This is an author version of the contribution published in:
Questa è la versione dell’autore dell’opera:
Oecologia, 2016, 10.1007/s00442-016-3637-y
The definitive version is available at:
La versione definitiva è disponibile alla URL:
http://link.springer.com/article/10.1007/s00442-016-3637-y
1 2 3 4 5 6 7 8 9 10 11 12 13Alpine bird distributions along elevation gradients: the consistency of climate
and habitat effects across geographic regions
Dan Chamberlain
1, Mattia Brambilla
2,3, Enrico Caprio
1, Paolo Pedrini
2&
Antonio Rolando
11
Dipartimento di Scienze della Vita e Biologia dei Sistemi, Università di Torino, Via
Accademia Albertina 13, 10123, Torino, Italy.
2
Museo delle Scienze di Trento, Sezione Zoologia dei Vertebrati, Corso del Lavoro e della
Scienza 3, 38122 Trento, Italy
3
Fondazione Lombardia per l'Ambiente, Settore biodiversità e aree protette, Largo 10 luglio
1976, 1 I-20822 Seveso (MB), Italy
Author contribution statement – All authors contributed significantly to the conception and planning
of the paper. DEC took a lead on analysing the data and writing the paper. MB and EC carried out data extraction in GIS. PP and AR managed fieldwork in Trentino and Piedmont respectively. 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36
Abstract Many species have shown recent shifts in their distributions in response to climate change.
Patterns in species occurrence or abundance along altitudinal gradients often serve as the basis for detecting such changes and assessing future sensitivity. Quantifying the distribution of species along altitudinal gradients acts as a fundamental basis for future studies on environmental change impacts, but in order for models of altitudinal distribution to have wide applicability, it is necessary to know the extent to which altitudinal trends in occurrence are consistent across geographically separated areas. This was assessed by fitting models of bird species occurrence across altitudinal gradients in relation to habitat and climate variables in two geographically separated Alpine regions, Piedmont and Trentino. The ten species studied showed non-random altitudinal distributions which in most cases were consistent across regions in terms of pattern. Trends in relation to altitude and
differences between regions could be explained mostly by habitat or a combination of habitat and climate variables. Variation partitioning showed that most variation explained by the models were attributable to habitat, or habitat and climate together, rather than climate alone or geographic region. The shape and position of the altitudinal distribution curve is important as it can be related to vulnerability where the available space is limited, i.e. where mountains are of not sufficient altitude for expansion. This study therefore suggests that incorporating habitat and climate variables should be sufficient to construct models with high transferability for many Alpine species.
Keywords Altitude, Climate change, Model performance, Model transferability, Variation
partitioning 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57
Introduction
Many species have shown recent shifts in their distributions in response to environmental change, in particular climate change (Parmesan and Yohe 2003; Chen et al. 2011), mostly towards higher latitudes and/or altitudes (Walther et al. 2002). Investigating the effect of climate change on biodiversity and ecosystems has thus become a key ecological research area, often underpinned by modelling approaches that seek to determine relationships between species occurrence or
population dynamics and climate, and to predict the future response to climate change (Bellard et al. 2012). Such approaches have been frequently applied to species distributions, which may be affected by a range of factors, but in particular by climatic variation and habitat availability. The effect of environmental factors such as climate, topography and land-cover are often considered for modelling species distributions (Guisan and Thuiller 2005), typically using correlative models, which relate the occurrence of a species to a set of environmental predictors, allowing for the re-projection of species occurrence under new, future environmental conditions. A critical issue in this approach is
represented by the extent to which a given model can be generally applied over different spatial and temporal contexts (i.e. model transferability). As it is impossible to test predictions on future data, model performance could be evaluated by means of a space-for-time substitution (Araujo and Rahbek 2006), using data from different regions and cross-checking models (e.g. Randin et al. 2006).
In mountain environments, where species distributions are often limited by temperature, increased warming has been accompanied by upward shifts in the distributions of many species, e.g. plants (Lenoir et al. 2008; Harsch et al. 2009), butterflies (Wilson et al. 2005), birds (Tryjanowski et al. 2005; Reif and Flousek 2012) and small mammals (Moritz et al. 2008). For several cold-adapted species, and in particular for those living in high altitude open habitats, such shifts may lead to a reduction in range as areas of suitable climate and habitat become smaller and more fragmented as they are pushed towards mountain summits. Such effects may have serious consequences for mountain biodiversity in the future (Sekercogliu et al. 2008; Dirnbock et al. 2011; Chamberlain et al. 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83
2013; Maggini et al. 2014; Brambilla et al. 2015), and indeed there is evidence that birds of high altitude are already showing declines (Lehikoinen et al. 2014, Flousek et al. 2015). It should, however, also be acknowledged that mountain biodiversity may be under other anthropogenic pressures (Chamberlain et al. 2016), such as changes in grazing regimes (Laiolo et al. 2004) and increasing disturbance (Caprio et al. 2011), although evidence for effects of these factors, either positive or negative, on mountain bird population trends is so far lacking.
Patterns in species occurrence or abundance along altitudinal gradients often serve as the basis for detecting changes (e.g. Maggini et al. 2011; Pernollet et al. 2015) and assessing future sensitivity (Chamberlain et al. 2013) of mountain species to climate change. Generally speaking, investigating elevational range limits is critical to understanding distributional patterns, and is needed to predict the likely effects of (and responses to) climate change in mountain species (Gifford and Kozak 2012). The altitudinal transect approach is useful for studying potential climatic effects on species distributions, because the altitudinal gradient provides a space-for-time substitution when considering conditions along the gradient (Hodkinson 2005), and also, complications involving broader-scale biogeographic processes, evident in geographic distribution shift studies, are largely avoided (Rahbek, 2005). However, given that conditions change rapidly over fine spatial scales along altitudinal gradients, data collected needs to be of a sufficiently high-resolution to be useful for monitoring and modelling distribution shifts (Chamberlain et al. 2012). In areas with strong altitudinal gradients, the use of models developed at finer spatial scales is required to avoid overestimation of habitat loss due to climate change (Randin et al. 2009).
Birds are undoubtedly a well-studied group in terms of impacts of environmental change generally. However, relative to other habitats, the factors that dictate bird distributions, population size and population trend in mountains are less well-known (EEA 2010), largely due to the logistical constraints of working in such an environment (Chamberlain et al. 2012). Even basic, but
nonetheless essential, information on species distributions along altitudinal gradients is scarce. In Europe, there is very little information on variation in bird distributions along altitudinal gradients in 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109
mountains, with a few exceptions (ring ouzel Turdus torquatus, marsh tit Poecile palustris and bullfinch Pyrrhula pyrrhula, Maggini et al. 2011; water pipit Anthus spinoletta, Melendez and Laiolo 2014; ptarmigan Lagopus muta, Pernollet et al. 2015). There is therefore a need to quantify the distribution of more species along altitudinal gradients in order to act as a fundamental basis for future studies on environmental change impacts. Furthermore, if models of altitudinal distribution are to be used for drawing inferences on wider impacts of environmental change, then it is necessary to know the extent to which altitudinal trends in occurrence are consistent across geographically separated areas and therefore the extent to which a model derived from one area can be used to make predictions in another (i.e. model transferability, Whittingham et al. 2007; Schaub et al. 2011). Finally, it would be useful to know whether relatively simple models based on altitude alone are sufficient to describe bird distributions along the gradient, or whether environmental variables (habitat and climate) are essential elements to modelling elevational distributions of birds.
This paper has three aims: (1) to describe the distributions of birds along altitudinal gradients in the European Alps at relatively high altitude (c. 1700-3100m) and to determine if they vary
between two geographically separated regions; (2) to assess the performance of models derived across the whole study area in order to determine whether bird distributions can be better explained by variations in altitude, habitat cover or climate, or combinations of these, along the gradient; (3) to assess the extent of unexplained variation attributable to regional differences (i.e. whether combinations of habitat and climate are sufficient to explain regional differences).
Materials and methods
Fieldwork was undertaken in two geographically distinct Alpine areas in northern Italy, Piedmont and Trentino (Fig 1). In both areas, dominant shrub species are typically juniper Juniperus communis and
Rhododendron spp. The natural treeline occurs at around 2200-2300m, although this varies
110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135
depending on local conditions. Furthermore, in many areas, the treeline is lower due to impacts of livestock grazing. Grasslands occur throughout both areas, consisting of seasonal pastures and higher altitude alpine meadows. Scree and rocky areas are common, especially at higher altitudes, and are typically dominant above c. 2700m. In Piedmont, the dominant tree species is European larch Larix
decidua whereas in Trentino it is spruce Picea abies. In general, trends in the cover of major habitat
types were similar between the two regions, although there were some notable differences, e.g. greater forest cover at lower altitudes (c. 1700-1900m), and higher rock cover throughout the gradient, in Trentino (Fig. S1 in the “Electronic Supplementary Material”). There was little difference in temperature between the two regions, although there was markedly higher precipitation in Piedmont (Fig. S2 in the “Electronic Supplementary Material”).
Bird and environmental data
Sampling took place over three years in Piedmont (2010-2012), and in a single year, 2011, in Trentino. Identical field survey methods were undertaken in each region. Point counts were carried out along transects, which were selected based on accessibility, and which were usually (although not always) along footpaths. Transects were separated by at least 1-km. A minimum altitude of 1700m was defined in both areas. In the field, the start of each transect was the closest suitable point above 1700m in altitude. Suitable points were those without any obvious disturbance (e.g. occupied human habitation, livestock) or where detectability may have been affected (e.g. large cliffs, noisy streams in spate) within 100m. Point locations were recorded on a handheld Garmin GPS. The next selected point along the transect was then the next suitable location after a minimum distance of 200m (i.e. to ensure no adjacent points were overlapping). Whilst a random selection of points was not practically feasible, this systematic technique at least avoided any possible selection based on the birds themselves.
136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160
Point counts (Bibby et al. 2000) were carried out from mid-May to mid-July, using a ten minute count period preceded by a five minute settling period. At each point, the observer recorded all birds seen and heard within a 100m radius (estimated with the aid of a laser range finder). Simple habitat data were also collected at each point, including the percentage cover of canopy (i.e.
vegetation above head height), shrubs (woody vegetation below head height), open grassland (i.e. no canopy), bare rock (including scree and other unvegetated areas) within 100m radius, and the number of mature trees (approximately greater than 20cm in diameter) within a 50m radius (in forested areas, it was not possible to count trees at a greater distance). Point counts commenced 1 to 1.5 hours after sunrise and continued until 13:00. No surveys were undertaken in wet or
excessively windy conditions. A total of 453 points were surveyed on 56 transects, 271 points from 34 transects in Piedmont and 182 points from 22 transects in Trentino.
Climate data were extracted from WorldClim (Hijmans et al. 2005), including five temperature variables and five precipitation variables. The temperature variables were: mean annual monthly temperature, maximum and minimum monthly temperature over the whole year, mean monthly temperature for the breeding season and mean monthly temperature for the winter. The precipitation variables were: total annual precipitation, maximum and minimum monthly precipitation over the whole year, mean monthly precipitation for the breeding season and mean monthly precipitation for the winter. Topographic data (aspect and slope) were extracted from a Digital Terrain Model of northern Italy at a 1ha scale, and calculated at the point level by calculating mean values of the squares overlapping each 100m radius point count location. Both easting and northing were considered, which were expressed as an index equal to –cos(A), where A is the aspect (east or south) expressed in radians (following Bradbury et al. 2011), where a value of 1 represents facing directly south or east, and -1 represents facing directly north or west. Altitude (expressed in m) at each point was recorded by the GPS in the field. Slope was measured in degrees. (A full list of environmental variables considered in the modelling procedure is presented in Table S1 in the 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185
“Electronic Supplementary Material”). All variables were standardized to a mean of zero prior to analysis. All analyses were carried out in R 3.01 (R Development Core Team 2013).
General modelling approach
The presence of a given species detected at each point count location was used to analyse the distribution of alpine birds along the altitudinal gradient. Initial analyses suggested that species with occurrence rates < 15% had consistently poor model performance (see below) and often, problems with model fitting (e.g. lack of convergence), therefore a species was only considered if it occurred on at least 15% of the sample for the relevant open or closed habitat type. For each species, we considered only the likely nesting habitat, which we defined broadly into ‘closed’ and ‘open’ habitats following Chamberlain et al. (2013). The former was defined as any habitat where the cover of canopy + shrubs > 0. Open habitats were defined as any habitat where the number of mature trees was zero – these species to some extent tolerate some woody vegetation (e.g. young trees, shrubs), but tend to avoid mature trees and other vertical structures. Applying this ‘habitat mask’ had the advantage of focussing only on likely breeding habitat (and thus omitting isolated records of non-breeding and/or dispersing individuals) and omitting redundant zeros which may cause model fitting problems (Zuur et al. 2009). Black redstart Phoenicurus ochrurus was recorded in a range of habitats, so for this species, the entire data set was considered.
Bird distributions were analysed using logistic regression with the lme4 package in R (Bates 2015). Multiple visits were made to some points in Piedmont, which was accounted for by using a successes/failures syntax (Crawley 2013). Transect was fitted as a random factor in all models to account for non-independence of points along the same transect. In all cases, model fitting was preceded by a procedure to detect and reduce the effect of collinearity between the variables. This was done by calculating Variance Inflation Factors (VIFs) for the variables and sequentially deleting the variable with the highest VIF, as described by Zuur et al. (2009), using a cut-off value of 3.0. The 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211
final variable sets, with minimal levels of collinearity, were used for model averaging and variation partitioning (see below).
Modelling altitudinal trends across regions (Aim 1)
In order to determine variations in species distributions in relation to altitude and region, a statistical hypothesis testing framework was adopted, with the null hypothesis that bird species were
distributed randomly in relation to altitude and region. Both linear and quadratic terms were included in the models. ‘Region’ was included as a two-level factor (Piedmont or Trentino). The interactions between region and both altitude and altitude2 were included in the initial model for each species, where significant interactions indicate differing trends along the gradient between regions. These initial models were subject to a model reduction procedure whereby non-significant terms were sequentially dropped from a model until only significant (p ≤ 0.05) terms remained.
Assessing model performance (Aim 2)
Altitude may be a proxy for a multitude of effects operating at various scales (Hodkinson 2005). The extent to which either habitat, climate or altitude, either alone or in combination, could predict bird distributions was assessed by testing the performance of different models derived from a randomly selected data set from 70% of observations against the observed distributions from the remaining 30% of observations. Models were derived from the model data set (i.e. 70% of the observations) for altitude (ALT), habitat (HAB – habitat cover and topographic variables), climate (CLIM – temperature and rainfall) and combined habitat and climate (HAB+CLIM) variables for each species (see Table S1 in the “Electronic Supplementary Material” for a complete list of variables in each set). In each case, variables causing inflated VIFs (see above) were omitted. Non-linear effects were included in models following visual assessment of scatterplots (following Zuur et al. 2009). Altitude and temperature, 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237
and altitude and precipitation (which were highly collinear) were not modelled together. In total, there were ten climate variables considered, and there was also a high level of collinearity within this data set. A preliminary step was therefore undertaken to select the best fitting temperature and best precipitation variable for each species by comparing univariate models (i.e. only one climate variable at a time) using AIC. The climate variables used in CLIM and HAB+CLIM models were then those whose univariate models had the lowest AIC (see the “Electronic Supplementary Material” for further details).
For each model type (ALT, HAB, CLIM, HAB+CLIM), a model averaging approach, considering all combinations of models, was used to derive parameter estimates using the shrinkage method with the MuMIn package in R (Bartoń 2013). These were then used to predict the probability of presence in the test data set (i.e. 30% of observations). Observed presences were compared with the probability estimates from the model, and AUC was calculated from the package
PresenceAbsence (Freeman 2015) to test the ability of the models to correctly predict observed presence. Models with AUC < 0.70 are considered to have limited predictive capacity (e.g. Swets 1988). To aid interpretation, we further classified models as having adequate (0.70 ≤ AUC < 0.80) and good (AUC ≥ 0.80) predictive capacity.
Variation partitioning (Aim 3)
The categorical variable ‘region’ was added to HAB+CLIM models from the above procedure – these models containing habitat, climate and region were defined as ‘full’ models. Variation partitioning for generalized linear models (e.g. Ficetola et al. 2007) was then carried out on the full models in order to assess the amount of variation explained by HAB and CLIM variables, and by the categorical variable region. A large amount of variation attributable to region indicates that overall differences between regions are not attributable to the other model variables (i.e. HAB and CLIM). Marginal R2 values (fixed effects) for generalized linear mixed effects models (Nagakawa & Schieleth 2013) were 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263
calculated for the full model, and for HAB variables, CLIM variables, region, and combinations of these for each species using the r.squareGLMM command in R package MuMIn (Bartoń 2013). The variation attributable to each component was determined using the approach outlined by Legendre (2008). Altitude was not considered in this analysis due to strong collinearity with climate variables.
Results
The occurrence rates for the commonest species (present on 15% of points in both regions) are shown in Table 1, along with the classification into ‘open’ and ‘closed’ habitat species. There were 10 species that were recorded on at least 15% of points in both regions (for a total of 847 records relative to those species).
Altitudinal trends across regions (Aim 1)
All ten species considered showed significant variation in probability of occurrence in relation to altitude (Fig. 2). Two, robin Erithacus rubecula and willow tit Poecile montanus, showed a significant interaction between region and altitude. Robin showed significant linear decreases in probability of occurrence with altitude in both regions, although the decrease was steeper in Trentino compared to Piedmont (Fig. 2D). Willow tit showed a non-linear relationship with altitude in Piedmont,
occurrence peaking at c. 1900m, but a decline in Trentino (Fig. 2I). Overall probability of occurrence varied significantly between regions for several species and was higher for water pipit, wheatear
Oenanthe oenanthe and chaffinch Fringilla coelebs in Piedmont and higher for dunnock Prunella modularis and Robin in Trentino. In general, therefore, trends across the elevation gradient were
similar for most species in the two regions, although overall occurrence rates often varied. 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288
Based on Fig. 2, the species can be broadly defined into lower altitude species (those
showing a decline along the gradient), transition zone species (those showing a non-linear trend with a peak in probability of occurrence around the treeline) and open habitat species (either showing a peak in probability of occurrence in open grassland or an increase with altitude). The majority of species were closed habitat species showing a significant decrease with altitude: wren Troglodytes
troglodytes, robin, chiffchaff Phylloscopus collybita, coal tit Periparus ater and chaffinch (Fig. 2). In
addition to willow tit in Piedmont, dunnock also showed an intermediate peak at the transition zone (c. 1900m, Fig. 2C). Both open habitat species considered showed a non-linear trend, with a peak in probability of occurrence at intermediate altitudes: water pipit (c. 2205m, Fig. 2A) and wheatear (c. 2350m. Fig. 2F). Black redstart was the only species considered initially to be a habitat generalist and therefore analysed across all habitats. This species showed a peak in probability of occurrence at relatively high altitudes (c. 2650m, Fig. 2E) suggesting it was more of an open habitat species. However, when the species was analysed considering only open habitats, there was no significant effect of altitude, suggesting that the significant variation in Fig. 2E is largely driven by the contrast in species occurrence between open and closed habitats.
Model performance (Aim 2)
AUC values for ALT, HAB, CLIM, and combined habitat and HAB+CLIM models, are given in Table 2 (details of the highest ranked model for each species are given in Table S2 in the “Electronic Supplementary Material”). For black redstart, model fits were considered inadequate (AUC < 0.70) for all models. There were also three species where no models were classified as ‘good’, dunnock, wheatear and chiffchaff. Forest species tended to have better performing models than species of more open habitats. Considering the best performing model (i.e. the highest AUC value, regardless of classification) for each species, it was clear that models which included habitat were better than those without, and in particular combinations of habitat and climatic variables (HAB+CLIM models) 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314
tended to have relatively high AUC values (Fig. 3A). When considering models classified according to AUC (i.e. poor, adequate or good), again it was clear that HAB+CLIM models tended to perform best, followed by HAB models. ALT and CLIM models performed less well (Fig. 3B).
Variation partitioning (Aim 3)
Variation partitioning was used to assess the contribution of each of HAB variables, CLIM variables and region in the full models (parameter estimates are given in Table S3 in the “Electronic
Supplementary Material”). There was a wide range of variation explained by the fixed effects of the full model, from 0.20 in water pipit to 0.75 in robin (Table 3). Much of the variation was attributable to habitat variables, and to a lesser extent climate (with the notable exception of robin). The pure effect of region was very low in most models, suggesting regional differences (e.g. Fig. 2) could be explained largely on the basis of habitat and climate. The variation attributable to interactions between region and climate variables was, however, reasonably high, and likely arose due to the sometimes marked differences in climate between the two regions (Fig. S2 in the “Electronic Supplementary Material”).
Discussion
Altitudinal trends
Alpine bird species show marked patterns in distribution along altitudinal gradients. For the widely distributed species considered (i.e. those occurring relatively commonly in both Piedmont and Trentino), these patterns were generally consistent across regions (wren, chiffchaff, coal tit), although there were some species for which overall rates of occurrence varied, but the shape of the
315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340
relationship between probability of occurrence and altitude was the same between regions (water pipit, dunnock, wheatear, chaffinch). This has important implications for modelling species distributions, as it suggests model transferability for several species, i.e. a model derived from one region could be used to project relative elevational shifts in a wider area. There were two species that showed significant differences in altitudinal trend between region, robin and willow tit. Differences in habitat, and in particular the number of trees may explain these patterns, especially for willow tit, where the patterns in species occurrence match very closely with trends in number of trees across the two regions (compare Fig. 2I and Fig. S3A in the “Electronic Supplementary
Material”).
Observer effects may be important in such surveys (e.g. Sauer et al. 1994; Farmer et al. 2014), and we cannot rule out that these may have influenced overall between-region differences for some species. However, we believe such effects are likely to have been minimised as methods were identical in the two areas, the species involved were relatively easily identifiable by song, and the use of a fairly course measure, presence/absence, will have reduced subjectivity that might arise from making estimates of abundance. It is also notable that differences were not uni-directional – there were some species with higher occurrence rates in Trentino (dunnock, robin) and others with higher occurrence rates in Piedmont (water pipit, wheatear, willow tit, chaffinch). There were also
differences in survey effort between regions in that many points were subjected to two or more visits in Piedmont, but there was only a single visit in Trentino. However, there was no evidence that this affected the outcome of the results (Fig. S4 in the “Electronic Supplementary Material”).
Model performance across regions
Altitude correlates with gradients in habitat cover and with trends in climate, and is therefore thought to be a good general surrogate for the multiple environmental variables that are likely to dictate species distributions (e.g. Hodkinson 2005) and therefore to be a good basis for studying 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366
environmental, and in particular climate, change (Shoo et al. 2006). Although the species here showed clear variations along the altitudinal gradient, altitude models with a simple habitat mask (i.e. removing unsuitable nesting habitat prior to modelling) did not perform especially well (only coal tit had an altitude model considered ‘good’; Table 2). Schaub et al. (2011), working across a longer, but lower, altitudinal gradient also found only weak evidence that altitude was a good predictor of farmland bird density. Similarly, climate-only models performed relatively poorly, and there was no species that had a ‘good’ climate model (Table 2). For both climate and altitude models, there were several species whose models were considered adequate, so it should not be concluded that such models, masked for wholly inappropriate habitats, are of no value. However, it is clear that incorporation of habitat cover in the models resulted in improved model performance in many species. Whilst models using climate alone have proved useful in estimating species
distributions at broad scales (e.g. Huntley et al. 2007), in many situations (and particularly when considering finer scales) climate and bird distributions are unlikely to be very tightly linked when vegetation distribution is subject to other limiting factors (in particular, grazing by domestic livestock) and when complex topography may mean strong influences of microclimatic conditions. Data derived from relatively broad scales may therefore be inadequate to model distributions over steep altitudinal gradients where mean climates can change over short distances. Climatic data collected in the field at scales more appropriate to the activity of the birds (e.g. delimited by territory size, foraging range or nest site) may therefore provide the basis of more informative climate-only models. However, given the effort involved in collecting such data, it is difficult to envisage a situation where simple habitat variables modelled in conjunction with larger-scale climate variables would not prove to be the best option in terms of both effort and model performance.
There was very little variation attributable to region compared to climate and habitat variables. This is likely because most of the variation caused by region (e.g. Fig. 2) is in fact due to habitat and climatic differences already taken into account in the models, so inclusion of region in addition to climate does not add any useful information. This is further evidence (along with the 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392
consistency in altitudinal trends) that habitat and climate act on species distributions in a consistent way across geographic regions. It also implies that other unmeasured differences, such as geology, soil type, current and past land management, and disturbance (e.g. through winter sports or
hunting), either are unimportant in dictating bird distributions, or they do not vary sufficiently across regions. Of course, we would caution against assuming such relationships are consistent across other regions with widely differing environmental pressures. Our results suggest model transferability for the Southern Alps for several widespread species, but it would be worthwhile to repeat the study on distribution data from regions in different countries.
Wider implications
The relatively poor performance of climate-only models (models were ‘adequate’ for five species and inadequate for the rest) implies that climate alone does not have a major role in directly limiting species distributions along the altitudinal gradient. Habitat, or a combination of habitat and climate, showed better performing models, suggesting that habitat management can be used to some extent to improve conditions, potentially mitigating the negative effects of climate change for some species (Braunisch et al. 2014). Habitat degradation and loss are often considered to be the key threats to biodiversity, rather than climate change per se (e.g. Sala et al. 2000; Jetz et al. 2007; Chamberlain et al. 2016), and indeed in an alpine context, there are a number of environmental pressures which are likely to affect habitat quality (e.g. winter sports, Rolando et al. 2007; land abandonment, Laiolo et al. 2004), but whose effects could be ameliorated via habitat management. Nevertheless, climate change is also a major driver of habitat change in the Alps, in particular via effects on shifts in vegetation zones (e.g. Cannone et al. 2007) which may have consequences for bird distributions in the future (Chamberlain et al. 2013). Whilst climate apparently plays a relatively minor role in limiting current species distributions at the altitudes and at the fine spatial scale considered here, it is very likely to be of greater importance over broader contexts – a longer altitudinal gradient and/or 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418
a broader spatial scale may well have revealed a greater importance of climate in the models. Although not the goal of this paper, identifying the point along the altitudinal distribution at which climate becomes limiting would help to improve longer-term forecasts of potential effects of climate change.
Although overall occurrence rates often varied between regions, the species studied showed non-random altitudinal distributions which for most species were consistent across regions in terms of pattern, which is a key finding in terms of evaluation of the potential effects of climate change and associated habitat shifts (Araújo and Rahbek 2006). The shape and position of the altitudinal
distribution curve is important as it can be related to vulnerability where the available space is limited, i.e. where mountains are of not sufficient altitude for expansion (e.g. Chamberlain et al. 2014; Pernollet et al. 2015). This study therefore suggests general consistency in response in terms of the shape of the curve, and that regional differences are largely driven by habitat and climate. Incorporating these variables should be sufficient to construct models with high transferability for many Alpine species, a particularly relevant finding in terms of modelling species response to habitat characteristics and environmental change (Randin et al. 2006). However, despite adequate model performance in many cases, there was nonetheless often a large amount of unexplained variation, suggesting that there is considerable scope for improving model performance further. We suggest that further detailed autecological studies of alpine bird species are needed in order to improve our ability to describe their distributions, in particular in terms of understanding what specific factors mostly affect their occurrence. This would help understand the capacity of bird species to buffer the effects of climate change by means of (micro)habitat selection (Moritz and Agudo 2013). Such knowledge would contribute to evaluating species' sensitivity (Chamberlain et al. 2013) and adaptation potential (Bellard et al. 2012), and ultimately to build more precise models on which to base future scenarios of environmental change and conservation planning. This is particularly compelling for alpine habitats and species, as fine-scaled modelling is highly desirable in areas with 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443
strong altitudinal gradients, where coarse models may overestimate the potential habitat loss due to climate change (Randin et al. 2009).
Acknowledgements The research in Piedmont was partly funded through a grant to DEC from the
People Programme (Marie Curie Actions) of the European Union’s Seventh Framework Programme FP7/2007-2013/. The research in Trentino was partially supported by Accordo di programma PAT, 2010-13 and MUSE 2011-12; we are grateful to Franco Rizzolli and Francesco Ceresa for field work in Trentino. We are grateful to Ola Olsson and an anonymous referee for their constructive comments on an earlier version of the manuscript.
References
Araújo MB, Rahbek C (2006) How does climate change affect biodiversity? Science 313:1396-1397 Bartoń K (2013) MuMIn: Multi-model inference. R package version 1.9.0 ed
Bates D, Maechler M, Bolker B, Walker S (2015) Fitting linear mixed-effects models using lme4. J Stat Softw 67:1-48
Bellard C, Bertelsmeier C, Leadley P, Thuiller W, Courchamp F (2012) Impacts of climate change on the future of biodiversity. Ecol Lett 15:365-377
Bibby CJ, Burgess ND, Hill DA, Mustoe SH (2000) Bird census techniques, 2nd edn. Academic Press, London.
Bradbury RB, Pearce-Higgins JW, Wotton S, Conway GJ, Grice PV (2011) The influence of climate and topography in patterns of territory establishment in a range-expanding bird. Ibis 153:336-344 Brambilla M, Bergero V, Bassi E, Falco R (2015) Current and future effectiveness of Natura 2000
network in the central Alps for the conservation of mountain forest owl species in a warming climate. Eur J Wildl Res 61:35-44
444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468
Braunisch V, Coppes J, Arlettaz R, Suchant R, Zellweger F, Bollmann K (2014) Temperate mountain forest biodiversity under climate change: compensating negative effects by increasing structural complexity. PloS ONE 9:e97718
Cannone N, Sgorbati S, Guglielmin M (2007) Unexpected impacts of climate change on alpine vegetation. Front Ecol Environ 5:360-364
Caprio E, Chamberlain DE, Isaia M, Rolando A (2011) Landscape changes caused by high altitude ski-spites affect bird species richness and distribution in the Alps. Biol Conserv 144:2958-2967 Chamberlain D, Arlettaz R, Caprio E, Maggini R, Pedrini P, Rolando A, Zbinden N (2012) The
altitudinal frontier in avian climate change research. Ibis 154:205-209
Chamberlain DE, Negro M, Caprio E, Rolando A (2013) Assessing the sensitivity of alpine birds to potential future changes in habitat and climate to inform management strategies. Biol Conserv 167:127-135
Chamberlain DE, Negro M, Caprio E, Rolando A (2014) Shifting habitats in the Alps and potential future consequences for birds. Proceedings of the BOU’s 2014 Annual Conference, Ecology and Conservation of Birds in Upland and Alpine Habitats (published online: http://www.bou.org.uk/bouproc-net/uplands/chamberlain-et-al.pdf).
Chamberlain DE, Pedrini P, Brambilla M, Rolando A, Girardello M (2016) Identifying key conservation threats to Alpine birds through expert knowledge. Peer J 4:e1723
Chen I-C, Hill JK, Ohlemüller R, Roy DB, Thomas CD (2011) Rapid range shifts of species associated with high levels of climate warming. Science 333:1024-1026
Crawley MJ (2007) The R book, 2nd edn. Wiley, Chichester.
Dirnböck T, Essl F, Babitsch W (2011) Disproportional risk for habitat loss of high-altitude endemic species under climate change. Global Change Biol 17:990-996
EEA (2010) Europe’s ecological backbone: recognising the true value of our mountains. EEA Report 6/2010. European Environment Agency, Copenhagen
469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493
Farmer RG, Leonard ML, Flemming JEM, Andersen SC (2014) Observer aging and long-term avian survey data quality. Ecol Evol 4:2563-2576
Ficetola GF, Thuiller W, Miaud C (2007) Prediction and validation of the potential global distribution of a problematic alien species – the American bullfrog. Divers Distrib 13: 476-485.
Flousek J., Telensky T, Hanzelka J, Reif J (2015). Population trends of Central European montane birds provide evidence for adverse impacts of climate change on high-altitude species. PLoS ONE 10: e0139465.
Freeeman E (2007) PresenceAbsence: An R package for presence-absence model evaluation. USDA Forest Service, Ogden, USA
Gifford ME, Kozak KH (2012) Islands in the sky or squeezed at the top? Ecological causes of elevational range limits in montane salamanders. Ecography 35:193-203
Guisan A, Thuiller W (2005) Predicting species distribution: offering more than simple habitat models? Ecol Lett 8:993–1009
Harsch MA, Hulme PE, McGlone MS, Duncan RP (2009) Are treelines advancing? A global meta-analysis of treeline response to climate warming. Ecol Lett 12:1040-1049
Hijmans RJ, Cameron SE, Parra JL, Jones PG, Jarvis A (2005) Very high resolution interpolated climate surfaces for global land areas. Int J Climatol 25:1965-1978
Hodkinson ID (2005) Terrestrial insects along elevational gradients: species and community responses to altitude. Biol Rev 80:489-513
Huntley B, Green RE, Collingham YC, Willis SG (2007) A climatic atlas of European breeding birds. Lynx Edicions, Barcelona
Jetz W, Wilcove DS, Dobson AP (2007). Projected impacts of climate and land-use change on the global diversity of birds. PLoS Biology 6:e157
Laiolo P, Dondero F, Ciliento E, Rolando A (2004) Consequences of pastoral abandonment for the structure and diversity of the alpine avifauna. J Appl Ecol 41:294-304
494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518
Legendre P (2008) Studying beta diversity: ecological variation partitioning by multiple regression and canonical analysis. J Plant Ecol 1:3-8
Lehikoinen A, Green M, Husby M, Kålås JE, Lindström Å (2014) Common montane birds are declining in Northern Europe. J Avian Biol 45:3-14
Lenoir J, Gégout JC, Marquet PA, de Ruffray P, Brisse H (2008) A significant upward shift in plant species optimum elevation during the 20th century. Science 320:1768-1771
Maggini R, Lehmann A, Kéry M, Schmid H, Beniston M, Jenni L, Zbinden N (2011) Are Swiss birds tracking climate change? Detecting elevational shifts using response curve shapes. Ecol Model 222:21-32
Maggini R, Lehman A, Zbinden N, Zimmerman NE, Bollinger J, Schröder B, Foppen R, Schmid H, Beniston M, Jenni L (2014) Assessing species vulnerability to climate and land use change: the case of the Swiss breeding birds. Divers Distrib 20:708-719
Melendez L, Laiolo P (2014) The role of climate in constraining the elevational range of the Water Pipit Anthus spinoletta in an alpine environment. Ibis 156:276-287
Moritz C, Patton JL, Conroy CJ, Parra JL, White GC, Beissinger SR (2008) Impact of a century of climate change on small-mammal communities in Yosemite National Park, USA. Science 322:261-264 Moritz C, Agudo R (2013) The future of species under climate change: resilience or decline?
Science 341:504-508
Nagakawa S, Schiezeth H (2013) A geneal and simple method for obtaining R2 from generalized linear mixed-effects models. Methods Ecol Evol 4:133-142
Parmesan C, Yohe G (2003) A globally coherent fingerprint of climate change impacts across natural systems. Nature 421:37–42
Pernollet CA, Korner-Nievergelt F, Jenni L (2015) Regional changes in elevational distribution of the Alpine Rock Ptarmigan Lagopus muta helvetica in Switzerland. Ibis 157:823-836
R Development Core Team (2013) R: A Language and Environment for Statistical Computing 3.01. R Foundation for Statistical Computing, Vienna
519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544
Rahbek C (2005) The role of spatial scale and the perception of large-scale species-richness patterns. Ecol Lett 8:224-239
Randin CF, Dirnböck T, Dullinger S, Zimmermann NE, Zappa M, Guisan A (2006) Are niche based ‐ species distribution models transferable in space? J Biogeogr 33:1689-1703
Randin CF, Engler R, Normand S, Zappa M, Zimmermann NE, Pearman PB, Vittoz P, Thuiller W, Guisan A (2009) Climate change and plant distribution: local models predict high-elevation persistence. Global Change Biol 15:1557–1569
Reif J, Flousek J (2012) The role of species’ ecological traits in climatically driven altitudinal range shifts of central European birds. Oikos 121:1053-1060
Rolando A, Caprio E, Rinaldi E, Ellena I (2007) The impact of high-altitude ski-runs on alpine grassland bird communities. J Appl Ecol 44: 210-219
Sala OE, Chapin FS, Armesto JJ, Berlow E, Bloomfield J, Dirzo R, Huber-Sanwald E, Huenneke LF, Jackson RB, Kinzig A, Leemans R, Lodge DM, Mooney HA, Oesterhel M, Poff NL, Sykes MT, Walker BH, Walker M, Wall DH (2000). Global biodiversity scenarios for the year 2100. Science 287:1770-1774
Sauer JR, Peterjohn BG, Link WA (1994) Observer differences in the North American Breeding Bird Survey. Auk 111:50–62
Schaub M, Kéry M, Birrer S, Rudin M, Jenni L (2011) Habitat-density associations are not geographically transferable in Swiss farmland birds. Ecography 34:693-704
Sekercioglu CH, Schneider SH, Fay JP, Loarie SR (2008) Climate change, elevational range shifts and bird extinctions. Cons Biol 22:140-150
Shoo LP, Williams SE, Hero J-M (2006) Detecting climate change induced range shifts: Where and how should we be looking? Austral Ecol 31:22-29
Swets JA (1988) Measuring the accuracy of diagnostic systems. Science 240:1285-1293
Tryjanowski P, Sparks TH, Profuc P (2005) Uphill shifts in the distribution of the white stork Ciconia
ciconia in southern Poland: the importance of nest quality. Divers Distrib 11:219-223
545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570
Walther GR, Post E, Convey P, Menzel A, Parmesan C, Beebee TJ, Fromentin JM, Hoegh-Guldberg O, Bairlein F (2002) Ecological responses to recent climate change. Nature 416:389–395
Whittingham MJ, Krebs JR, Swetnan RD, Vickery JA, Wilson JD, Freckleton RP (2007) Should conservation strategies consider spatial generality? Farmland birds show regional not national patterns of habitat association. Ecol Lett 10:25-35
Wilson RJ, Gutiérrez D, Gutiérrez J, Martinez D, Agudo R, Monserrat VJ (2005) Changes to the elevational limits and extent of species ranges associated with climate change. Ecol Lett 8:1138-1146
Zuur AF, Ieno EN, Walker NJ, Saveliev AA, Smith GM (2009) Mixed effects models and extensions in ecology with R. Springer, New York
571 572 573 574 575 576 577 578 579 580
Table 1 Species occurrence (proportion of points where present) for species which occurred on at least 15%
of points in at least one region. Species were defined according to general nesting habitat (see text for details). Species occurrence is based on the total points surveyed per habitat. n is the number of points for each habitat and region.
Species Habitat Total n Piedmont n Trento n
Water pipit Anthus spinoletta OPEN 0.46 246 0.57 140 0.30 106
Wren Troglodytes troglodytes CLOSED 0.22 305 0.26 191 0.17 114
Dunnock Prunella modularis CLOSED 0.30 305 0.20 191 0.46 114
Robin Erithacus rubecula CLOSED 0.15 305 0.15 191 0.18 114
Black redstart Phoenicurus ochrurus ALL 0.21 453 0.24 271 0.16 182
Wheatear Oenanthe oenanthe OPEN 0.35 246 0.45 140 0.22 106
Chiffchaff P. collybita CLOSED 0.21 305 0.18 191 0.25 114
Coal tit Periparus ater CLOSED 0.28 305 0.27 191 0.28 114
Willow tit Poecile montanus CLOSED 0.30 305 0.37 191 0.18 114
Chaffinch Fringilla coelebs CLOSED 0.42 305 0.54 191 0.21 114
581 582 583 584 585 586 587 588 589
Table 2 Performance of different models describing probability of occurrence of species along altitudinal
gradients. Values presented are AUC ± SE . ALT includes altitude and altitude2; HAB includes habitat and topographic variables (after variable reduction according to VIF); CLIM includes climate variables (temperature and precipitation). HAB+CLIM includes both habitat and climate variables. AUC is coded according to thresholds of inadequate (white), adequate (shaded, normal text) and good (shaded, bold text) model performance.
Species ALT HAB CLIM HAB+CLIM
Water pipit 0.74 ± 0.06 0.74 ± 0.06 0.79 ± 0.05 0.84 ± 0.05 Wren 0.71 ± 0.06 0.67 ± 0.07 0.68 ± 0.07 0.57 ± 0.07 Dunnock 0.65 ± 0.06 0.56 ± 0.06 0.75 ± 0.05 0.76 ± 0.05 Robin 0.78 ± 0.06 0.76 ± 0.07 0.64 ± 0.07 0.77 ± 0.06 Black redstart 0.59 ± 0.05 0.67 ± 0.05 0.65 ± 0.06 0.64 ± 0.06 Wheatear 0.74 ± 0.06 0.51 ± 0.08 0.75 ± 0.06 0.65 ± 0.07 Chiffchaff 0.60 ± 0.07 0.76 ± 0.06 0.60 ± 0.06 0.74 ± 0.06 Coal tit 0.87 ± 0.04 0.83 ± 0.05 0.77 ± 0.05 0.89 ± 0.05 Willow tit 0.68 ± 0.06 0.87 ± 0.04 0.59 ± 0.06 0.84 ± 0.04 Chaffinch 0.69 ± 0.06 0.90 ± 0.04 0.77 ± 0.05 0.91 ± 0.03 590 591 592 593 594 595 596 597 598 599
Table 3. Variation partitioning based on R2 for generalised linear models (Nagakawa & Schielzeth 2013) derived from the fixed effects of full models, i.e. including habitat variables (HAB), climate variables (CLIM) and region (REG). The variation attributable to each individual group of variables is given by REG, CLIM and HAB. The variation attributable to interacting pairs of variables is given by REGHAB (region and habitat), CLIMHAB (climate and habitat) and REGCLIM (region and climate). The variation attributable to the
interaction between all three variables is given as 3-WAY. Negative values are interpreted as zeros (Legendre 2008).
Species Full REG CLIM HAB REGHAB CLIMHAB REGCLIM 3-WAY
Water pipit 0.200 0.004 0.028 0.114 -0.001 0.011 0.054 -0.010 Wren 0.299 0.001 0.022 0.148 0.002 0.119 0.128 -0.121 Dunnock 0.471 0.001 0.117 0.105 0.009 0.094 0.248 -0.103 Robin 0.755 0.061 0.309 0.107 0.028 0.328 0.278 -0.356 Black redstart 0.522 0.016 0.008 0.374 -0.018 0.137 0.124 -0.119 Wheatear 0.321 0.000 0.113 0.161 0.000 0.022 0.047 -0.022 Chiffchaff 0.429 0.015 0.069 0.180 0.016 0.138 0.165 -0.154 Coal tit 0.549 0.022 0.154 0.171 -0.009 0.217 0.202 -0.208 Willow tit 0.394 0.047 0.051 0.119 -0.012 0.164 0.177 -0.152 Chaffinch 0.648 0.004 0.030 0.200 -0.004 0.274 0.414 -0.270 600 601 602 603 604 605 606 607
Figure legends
Fig. 1 Location of the transects in the provinces of Turin (Piedmont) (A) and Trentino (B).
Fig. 2 Trends in the probability of occurrence in relation to altitude. Where there was an effect of region, or
where a species was analysed in only one region, dashed lines indicate Piedmont and dotted lines indicate Trentino. A solid line indicates a trend fitted from both regions combined (i.e. where there was no
significant difference). Observed presences and absences are shown as black squares (Piedmont) and grey squares (Trentino) on the y-axis maximum and minimum respectively. These are summarised into
frequencies for each category (region and presence/absence) calculated at each 100m interval. Symbol size is representative of the number of points, divided into five groups, where 1 (i.e. the smallest) = 1 point, 2 = 2-5 points, 3 = 6-10 points, 4 = 11-15 points and 5 = 16 points or greater.
Fig. 3 Performance of different models measured by AUC. A. Number of species for which a given model
had the maximum value of AUC. B. Number of species for which a given model was classified as having inadequate (AUC < 0.70, white bars), adequate (0.70 ≤ AUC < 0.80, grey bars) and good (AUC ≥ 0.80, black bars) predictive capacity. ALT = altitude only, HAB = habitat only, CLIM = climate variables only, HAB+CLIM= combined habitat and climate variables.
608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626
Fig. 1 627
P ro b ab ili ty o f p re se n ce 1700 2100 2500 2900 1700 2100 2500 2900 Fig. 2 628 629 630 631
Fig. 3 632
Online Supplementary Material
Part 1. Trends in habitat and climate along the elevational gradient
A summary of all environmental variables considered is given in Table S1. Variations in environmental variables (habitat cover and climate) were analysed in relation to altitude, altitude2, region and their interactions, using a similar mixed model structure and statistical hypothesis testing approach as used for birds, but specifying different error distributions. For grass and rock cover, percentage data were arcsine-square-root transformed prior to analysis with Gaussian mixed models using the lmer command in the R package lme4. For trees, canopy cover and shrub cover, only closed habitats were considered in order to reduce zero-inflation in the models. Shrub and canopy cover were arcsine-square-root transformed prior to analysis as above. The number of trees was analysed with a negative binomial model (preliminary analysis suggested overdispersion in the Poisson model) using the glmmADMB package (Skaug et al. 2006).
Fitted trends in habitat cover and number of trees in relation to altitude are shown in Fig. S1. There were significant interactions between altitude and region for number of trees, shrub cover and rock cover. The number of trees declined more sharply in Trentino than Piedmont (Fig. S1A). Canopy cover was higher in Piedmont than in Trentino for a given altitude, but there was no interaction (Fig. S1B). Shrub cover also showed a significant interaction - whilst there was a quadratic relationship with altitude in Piedmont, with a peak in shrub cover at c. 2200m (Fig. S1C), there was no significant relationship with altitude in Trentino. Grass cover was higher in Trentino than Piedmont, but showed no interaction, grass cover peaking at c. 2350m (Fig. S1D). Rock cover showed a significant interaction with region, where rock cover was higher in Trentino compared to Piedmont along most of the altitudinal gradient (Fig. S1E).
The mean temperature averaged over all points was similar between the two regions (Piedmont = 1.78˚C ± 1.77 SD, n = 271; Trento = 1.77˚C ± 1.48 SD, n = 182). There was a significant interaction between altitude and region on temperature (F1,417.2 = 17.78, p < 0.001), and temperatures were on average higher for a given altitude in Piedmont than Trentino, although the decline in temperature along the altitudinal gradient was similar in the two regions (Fig. S2A). There was a large difference in total precipitation between the two regions (Piedmont = 1473mm ± 172 SD, n = 271; Trento = 885mm ± 1.21 SD, n = 182). There was a significant interaction between precipitation and region (F1,407.4 = 64.23, p < 0.001), with a greater increase in precipitation towards higher altitudes in Trento, but precipitation was almost never greater in Trento than Piedmont at any altitude (Fig S2B).
634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665
Table S1 A full list of variables considered in the modelling procedure, and the model type to which they
are relevant. In subsequent figures and tables HAB+CLIM refers to models combining both HAB and CLIM variables. Mean temperature and rainfall for each month were extracted from WorldClim. Subsequent means, minima and maxima were based on monthly means (e.g. mean annual temperature is the mean of means for each month of the calendar year). Note that all variables were standardized to a mean of zero before modelling.
Model set Model code Variable Description
Altitude ALT Altitude Altitude above sea-level (km)
Climate CLIM Temp Mean annual temperature (˚C)
MaxT Maximum monthly temperature (˚C) MinT Minimum monthly temperature (˚C)
BreedT Mean breeding season (May-July) temperature (˚C) MinT Mean winter (December-February) temperature (˚C) Rain Total annual precipitation (mm)
MaxR Maximum monthly precipitation (mm) MinR Minimum monthly precipitation (mm)
BreedR Total breeding season (May-July) precipitation (mm) WinR Total winter (December-February) precipitation (mm) Habitat HAB Trees Number of mature trees (c. 20cm DBH) within 50m radius
Canopy %cover of canopy (vegetation >2m in height) within 100m radius
Shrubs %cover of shrubs (woody vegetation <2m in height) within 100m radius
Grass %cover of open grassland (i.e. no canopy) within 100m radius Rock %cover of unvegetated rock or scree within 100m radius Slope Slope at the point location estimated from DTM (angular ˚) E Extent of easterly exposition (radians, where 1 = directly east,
0 = directly west)
S Extent of southerly exposition (radians, where 1 = directly south, 0 = directly north)
33
A. Trees
Fig. S1 Variations in habitat extent in relation to altitude in Piedmont (dashed line, squares) and
Trentino (dotted line, triangles). All variables are measured as percentage cover in 100m radius, 666 667 668 669 670 671 672 673 674 675 676 677 678 679 65
B. Canopy
C. Shrubs
H
ab
it
at
e
xt
e
n
t
Altitude
Fig. S1 Continued.H
ab
it
at
e
xt
e
n
t
Altitude
E. Rock
D. Grass
680Altitude
Te
m
p
er
at
u
re
(
˚C
)
P
re
ci
p
it
ati
o
n
(
m
m
)
A.
B.
Fig. S2 Variations in (A) mean monthly temperature and (B) total annual precipitation in relation to
altitude in Piedmont (dashed line) and Trentino (dotted line). 681
Part 2. Details of the modelling procedure
Details of the best performing model for each species (i.e. that with the highest value of AUC in Table 2) are given in Table S3. Details of the full models for each species (i.e. including CLIM, HAB and REG) are given in Table S3.
Deriving CLIM models
For climate models, temperature and precipitation variables were extracted from WorldClim (Hijmans et al. 2005), as means and sums respectively, for each month of the year. These values were then used to calculate five different variables for temperature: annual mean temperature (i.e. mean of mean monthly temperature over the whole year), annual maximum temperature, annual minimum temperature, mean breeding season (May-July) temperature, and mean winter (Dec-Feb) temperature. Similarly, five different variables were calculated for rainfall, based on total amount: annual, annual maximum, annual minimum, breeding season and winter total rainfall. A range of variables were considered, as species populations are more closely linked to maximum, minimum or season-specific values (e.g. Pearce-Higgins et al. 2015). The goal was to identify climate models for each species which could explain the largest amount of variation in species distributions along the altitudinal gradient, but which had a low level of collinearity. There was a high level of collinearity with each set of temperature and precipitation variables, so models of species presence were first compared with AIC for each temperature variable, and for each precipitation variable, in turn. These models also included a quadratic terms if there was evidence of non-linear effects according to visual inspection of scatter plots. The temperature variable with the lowest AIC was then combined with the precipitation variable with the lowest AIC in a single model. In cases where VIF was above the threshold for temperature and precipitation, both were tested, and the univariate model with the lowest AIC was used as the CLIM model (see main text). Altitude and temperature, and altitude and precipitation (which were highly collinear) were not modelled together.
Fitting region*environment interactions
There was evidence that trends along the altitudinal gradient varied between regions for robin and willow tit (i.e. there was a significant altitude x region interaction). For the full models (i.e. that included region), interactions between region and habitat and region and climate were considered for these two species. Including all two-way interactions with region often resulted in model fitting problems (e.g.
non-convergence) and was also very time-consuming for the model-averaging procedure, in particular considering HAB models and HAB+CLIM models. We therefore adopted a pragmatic approach whereby 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716
interaction term which resulted in the model with the lowest model AIC was added to the full model, specifically region*shrubs for robin and region*temperature willow tit. In neither case did the addition of the interaction improve model performance in terms of AUC, and indeed it was noticeably worse for robin (robin without interaction = 0.67 ± 0.08, with interaction = 0.58 ± 0.09; willow tit without interaction = 0.84 ± 0.04, with interaction = 0.81 ± 0.04). Interactions were therefore not considered further in the full model for these or any other species.
Model performance using TSS
Reliance on a single measure of model performance is not generally recommended (Elith and Graham 2009). We therefore also considered an alternative measure, the True Skill Statistic (TSS; Allouche et al. 2006) to assess consistency between different measures of model fit. The optimal thresholds required for calculation of TSS were defined as the value of the observed prevalence in the model data set, one of the methods recommended by Liu et al. (2005). Concordance in ranking the different models across species according to the different measures of model performance (AUC v TSS) was assessed using Kendall’s coefficient for concordance between groups (Zar 1984), P-values being derived from a null hypothesis of no positive concordance (1-tailed test). Results were similar, in particular in that CLIM models performed less well over all (Table S4), and there was overall significant concordance in ranking of model performance between AUC and TSS (W = 0.10, z = 1.77, p = 0.038).
718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738
Table S2 Model-averaged parameter estimates ± SE for the best performing model (i.e. the top ranked
model according to AUC presented in Table 2, given in parenthesis after the species name), z-scores and associated p-values, and model variable weights for each species. Where there was a tie in AUC value, the model with the smaller standard error is presented. In one case, Wheatear, AUC estimates and standard errors were equal, so both models are presented. ALT includes altitude and altitude2; HAB includes only habitat and topographic variables; CLIM includes climate variables (temperature and precipitation). HAB+CLIM includes both habitat and climate variables. Models were derived from a model-averaging procedure based on a full model which had been subject to variable reduction according to VIF (Zuur et al. 2009) to minimise collinearity in the independent variable set. n/a indicates that a model weight was not calculated (for the intercept, and for those cases where there was only a single fixed effect). Further details are given in the text.
(a) Water pipit (HAB+CLIM)
Variable Estimate SE z-score p-value Model weight
Intercept 0.015 0.432 0.034 0.973 n/a Canopy -0.026 0.084 0.313 0.754 0.31 E -0.013 0.155 0.081 0.936 0.26 Rain 0.028 0.055 0.508 0.612 0.38 Rain2 -0.041 0.031 1.336 0.182 0.78 Rock -0.027 0.008 3.240 0.001 0.99 S 0.074 0.224 0.332 0.740 0.30 Shrubs -0.019 0.014 1.373 0.170 0.80 Slope 0.002 0.011 0.203 0.839 0.27 (b) Wren (ALT)
Variable Estimate SE z-score p-value Model weight
Intercept -2.176 0.313 -6.950 <0.001 n/a Altitude -5.484 1.183 -4.637 <0.001 n/a 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755
(c) Dunnock (HAB+CLIM)
Variable Estimate SE z-score p-value Model weight
Intercept -1.210 0.355 3.410 <0.001 n/a E -0.669 0.464 1.441 0.150 0.82 Grass -0.012 0.012 1.004 0.316 0.66 MinT 0.050 0.213 0.235 0.814 0.29 MinT2 -0.557 0.363 1.534 0.125 0.89 Rain -0.336 0.102 3.291 0.001 0.99 Rock -0.007 0.014 0.529 0.597 0.40 S 0.156 0.333 0.469 0.640 0.38 Shrubs 0.018 0.013 1.415 0.157 0.80 Slope 0.012 0.022 0.532 0.594 0.39 Slope2 -0.003 0.003 0.896 0.370 0.62 Trees -0.001 0.005 0.111 0.912 0.35 (d) Robin (ALT)
Variable Estimate SE z-score p-value Model weight
Intercept -2.586 0.290 -8.933 <0.001 n/a
Altitude -6.047 1.381 -4.379 <0.001 n/a
(e) Black redstart (HAB)
Variable Estimate SE z-score p-value Model weight
Intercept -2.587 0.362 7.154 <0.001 n/a E -0.081 0.197 0.413 0.680 0.34 Rock 0.012 0.009 1.333 0.182 0.77 S -0.101 0.232 0.435 0.663 0.35 Shrubs -0.001 0.005 0.255 0.800 0.30 Slope 0.017 0.022 0.757 0.449 0.52 Slope2 -0.0002 0.0009 0.263 0.792 0.30 Trees -0.051 0.016 3.141 0.002 1.00 756 757 758 759 760 761 762
(f) Wheatear (CLIM)
Variable Estimate SE z-score p-value Model weight
Intercept -0.168 0.300 0.559 0.576 n/a
MinT -0.231 0.239 0.968 0.333 0.64
MinT2 -0.479 0.162 2.957 0.003 1.00
(g) Chiffchaff (HAB)
Variable Estimate SE z-score p-value Model weight
Intercept -2.792 0.435 6.424 <0.001 n/a E -0.801 0.621 1.290 0.197 0.77 Grass -0.030 0.013 2.319 0.020 0.95 Rock -0.082 0.028 2.964 0.003 1.00 S 0.130 0.382 0.341 0.732 0.33 Shrubs 0.009 0.013 0.691 0.489 0.49 Slope 0.004 0.018 0.239 0.811 0.28 Slope2 0.0004 0.001 0.301 0.764 0.29 Trees -0.001 0.005 0.178 0.858 0.33
(h) Coal tit (HAB+CLIM)
Variable Estimate SE z-score p-value Model weight
Intercept -1.751 0.466 3.758 <0.001 n/a E -0.188 0.297 0.286 0.775 0.45 Grass -0.028 0.011 2.676 0.007 0.97 Rain -0.165 0.151 1.088 0.277 0.71 Rain2 -0.028 0.049 0.566 0.572 0.45 Rock -0.063 0.020 3.132 0.002 1.00 S -0.062 0.216 0.286 0.775 0.30 Shrubs -0.026 0.013 2.032 0.042 0.91 Slope 0.038 0.030 1.277 0.202 0.76 Temp 0.391 0.294 1.330 0.184 0.78 Temp2 -0.011 0.069 0.166 0.868 0.28 Trees 0.003 0.006 0.582 0.561 0.43
(i) Willow tit (HAB)
Variable Estimate SE z-score p-value Model weight
Intercept -1.659 0.245 6.757 <0.001 n/a E -0.073 0.207 0.354 0.723 0.32 Grass -0.021 0.010 2.050 0.040 0.94 Rock -0.034 0.018 1.828 0.068 0.91 S -0.074 0.217 0.341 0.733 0.31 Shrubs -0.006 0.010 0.614 0.539 0.46 Slope 0.008 0.018 0.462 0.644 0.36 Slope2 -0.0003 0.001 0.249 0.803 0.28 763 764 765 766 767 768 769 770 771
Trees 0.016 0.006 2.698 0.007 0.97
(j) Chaffinch (HAB+CLIM)
Variable Estimate SE z-score p-value Model weight
Intercept 0.011 0.716 0.015 0.988 n/a E 0.335 0.395 0.846 0.397 0.58 Grass -0.046 0.011 4.414 <0.001 1.00 Rock -0.092 0.019 4.711 <0.001 1.00 S 0.001 0.231 0.003 0.997 0.28 Shrubs -0.039 0.012 3.145 0.002 0.99 Slope 0.007 0.017 0.382 0.703 0.32 Temp 0.482 0.384 1.254 0.210 0.76 Temp2 -0.088 0.127 0.695 0.487 0.50 Trees 0.004 0.006 0.702 0.482 0.49 WinR 1.301 1.328 0.980 0.327 0.64 WinR2 -4.435 4.526 0.980 0.327 0.63 772 773 774
Table S3 Parameter estimates ± se for full models (i.e. HAB+CLIM + region) which were the basis of
variation partitioning. Region is a two-level categorical variable (Piedmont and Trentino), where the parameter estimate is for Piedmont (i.e. Trentino is the reference category). z-scores and associated p-values are also given for each species. Variables are defined in Table S1.
(a) Water pipit
Variable Estimate SE z-score p-value
Intercept -0.137 0.436 -0.314 0.753 Canopy -0.086 0.129 -0.669 0.504 E -0.168 0.318 -0.528 0.597 Rain -0.102 0.169 -0.603 0.547 Rain2 -0.038 0.026 -1.466 0.142 Rock -0.024 0.008 -2.923 0.003 S 0.302 0.372 0.814 0.416 Shrubs -0.025 0.124 -1.938 0.053 Slope 0.007 0.020 0.334 0.739 (b) Wren
Variable Estimate SE z-score p-value
Intercept -1.725 0.483 -3.576 <0.001 E -0.831 0.485 -1.713 0.087 Grass -0.014 0.010 -1.459 0.144 MaxR 0.186 4.865 0.038 0.969 MaxR2 -6.378 13.756 -0.464 0.642 Region 0.734 0.611 0.120 0.904 Rock -0.010 0.019 -0.542 0.588 S -0.154 0.517 -0.297 0.766 Shrubs 0.003 0.012 0.227 0.820 Slope 0.044 0.029 1.506 0.132 Slope2 -0.004 0.002 -1.839 0.066 Temp 0.284 0.433 0.654 0.513 Temp2 -0.117 0.146 -0.798 0.425 Trees 0.018 0.006 2.744 0.006 (c) Dunnock
Variable Estimate SE z-score p-value
Intercept -0.827 0.536 -1.544 0.123 E -0.937 0.394 -2.379 0.017 Grass -0.017 0.011 -1.485 0.138 MinT 0.752 0.838 0.898 0.369 MinT2 -0.584 0.331 -1.765 0.078 Rain 0.160 0.661 0.242 0.809 Region -1.267 1.735 -0.730 0.465 Rock -0.017 0.017 -0.993 0.321 S 0.494 0.405 1.220 0.223 775 776 777 778 779 780 781 782 783 784 785
