Stochastic and deterministic drivers of spatial and temporal turnover in breeding bird communities

Abstract

Aim A long‐standing challenge in ecology is to identify the suite of factors that lead to turnover in species composition in both space and time. These factors might be stochastic (e.g. sampling and priority effects) or deterministic (e.g. competition and environmental filtering). While numerous studies have examined the relationship between turnover and individual drivers of interest (e.g. primary productivity, habitat heterogeneity, or regional – ‘gamma’ – diversity), few studies have disentangled the simultaneous influences of multiple stochastic and deterministic processes on both temporal and spatial turnover. If turnover is governed primarily by stochastic sampling processes, removing the sampling effects of gamma diversity should result in non‐significant relationships between turnover and environmental variables. Conversely, if deterministic processes govern turnover patterns, removing sampling effects will have little influence on turnover gradients. Here, we test these predictions. Location The United States. Methods Continental‐scale, multidecadal data were used to quantify spatial and temporal turnover in avian community composition within 295 survey routes. A series of regression and structural equation models were coupled with a null model to construct statistical models describing turnover patterns. Results Examining explanatory variables alone or in combination showed that spatial and temporal turnover increased together, decreased with primary productivity and increased with habitat heterogeneity. The relationships between turnover and all variables became weaker when sampling effects were removed, but relationships with primary productivity and habitat heterogeneity remained relatively strong. In addition, spatial turnover increased strongly with spatial gamma diversity after sampling effects were removed. Main conclusions Our results show that spatial and temporal turnover are related to each other through a stochastic sampling process, but that each type of turnover is further influenced by deterministic processes. The relative influence of deterministic processes appears, however, to decrease with primary productivity and increase with habitat heterogeneity across an east–west longitudinal gradient in North America.

Biogeographic, climatic and spatial drivers differentially affectα-,β- andγ-diversities on oceanic archipelagos

“…However, because b A /a can vary independently from g, their positive relationship therefore suggests biological mechanisms beyond the additive relationship. For example, such a relationship could be caused by dispersal limitation, niche specialization or competitive exclusion [24,27]. Moreover, adaptive radiations, taxon-cycle dynamics and island-hopping [3,5,6] are also important for species turnover [17], and our results suggest that they indirectly affect archipelagic g-diversity.…”

“…The key role of b as a driver of g, evident from the direct positive effect of the relative importance of b A compared with a (b A /a) on g (figure 3c), has also been recently reported for mainland birds [27] and trees ( [41], but see [48]). The effect of b can be at least partially attributed to the additive partitioning [30,41].…”

“…isolation and area) and climate (e.g. temperature) still affect archipelagic g-diversity directly, as has been recently suggested for breeding birds at a continental scale [27].…”

“…[23,24]) and spatial landscape structure [25,26]. Considering that g-diversity results from both its a-and b-components, potential abiotic drivers may actually affect g indirectly via a and b rather than directly [27]. Spatial intra-archipelagic structure may be important for g-diversity because of its effects on the b-component, namely species turnover among individual islands.…”

Governance of Land Use Planning to Reduce Fire Risk to Homes Mediterranean France and California

“…They have similar climates, and both areas face frequent fire events [28][29][30]. Both have large amounts of housing in the Wildland Urban Interface (WUI) [4,31], where many local forests are not managed for optimal production and therefore may be highly dense and fire prone [32]. While the social systems are in many ways similar, the property rights regimes that underlie land use planning are quite different.…”