previous information on the species daily activity patterns (Pita et al., 2011a), Cabrera voles should spend more time moving, and move over larger distances in daytime, especially during the wet season; ii) Higher movement rates and distances should be associated with an increased use of open grass swards that provide the preferred foods for voles (e.g., Rosário et al., 2008), while periods of resting should be associated with habitats dominated by shrubs and tussockforming grasses that offer more cover against predators; and iii) According to the predominantly monogamous mating system of the species (e.g., Pita et al., 2014), the circadian variations in habitat use and movement patterns should not differ between sexes. Support for such behavioural patterns along the 24-h cycle may have important implications for local habitat management, particularly regarding the relative amount, and spatial arrangement of main vegetation types favouring the persistence of resident individuals. In addition, the patterns of movement and habitat use at fine spatio-temporal scales may provide useful behavioural indicators to infer species population status, and to monitor and predict their responses to environmental change (e.g., Caro, 2007; Sutherland, 1998).
out in summer, whereas permanent water bodies are scarce and mostly associated with irrigation infrastructure such as concrete channels and reservoirs (Ferreira and Beja, 2013; Pita et al., 2010). Over the past two decades, agricultural practices have intensified considerably, with largely negative consequences for farmland biodiversity (e.g., Beja and Alcazar, 2003; Beja et al., 2014; Ferreira and Beja, 2013; Pita et al., 2009). Study animals We used data from 25 radio-tracked adult Cabrera voles (mean ± se weight = 50.5 ± 1.39 g) exhibiting stable home-ranges (465.90 ± 69.03 m2 , based on 95% Kernel) encompassing mosaics of wet grassy and shrubby vegetation (see Pita et al., 2010, 2011a,b). Animals were surveyed within 11 colonies in two farmland areas in south-west Portugal (37◦ 58 –37◦ 54 N, 08◦ 48 –08◦ 44 W and 37◦ 38–37◦ 35 N, 08◦ 49 –08◦ 46 W), between April-2006 and April2008. Overall, 10 voles (7 females and 3 males) were tracked during the dry season (May-September, 4 colonies), and 15 voles (9 females and 6 males) were tracked during the wet season (October–April, 7 colonies). Details regarding capture and handling of animals are provided in Pita et al. (2010, 2011a,b). In brief, we used Sherman live-traps baited with apple and supplied with hay and hydrophobic cotton for bedding, with a mean ± se traps × days per sampled colony of 403.7 ± 84.2. At capture, voles were weighed, sexed, and fitted with collar radio-transmitters (Wildlife Materials, Inc., Murphysboro, IL, USA), adding no more than 5% of the animals weigh (Gannon and Sikes, 2007). All animals were lightly sedated with a subcutaneous injection of Dormitor® (0.2 mg/kg) to reduce handling stress. After transmitter attachment, voles were induced out of anaesthesia using an equivalent dose of Antisedam® . Before release, animals were kept under observation for at least 2 h to ensure they were suffering no ill effects or loss of mobility. Voles were released at their point of capture and radiotracking started at least 4 h after release, to avoid sampling atypical behaviours while animals became familiar with collars and recovered from stress. A total of 48% of collars were recovered from re-trapping by the end of surveys, with no signs of injuries on voles ever detected. In addition, 8% of collars were found within home ranges, presumably because voles managed to remove them. Another 16% of collars were found outside home ranges and showed damaged coats or cable ties, presumably because of predation. Direct evidence of predation was only observed in the case of one female consumed by a ladder snake, Elaphe scalaris, 7 days after collar attachment. The remaining collars (24%) were lost because of vole dispersal, battery discharge or untraced predation (Pita et al., 2011a). Sampling design and variables
Methods Study area The study was conducted in southwest Portugal, which is included in the thermo-Mediterranean bioclimatic zone and is characterized by strong seasonality in environmental conditions. Mean monthly temperature range between 9 and 23 ◦ C and average annual rainfall is around 680 mm, of which about 80% occurs between October and April (Sistema Nacional de Informac¸ão de Recursos Hídricos, http://snirh.pt). The region is predominantly flat and dominated by agriculture, with almost half of the land devoted to irrigated annual crops. Cattle production is also important, resulting in large areas occupied by pastures, fodder crops and silage corn or sorghum. Wood cover is restricted to arboreal windbreaks and a few woodlots. Surface waters are mostly associated with temporary ponds which flood during the rainy season and dry
Routine movement patterns and habitat use by voles within their home ranges were sampled in six different 4h-periods of the day covering the whole 24 h cycle, and thus reflecting different sunlight, climate and biotic conditions: 02 h15–06 h00 (late night/early morning); 06 h15–10 h00 (morning); 10 h15–14 h00 (noon); 14 h15–18 h00 (afternoon); 18 h15–22 h00 (evening); and 22 h15–02 h00 (night). Each animal was sampled in a different period every day along a mean ± se of 11.8 ± 1.6 days per vole (range: 3–23 days). During each sampling period, locations were carried out at 15-min intervals, totalling 16 locations per sampling period. Voles were located using a TRX-100S receiver and an external 3-element Yagi directional antenna (Wildlife Materials, Inc.). Locations were made by homing and by multiple triangulations when the tracker was close to the animals (about 50 cm). Voles were frequently seen during tracking, and appeared little affected by the presence of the observer. At each location, the GPS coordinates were recorded (±4 m error) along with the dominant vegetation
A.R. Grácio et al. / Mammalian Biology 83 (2017) 21–26
23
type within 50 cm-radius plots (Pita et al., 2010, 2011a,b). For this, two main vegetation types were considered: i) herbaceous swards dominated by wet grasses and forbs, usually >30 cm tall (hereafter grasses); and ii) shrubs (e.g., bramble Rubus thickets) or tussockforming herbs (e.g., sedges, rushes, or reeds), usually >60 cm tall (hereafter shrubs). From these data we extracted, for each sampling period and animal, the following dependent variables: i) Proportion of time spent moving (PM), given by the number of times the animal moved in each period divided by the total number of observations; ii) Overall movement length (OM), given by summing up the distances travelled between successive point locations; and iii) Grass selectivity index (GS), given as the frequency of point locations (habitat used) dominated by grassy vegetation relative to respective overall frequency (habitat available), recorded for each animal throughout overall sampling time. The latter corresponds to the Ivlev selec˜ et al., 2010), and measures the strength tivity index (e.g., Magana of third order selection of grasses relative to availability (Johnson, 1980).
Table 1 Ranking of the models for each season, specifically the ‘null’ models and the ‘circadian’ models (including the period of the day as a fixed effect).
Data analyses
Results
To investigate whether the period of day affected the three variables describing vole movements and habitat use in each season, we used mixed-effects multivariate analysis of variance (MANOVAs), which is a generalization of analysis of variance (ANOVA) that allows the analysis of multiple dependent variables, while accounting for possible correlation among them (Bray and Maxwell, 1985). Survey years, colonies, and animals were included as random intercept effects, i.e. grouping variables with random variations around an overall population mean (see e.g. Gillies et al., 2006). To reduce the influence of extreme values and enhance normality of model residuals, the scaled dependent variables were arcsin-transformed in the case of the proportion of time spent moving, and logtransformed in the case of the overall movement length (e.g. Wilson et al., 2010). This also resulted in improved model mixing and convergence. In model building, the intercept was suppressed as suggested by Hadfield (2016). Analyses were conducted for each season separately, because previous studies have shown considerable seasonal differences in both circadian activity rhythms and habitat selection by Cabrera voles (Pita et al., 2011a,b). For each season we first built a ‘circadian’ model, in which the period of the day was included as a fixed effect. In order to assess the support of the ‘circadian’ models, we then built concurrent ‘null’ models (i.e. including only the random effects), and checked whether the ‘circadian’ models yielded deviance information criterion (DIC) at least 5 units lower than that of the corresponding ‘null’ models (Spiegelhalter et al., 2002). To test the importance of the sex in the circadian variation of habitat use and movement patterns, we built an additional model for each season including the interaction between the period of the day and the sex of tracked voles (‘circadian*sex’ model). We also assessed eventual differences in mean circadian movement patterns and habitat use between males and females, by running a further model including the main effect of sex (‘circadian + sex’ model). Support for any of these models was assessed by checking whether the corresponding DIC values were at least 5 units lower than that of the simple ‘circadian’ models. Finally, model fit of the most supported models was estimated using pseudo-R2 (Johnson, 2014), and the respective posterior distributions were used to estimate correlations among the three dependent variables in each season (e.g., Hadfield, 2010; Wilson et al., 2010), with significant correlations determined by the 95% credible intervals not overlapping with zero. All models were run with Bayesian Markov Chain Monte Carlo simulation using the package ‘MCMCglmm’ version 2.22 (Hadfield, 2010) and the program R 3.1.3 (R Core Team, 2016), keeping
A total of 3,664 location records (2,080 in the wet season and 1,584 in the dry season) were obtained for the 25 voles radio tracked, with a mean ± se of 146.56 ± 14.68 locations per animal. This corresponded to a total of 130 and 99 experimental units (4-h periods) sampled in the wet and dry seasons respectively. Despite the variability observed in individual responses (see Fig. SM1, Supplementary material), there were some differences in the mean values of the dependent variables estimated among the six different 4-h sampling periods considered, for both seasons. In general, voles tended to spend more time moving and moved over larger distances in daytime, which is also when they used grassy areas more often (Table SM1, Supplementary material). This was confirmed by the mixed-effects MANOVA, which showed that the ‘circadian’ models for each season were strongly supported when compared to the respective ‘null’ models (Table 1). In particular, using the period between 02 h15–06 h00 as baseline, posterior estimates of model coefficients regarding the proportion of time animals spent moving during the wet season were significantly higher between 06 h15 and 22 h00, with the highest values between 10 h15 and 18 h00 (noon and afternoon) (Table SM2, Supplementary material), resulting in higher predicted estimates during these periods (Fig. 1a). These results were generally also observed during the dry season, though the highest posterior estimates of model coefficients were recorded between 06 h15–10 h00 (morning) (Table SM2, Supplementary material; Fig. 1b). Very similar patterns were found for both the overall movement length (Fig. 1c–d) and the strength of selection of grassy areas over shrubs, with significantly higher values between 06 h15 and 22 h00 during both seasons (Table SM2, Supplementary material; Fig. 1e–f). Overall movement length was higher between 14 h15–18 h00 during the wet season (afternoon), and between 06 h15–10 h00 (morning) during the dry season (Table SM2, Supplementary material; Fig. 1c–d). In general, the selection of grasses was particularly strong between 06 h15–10 h00 during both seasons (Table SM1, Supplementary material; Fig. 1e–f). Despite the general consistency in diel movement patterns and habitat use by voles in each season, the three random effects considered in the models, while not being explanatory, also introduced variation influencing response variables. In general, posterior mean estimates of explained variation suggested a greater importance of inter-annual variation, followed by inter-colony and interindividual variation (see Table SM2, Supplementary material). The ‘circadian*sex’ models yielded much higher DIC values than the simple ‘circadian’ model for both the wet (DIC = 22.73) and
Season
Model
DIC
DIC
Adjusted pseudo R2
Wet
‘null’ model ‘circadian’ model
666.49 634.66
31.83
– 0.39
Dry
‘null’ model ‘circadian’ model
661.55 609.67
51.88
– 0.51
>1000 posterior samples (Hadfield, 2016). Models were run using default uninformative flat priors, considering acceptable levels of first order autocorrelation (generally <0.01 for successive iterations, Hadfield, 2010; Plummer et al., 2006) and until they reached convergence, as assessed computationally using Geweke’s convergence diagnostic (Plummer et al., 2006). This implied runs with 20,000 iterations, burn-in size of 2000 and thinning interval of 10 iterations. Adjusted pseudo-R2 values were estimated with ´ 2016). ‘MuMin’ (Barton,
24
A.R. Grácio et al. / Mammalian Biology 83 (2017) 21–26
Fig. 1. Posterior estimates and 95% confidence intervals of predicted responses in each of the six 4-h periods of the day considered, according to the ‘circadian’ models for the wet and dry seasons: Proportion of time spent moving (a and b); overall movement length (c and d); and grass selectivity index (e and f).
Table 2 Correlations among the proportion of time spending moving (PM), overall movement length (OM) and grass selectivity index (GS) extracted from the ‘circadian’ models for each season. Season
Variables
Correlation(posterior mode)
95% CI
Wet
PM vs. OM PM vs. GS OM vs. GS
0.73 0.48 0.39
0.65–0.84 0.29–0.62 0.23–0.58
Dry
PM vs. OM PM vs. GS OM vs. GS
0.78 0.31 0.41
0.67–0.85 0.13–0.56 0.20–0.56
dry (DIC = 14.97) seasons, indicating that sex had little effect on circadian movement and habitat use patterns (see Table SM3 in Supplementary material). The ‘circadian + sex’ models also yielded higher DIC values than the ‘circadian’ models, though the respective DIC were lower than 5 in both the wet (DIC = 2.97) and dry (DIC = 0.47) seasons. These models suggested some tendency for sex differences in movement activity and habitat use, though these were never significant for either season (see Table SM4 in Supplementary material). There were significant pairwise correlations between all dependent variables estimated using the posterior distributions of the ‘circadian’ models for both seasons (Table 2). Discussion Patterns of habitat use and movement are key topics in ecological and conservation studies, because they affect animals’ distribution at different temporal and spatial scales, and can strongly influence their fitness and survival (e.g., Morales et al., 2010; Russell et al., 2003). These patterns are usually suggested to result from trade-offs between food acquisition efficiency and mor-
tality risk, which are expected to vary across the 24-h cycle for most taxa, according to endogenous biological rhythms and external environmental factors (Railsback et al., 2005). Although seasonal variation in resource selection and circadian activity rhythms of Cabrera voles have been reported previously (Pita et al., 2011a,b; Santos et al., 2005), our study provides the first integrated approach showing the link between habitat use and movement patterns within individuals home-ranges, and how joint decisions of what habitat to occupy and how to move within it may reflect circadian variations in foraging versus hiding/resting activities across seasons (Pépin et al., 2009; Tigas et al., 2002). Overall, results were consistent with our first prediction that, despite the likely influence of inter-annual, inter-patch and interindividual variations, Cabrera voles should spend more time moving and move over larger distances in daytime than at night. This unimodal (diurnal) movement pattern was particularly evident during the wet season, and it is in line with previous studies on the circadian activity pattern of this species (Pita et al., 2011a). Also in line with previous research, the movement during the dry season was largely bimodal (crepuscular), peaking in the morning (06 h15–10 h00), declining at mid-day (10 h15–14 h00), and then increasing again in the evening (18 h15–22 h00), while it was minimal at night (22 h15–06 h00). The mid-day decline in movement during the dry season probably reflects a behavioural response to minimize exposure to high temperatures, helping individuals to cope with physiological and metabolic needs during the hottest hours (Mathias et al., 2003; Pita et al., 2011b). Our second prediction that diel movement and habitat use patterns are linked and cyclic was also confirmed, as results showed that increased movements during the day were associated with a higher use of grassy areas, where individuals were probably foraging, while reduced movements were associated with the use of shrubby areas, where
A.R. Grácio et al. / Mammalian Biology 83 (2017) 21–26
individuals were probably resting. Considering that individuals select the combination of habitat and movement that maximizes their fitness (e.g., Rhodes et al., 2005), it is likely that these temporal shifts were related to a higher forage abundance and quality of grassy areas to fulfil food intake requirements while foraging, and a higher value of shrubby areas for reducing predation risk (LuqueLarena and López, 2007; Pita et al., 2006; Rosário et al., 2008). Resting in shrubby areas should be particularly favourable, because shrubs and tussock-forming herbs may provide concealment and some protection to voles in periods when vigilance is presumably low and thus vulnerability might be high (Ganskoop and Johnson, 2007; Semeniuk and Dill, 2005). Finally, there was support for our prediction that male and female Cabrera voles should have similar circadian patterns of movement and habitat use. This is probably a consequence of the species’ predominantly monogamous mating system and biparental care (e.g., Fernández-Salvador et al., 2001; Pita et al., 2014), which implies that both sexes likely share much the same time intervals to search for feeding areas and to secure resting and nesting. Taken together, results from this study support the view that the way animals use their environment depends on the characteristics of the resources available at fine scales, and that circadian variations in environmental conditions may impose differences in individuals behavioural choices related to movement patterns and habitat use (e.g., Bennitt et al., 2015). In particular, our study showed that within their home-ranges, Cabrera voles may distinguish between particular vegetation types based on their structural and functional value, which in turn affect their suitability for particular behaviours along the 24 h cycle. While we acknowledge that replicating surveys across a larger number of animals, patches and years would have probably increased the robustness and consistency of our findings, we believe that the relationships uncovered here provide useful information for a more complete understanding of movement activity and habitat use by Cabrera voles. This highlights the need to include individual behavioural traits when analysing resource use by the species at the population level, as some vegetation types may be more suited for particular behaviours than others (e.g., Bjørneraas et al., 2011; Dzialak et al., 2012). Implications for conservation and monitoring Although we cannot exclude the influence of other sources of variation not considered here (e.g., geographic variation, and occurrence of predators and intra- and interspecific abundance at local scales) (e.g., Ager et al., 2003), our study clearly shows that the period of the day is an important driver of changes in movement and habitat use of Cabrera voles. Therefore, based on the patterns found, a set of recommendations may be posited for improving local conservation management and monitoring programs targeting the species. In particular, our findings suggest the need to maintain or create potential habitat patches composed of a mosaic of wet grasses and shrubs or tussock-forming herbs, likely to guarantee high food intake, while minimizing foraging space and travel time to safer sites during voles’ routine activities within their homeranges (e.g., Ferron and Ouellet, 1992). This may require occasional management of local habitat patches, such as for instance shrub control to prevent encroachment into grassy, food-rich areas (Pita et al., 2007). It should be noted, however, that possible management actions should take into account the dynamic nature of habitat selection and movement patterns by the species across the diel cycle. Another important contribution of our study is that, by analysing the fine-scaled shifts in diel movement and habitat selection of adult individuals with stable home-ranges, it provides baseline information for future research aiming to assess individual fitness after disturbances, or translocation programs, as well as to identify
25
deviations from the expected behaviours at particular life-history stages such as dispersal or colonization (Caro, 2007; Sutherland, 1998). Such behavioural indicators of individual conditions may in turn be used to infer the overall population status at early stages, i.e. before demographic indicators become noticed (e.g., Haroldson and Fritzell, 1984; Onorato et al., 2011; Di Stefano et al., 2009). The circadian changes in Cabrera voles’ movement and habitat use shown here may also help refining the sampling and monitoring programs focusing on particular behaviours or ecological processes, namely in what concerns the choice of the temporal scale and optimal allocation of resources for sampling. For instance, in telemetrybased studies focusing on foraging voles and their interactions with predators, it may be preferred to sample more often during daytime or more seldom during night time, according to the species diel behavioural patterns (e.g., Moe et al., 2007). Overall, therefore, our study illustrates the practical utility of incorporating fine-scaled shifts in animal behaviour across the diel cycle as part of spatial prediction of species persistence ability and overall state of populations, thereby contributing to improving species conservation and monitoring programs (e.g., Berger-Tal et al., 2011; Kotler et al., 2016). Acknowledgements This study was funded by the European Regional Development Fund (ERDF) through the Operational Program for Competitiveness Factors (COMPETE), and National Funds through the Foundation for Science and Technology (FCT), under the projects ‘NetPersist’ (PTDC/AAG-MAA/3227/2012) and ‘MateFrag’ (PTDC/BIA-BIC/6582/2014). RP was supported by the grants SFRH/BPD/73478/2010 and SFRH/BPD/109235/2015. PB was supported by the EDP Biodiversity Chair. Capture and handling of voles were carried-out under the permission of the Portuguese nature conservation authority (ICNF) and were supervised by Dr. Manuel Mestre from ZooMédica Veterinary Clinic. All procedures conformed to the guidelines approved by the American Society of Mammalogists for the use of wild mammals in research. We thank R. Brito and M. Duarte for help during field surveys. We also thank Julian Di Stefano and two anonymous reviewers for helpful comments to improve the paper. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.mambio.2016.11. 008. References Ager, A.A., Johnson, B.K., Kern, J.W., Kie, J.G., 2003. Daily and seasonal movements and habitat use by female rocky mountain elk and mule deer. J. Mammal. 84, 1076–1088. ´ K., 2016. Multi-Model Inference. Package MuMIn. https://cran.r-project. Barton, org/web/packages/MuMIn/index.html. Beja, P., Alcazar, R., 2003. Conservation of Mediterranean temporary ponds under agricultural intensification: an evaluation using amphibians. Biol. Conserv. 114, 317–326. Beja, P., Schindler, S., Santana, J., Porto, M., Morgado, R., Moreira, F., Pita, R., Mira, A., Reino, L., 2014. Predators and livestock reduce bird nest survival in intensive Mediterranean farmland. Eur. J. Wildl. Res. 60, 249–258. Bennitt, E., Bonyongo, M.C., Harris, S., 2015. Behaviour-related scalar habitat use by cope buffalo (Syncerus caffer caffer). PLoS One 10, e0145145. Berger-Tal, O., Polak, T., Oron, A., Lubin, T., Kotler, B.P., Saltz, D., 2011. Integrating animal behavior and conservation biology: a conceptual framework. Behav. Ecol. 22, 236–239. Bestley, S., Jonsen, I.D., Hindell, M.A., Guinet, C., Charrassin, J.B., 2012. Integrative modelling of animal movement: incorporating in situ habitat and behavioural information for a migratory marine predator. Proc. R. Soc. B: Biol. Sci. 280, 20122262. Bjørneraas, K., Solberg, E.J., Herfindal, I., Moorter, B.V., Rolandsen, C.M., Tremblay, J.-P., Skarpe, C., Sæther, B.-E., Eriksen, R., Astrup, R., 2011. Moose Alces Alces
26
A.R. Grácio et al. / Mammalian Biology 83 (2017) 21–26
habitat use at multiple temporal scales in a human-altered landscape. Wildl. Biol. 17, 44–54. Bray, J.H., Maxwell, S.E., 1985. Multivariate Analysis of Variance. SAGE Publications California, Newbury Park, CA. Caro, T., 2007. Behavior and conservation: a bridge too far? Trends Ecol. Evol. 22, 394–400. Di Stefano, J., York, A., Swan, M., Greenfield, A., Coulson, G., 2009. Habitat selection by the swamp wallaby (Wallabia bicolour) in relation to diel period, food and shelter. Austral. Ecol. 34, 143–155. Dzialak, M.R., Olson, C.V., Harju, S.M., Webb, S.L., Winstead, J.B., 2012. Temporal and hierarchical spatial components of animal occurrence: conserving seasonal habitat for greater sage-grouse. Ecosphere 3, 30. Fernández-Salvador, R., García-Perea, R., Ventura, J., 2001. Reproduction and postnatal growth of the Cabrera vole, Microtus cabrerae, in captivity. Can. J. Zool. 79, 2080–2085. Ferreira, M., Beja, P., 2013. Mediterranean amphibians and the loss of temporary ponds: are there alternative breeding habitats? Biol. Conserv. 165, 179–186. Ferron, J., Ouellet, J.-P., 1992. Daily partitioning of summer habitat and use of space by the snowshoe hare in southern boreal forest. Can. J. Zool. 70, 2178–2183. Gannon, W.L., Sikes, R.S., 2007. Animal care and use committee of the american society of mammalogists guidelines of the american society of mammalogists for the use of wild mammals in research. J. Mammal. 88, 809–823. Ganskoop, D.C., Johnson, D.D., 2007. GPS error in studies addressing animal movements and activities. Rangel. Ecol. Manag. 60, 350–358. Gillies, C.S., Hebblewhite, M., Nielsen, S.E., Krawchuk, M.A., Aldridge, C.L., Frair, J.L., Saher, D.J., Stevens, C.E., Jerde, C.L., 2006. Application of random effects to the study of resource selection by animals. J. Anim. Ecol. 75, 887–898. Godvik, I.M.R., Loe, L.E., Vik, J.O., Veiberg, V., Langvatn, R., Mysterud, A., 2009. Temporal scales, trade-offs, and functional responses in red deer habitat selection. Ecology 90, 699–710. Hadfield, J.D., 2010. MCMC methods for multi–response generalised linear mixed models: the MCMCglmm R package. J. Stat. Softw. 33, 1–22. Hadfield, J., 2016. MCMC Generalised Linear Mixed Models. Package MCMCglmm (Accessed 9 June 2016) https://cran.r-project.org/web/packages/MCMCglmm/ index.html. Haroldson, K.J., Fritzell, E.K., 1984. Home ranges, activity, and habitat use by gray foxes in an oak-hickory forest. J. Wildl. Manag. 48, 222–227. Horton, T.H., 2001. Conceptual issues in the ecology and evolution of circadian rhythms. In: Takahashi, J.S., Turek, F.W., Moore, R.Y. (Eds.), Handbook of Neurobiology: Circadian Clocks 12. Kluwer Academic/Plenum Publishers, New York, pp. 45–57. Johnson, D.H., 1980. The comparison of usage and availability measurements for evaluating resource preference. Ecology 61, 65–71. Johnson, P.C.D., 2014. Extension of Nakagawa & Schielzeth’s R2 GLMM to random slopes models. Method Ecol. Evol. 5, 44–946. Kotler, B.P., Morris, D.W., Brown, J.S., 2016. Direct behavioral indicators as a cconservation and management tool. In: Berger-Tal, O., Saltz, D. (Eds.), Conservation Behaviour: Applying Behavioural Ecology to Wildlife Conservation and Management. Cambridge University Press, Cambridge, pp. 307–353. Kronfeld-Schor, N., Dayan, T., 2003. Partitioning of time as an ecological resource. Annu. Rev. Ecol. Syst. 34, 153–181. Lima, S.L., Dill, L.M., 1990. Behavioural decisions made under the risk of predation. Can. J. Zool. 68, 619–640. Luque-Larena, J.J., López, P., 2007. Microhabitat use by wild-ranging Cabrera voles Microtus cabrerae as revealed by live trapping. Eur. J. Wildl. Res. 53, 221–225. ˜ M., Alonso, J.C., Martin, C.A., Bautista, L.M., Martin, B., 2010. Nest-site Magana, selection by Great Bustards Otis tarda suggests a trade-off between concealment and visibility. Ibis 152, 77–89. Manly, B., McDonald, L., Thomas, D., MacDonald, T., Erickson, W., 2002. Resource Selection by Animals. Statistical Design and Analysis for Field Studies. Kluwer Academic Publisher, Springer, Neaderlands. Mathias, M.L., Klunder, M., Santos, S.M., 2003. Metabolism and thermoregulation in the Cabrera vole (Rodentia: Microtus cabrerae). Comp. Biochem. Phys. A 136, 441–446. Mayor, S.J., Schneider, D.C., Schaefer, J.A., Mahoney, S.P., 2009. Habitat selection at multiple scales. Écoscience 16, 238–247. Moe, T.F., Kindberg, J., Jansson, I., Swenson, J.E., 2007. Importance of diel behaviour when studying habitat selection: examples from female Scandinavian brown bears (Ursus arctos). Can. J. Zool. 85, 518–525.
Morales, J.M., Moorcroft, P.R., Matthiopoulos, J., Frair, J.L., Kie, J.G., Powell, R.A., Merrill, E.H., Haydon, D.T., 2010. Building the bridge between animal movement and population dynamics. Philos. Trans. R. Soc. B 365, 2289–2301. Morris, D.W., 2014. Can foraging behaviour reveal the eco-evolutionary dynamics of habitat selection? Evol. Ecol. Res. 16, 1–18. Onorato, D.P., Criffield, M., Lotz, M., Cunningham, M., McBride, R., Leone, E.H., Bass Jr, O.L., Hellgren, E.C., 2011. Habitat selection by critically endangered Florida panthers across the diel period: implications for land management and conservation. Anim. Conserv. 14, 196–205. Ordiz, A., Kindberg, J., Sæbø, S., Swenson, J.E., Støen, O.-Gunnar, 2014. Brown bear circadian behaviour reveals human environmental encroachment. Biol. Conserv. 173, 1–9 (Elsevier). Pépin, D., Morellet, N., Goulard, M., 2009. Seasonal and daily walking activity patterns of free-ranging adult red deer (Cervus elaphus) at the individual level. Eur. J. Wildl. Res. 55, 479–486. Pita, R., Mira, A., Beja, P., 2006. Conserving the Cabrera vole, Microtus cabrerae, in intensively used Mediterranean landscapes. Agric. Ecosyst. Environ. 115, 1–5. Pita, R., Beja, P., Mira, A., 2007. Spatial population structure of the Cabrera vole in Mediterranean farmland: the relative role of patch and matrix effects. Biol. Conserv. 134, 383–392. Pita, R., Mira, A., Moreira, F., Morgado, R., Beja, P., 2009. Influence of landscape characteristics on carnivore diversity and abundance in Mediterranean farmland. Agric. Ecosyst. Environ. 132, 57–65. Pita, R., Mira, A., Beja, P., 2010. Spatial segregation of two vole species (Arvicola sapidus and Microtus cabrerae) within habitat patches in a highly fragmented farmland landscape. Eur. J. Wildl. Res. 56, 651–662. Pita, R., Mira, A., Beja, P., 2011a. Circadian activity rhythms in relation to season, sex and interspecific interactions in two Mediterranean voles. Anim. Behav. 81, 1023–1030. Pita, R., Mira, A., Beja, P., 2011b. Assessing habitat differentiation between coexisting species: the role of spatial scale. Acta Oecol. 37, 124–132. Pita, R., Mira, A., Beja, P., 2014. Microtus cabrerae (Rodentia: Cricetidae). Mamm. Species 46, 48–70. Plummer, M., Best, N., Cowles, K., Vines, K., 2006. CODA: convergence diagnosis and output analysis for MCMC. Newsl. R. Proj. 6, 7–11. R Core Team, 2016. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria (Accessed 9 June 16) http://www.R-project.org/. Railsback, S.F., Harvey, B.C., 2002. Analysis of habitat-selection rules using an individual-based model. Ecology 83, 1817–1830. Railsback, S.F., Harvey, B.C., Hayse, J.W., LaGory, K.E., 2005. Tests of theory for diel variation in salmonid feeding activity and habitat use. Ecology 86, 947–959. Rhodes, J.R., McAlpine, C.A., Lunney, D., Possingham, H.P., 2005. A spatially explicit habitat selection model incorporating home range behavior. Ecology 86, 1199–1205. Rosário, I.T., Cardoso, P.E., Mathias, M.L., 2008. Is habitat selection by the Cabrera vole (Microtus cabrerae) related to food preferences? Mamm. Biol. 73, 423–429. Russell, R.E., Swihart, R.K., Feng, Z., 2003. Population consequences of movement decisions in a patchy landscape. Oikos 103, 142–152. Santos, S.M., Rosário, I.T., Mathias, M.L., 2005. Microhabitat preference of the Cabrera vole in a Mediterranean cork oak woodland of southern Portugal. Vie et Milieu 55, 53–59. Semeniuk, C.A.D., Dill, L.M., 2005. Cost/benefit analysis of group and solitary resting in the cowtail stingray, Pastinachus sephen. Behav. Ecol. 16, 417–426. Spiegelhalter, D.J., Best, N.G., Carlin, B.P., Van Der Linde, A., 2002. Bayesian measures of model complexity and fit. J. R. Stat. Soc. B 64, 583–639. Sutherland, W.J., 1998. The importance of behavioural studies in conservation biology. Anim. Behav. 56, 801–809. Tigas, L.A., Vuren, D.H.V., Sauvajot, R.M., 2002. Behavioral respondes of bobcats and coyotes to habitat fragmentation and corridors in an urban environmental. Biol. Conserv. 108, 299–306. Verdolin, J.L., 2006. Meta-analysis of foraging and predation risk trade-offs in terrestrial systems. Behav. Ecol. Sociobiol. 60, 457–464. Wilson, A.J., Réale, D., Clements, M.N., Morrissey, M.M., Postma, E., Walling, C.A., Kruuk, L.E.B., Nussey, D.H., 2010. An ecologist’s guide to the animal model. J. Anim. Ecol. 79, 13–26.