Abstract
The relationship between net primary productivity (NPP) and species richness has been the subject of long-running debate. A changing climate gives added impetus to resolving this debate, as it becomes increasingly necessary to predict biodiversity responses that might arise from shifts in productivity or its climatic correlates. It has become increasingly clear that at small scales productivity-species richness relationships (PSRs) are variable, while at macro scales relationships are typically positive. We demonstrate the importance of explicitly considering scale in discussions on PSRs even at large scales by showing that distinct patterns emerge in a global dataset of terrestrial ecoregions when ecoregions are binned into size classes. At all sizes, PSRs in ecoregions are positive, but the strength of the PSR scales positively with ecoregion size. In small ecoregions (10(3)-10(4) km(2)), factors correlating with productivity play only a minor role in species richness patterns, while in large ecoregions (>10(5) km(2)), NPP modelled from remotely sensed data is able to explain most of the variation in species richness. Better understanding the effects of scale on PSRs contributes to the debate on the relationship between species richness and productivity, which in turn allows us to better predict how both long- and short-term biodiversity patterns and ecosystem functioning might be altered under global change scenarios. This gives focus on future research to clarify causal pathways between species richness and productivity with appropriate attention to scale as an important focusing element.