Beyond description: the active and effective way to infer processes from spatial patterns.

The ecological processes that create spatial patterns have been examined by direct measurement and through measurement of patterns resulting from experimental manipulations. But in many situations, creating experiments and direct measurement of spatial processes can be difficult or impossible. Here, we identify and define a rapidly emerging alternative approach, which we formalize as "space as a surrogate" for unmeasured processes, that is used to maximize inference about ecological processes through the analysis of spatial patterns or spatial residuals alone. This approach requires three elements to be successful: a priori hypotheses, ecological theory and/or knowledge, and precise spatial analysis. We offer new insights into a long-standing debate about process-pattern links in ecology and highlight six recent studies that have successfully examined spatial patterns to understand a diverse array of processes: competition in forest-stand dynamics, dispersal of freshwater fish, movement of American marten, invasion mechanisms of exotic trees, dynamics of natural disturbances, and tropical-plant diversity. Key benefits of using space as a surrogate can be found where experimental manipulation or direct measurements are difficult or expensive to obtain or not possible. We note that, even where experiments can be performed, this procedure may aid in measuring the in situ importance of the processes uncovered through experiments.