Integrating experimental and gradient methods in ecological climate change research

Field-based research on the responses of ecosystems to anthropogenic climate change has primarily used either natural gradient or experimental methods. Taken separately, each approach faces methodological, spatial, and temporal limitations that potentially constrain the generality of results and predictions. Integration of the two approaches within a single study can overcome some of those limitations and provide ways to distinguish among consistent, dynamic, and context-dependent ecosystem responses to global warming. A simple conceptual model and two case studies that focus on climate change impacts on flowering phenology and carbon cycling in a subalpine meadow ecosystem illustrate the utility of this type of integration.