Seasonality drives global-scale diversity patterns in waterfowl (Anseriformes) via temporal niche exploitation

被引:53
作者
Dalby, Lars [1 ,4 ]
McGill, Brian J. [2 ,3 ]
Fox, Anthony David [1 ]
Svenning, Jens-Christian [4 ]
机构
[1] Aarhus Univ, Dept Biosci, DK-8410 Ronde, Denmark
[2] Univ Maine, Sch Biol & Ecol, Orono, ME 04469 USA
[3] Univ Maine, Sustainabil Solut Initiat, Orono, ME 04469 USA
[4] Aarhus Univ, Dept Biosci, DK-8000 Aarhus C, Denmark
来源
GLOBAL ECOLOGY AND BIOGEOGRAPHY | 2014年 / 23卷 / 05期
基金
美国国家科学基金会;
关键词
winter range; productivity; NDVI; climate; seasonality; migratory birds; latitudinal diversity gradient; bird species richness; Anatidae; SPECIES RICHNESS; PREDATION RISK; CLIMATE; ENERGY; BIRDS; DISTRIBUTIONS; MIGRATION; DETERMINANTS; PRODUCTIVITY; BIODIVERSITY;
D O I
10.1111/geb.12154
中图分类号
Q14 [生态学(生物生态学)];
学科分类号
071012 ; 0713 ;
摘要
Aim Birds (class Aves) overall follow the expected latitudinal gradient in species richness. However, there is a poorly understood secondary peak in bird richness at about 45 degrees N latitude, at least in North America. Waterfowl (Anseriformes) species richness conforms to this secondary peak, yet this group is commonly excluded from bird species richness studies and consequently the drivers of waterfowl richness are poorly understood. Here, we explore the drivers of waterfowl richness on a global scale, emphasizing the secondary peak. Location Global. Methods We mapped total waterfowl species richness for the breeding and non-breeding seasons. Considering a wide range of potential environmental drivers (climate, productivity and habitat), we initially ran univariate ordinary least squares models for the candidate set of predictors. We then used regression trees (RT) to multivariately assess their importance for the richness pattern while allowing for nonlinear and non-stationary processes. Results We found seasonal variability in plant productivity (measured by the normalized difference vegetation index, NDVI) to be the most important predictor of breeding season richness in both univariate regressions and multivariate RT models. For the non-breeding season, winter actual evapotranspiration (AET) was the best predictor. In the regions of highest richness, including the secondary mid-latitude peak, an overwhelming portion of the species were migratory species. Main conclusions Predictors commonly used to explain the large-scale richness patterns of birds, such as annual AET or annual NDVI explained little of the variation in richness in Anseriformes. Instead, measures reflecting intra-annual variability and seasonal productivity in the relevant season were the best predictors. This finding, combined with the high proportion of migrants in the richness peaks, strongly suggests that the patterns of richness in Anseriformes reflect the group's exploitation of seasonal environmental variability via short- and long-distance migration, i.e. temporal niche exploitation at an annual and global scale.
引用
收藏
页码:550 / 562
页数:13
相关论文
共 71 条
[1]  
AEWA, 2008, AGR CONS AFR EUR MIG
[2]  
[Anonymous], 1990, INT WATERFOWL RES BU
[3]  
[Anonymous], 2013, R LANG ENV STAT COMP
[4]  
[Anonymous], 2011, BIRD SPEC DISTR MAPS
[5]  
[Anonymous], STAT BIOL HLTH ANAL
[6]  
[Anonymous], 2007, TREES WATER SMALLHOL
[7]  
[Anonymous], 2010, Global soil water balancegeospatial database
[8]  
[Anonymous], 2012, GEOSPHERE SPHERICAL
[9]   Species and area [J].
Arrhenius, O .
JOURNAL OF ECOLOGY, 1921, 9 :95-99
[10]   Can niche-based distribution models outperform spatial interpolation? [J].
Bahn, Volker ;
McGill, Brian J. .
GLOBAL ECOLOGY AND BIOGEOGRAPHY, 2007, 16 (06) :733-742