Multi‐extent analysis of the relationship between pteridophyte species richness and climate

Abstract

Aim To determine the relationship between the distribution of climate, climatic heterogeneity and pteridophyte species richness gradients in Australia, using an approach that does not assume potential relationships are spatially invariant and allows for scale effects (extent of analysis) to be explicitly examined.Location Australia, extending from 10 ° S to 43 ° S and 112 ° E to 153 ° E.Method Species richness within 50 × 50 km grid cells was determined using point distribution data. Climatic surfaces representing the distribution and availability of water and energy at 1 km and 5 km cell resolutions were obtained. Climate at the 50 km resolution of analysis was represented by their mean and standard deviation in that area. Relationships were assessed using geographically weighted linear regression at a range of spatial bandwidths to investigate scale effects. ResultsThe parameters and the predictive strength of all models varied across space at all extents of analysis. Overall, climatic variables representing water availability were more highly correlated to pteridophyte richness gradients in Australia than those representing energy. Their variance in cells further increased the strength of the relationships in topographically heterogeneous regions. Relationships with water were strong across all extents of analysis, particularly in the tropical and subtropical parts of the continent. Water availability explained less of the variation in richness at higher latitudes. Main conclusionsThis study brings into question the ability of aspatial and single-extent models, searching for a unified explanation of macro-scaled patterns in gradients of diversity, to adequately represent reality. It showed that, across Australia, there is a positive relationship between pteridophyte species richness and water availability but the strength and nature of the relationship varies spatially with scale in a highly complex manner. The spatial variance, or actual complexity, in these relationships could not have been demonstrated had a traditional aspatial global regression approach been used. Regional scale variation in relationships may be at least as important as more general relationships for a true understanding of the distribution of broad-scale diversity.

Multi-scale mismatches between urban sprawl and landscape fragmentation create windows of opportunity for conservation development

“…The GWR models showed a decreasing scaling curve for extent, i.e., with sprawl and fragmentation patterns matching more closely at finer scales. This type of response has been found in previous studies that used various bandwidths of analysis to explore extent effects with GWR models (Bickford and Laffan 2006). As the extent decreases, the analysis becomes increasingly local and the model performance becomes increasingly inflated, which may obscure real differences in predictability between models (Jetz et al 2005).…”

Eight (and a half) deadly sins of spatial analysis

“…That is, in non-stationary data the single, semilocal regression coefficients (sensu Fotheringham et al, 2002) generated by spatial regression cannot be interpreted with confidence because a unique local coefficient does not exist in the data. Much like spatial autocorrelation itself, this fundamental property of geographical data was largely ignored until recently (Foody, 2004;Bickford & Laffan, 2006;Cassemiro et al, 2007;Beale et al, 2010;Landeiro & Magnusson, 2011), and it remains very uncommon for geographical ecology papers to report that they have determined whether their data are stationary or not. This is despite the fact that nonstationarity is common in broad-scale data; e.g.…”

Exploring Spatial Scale, Autocorrelation and Nonstationarity of Bird Species Richness Patterns

“…More recently, local spatial statistics have been used to investigate species richness patterns in relation to spatial autocorrelation [12][13][14] and spatial nonstationarity [15][16][17][18][19]. One of the most important parameters in local spatial statistics is the spatial weights matrix, which is used to determine "neighboring" observations.…”

“…If a relationship is spatially nonstationary, then the global statistical measures used to investigate it will more likely generate "smoothed" model results that may only be applicable to some of the study area or none at all. Nonstationarity of species richness-environment relationships has been studied for a variety of species using geographically weighted regression (GWR) [15][16][17][18][19]34].…”