Experiment-Model Integration: A Case Study with EDGE Project

Site-level studies (e.g., a temperature manipulative experiment in a forest or measurements made in a grassland across years with different precipitation) can provide excellent information concerning ecological responses to climate change on small spatial scales. However, substantial limitations exist when trying to scale up to form regional and global predictions. Indeed, we know much more about how climate drivers are likely to be altered at various spatial scales (Murphy 2000, Schoof et al. 2010) than how the resulting impacts on ecosystem processes will play out to affect ecosystem services.

There are a number of major challenges associated with extending field- based findings to larger spatial scales. Two major ones are (1) understanding the mechanisms driving ecosystem responses to alterations in environmental variables and (2) obtaining knowledge about how these mechanisms vary across ecosystems. These challenges lead to an important overarching question: what are the relative strengths of climatic context (e.g., wet versus dry systems) versus ecosystem attributes (e.g., types of plant community) in driving how sensitive an ecosystem will be to changes in climate? For example, an arid system may have high sensitivity to drought because conditions are already dry and water is highly limiting. Yet on the other hand, xeric plant species may buffer primary production due to their having plant traits enabling growth even under stressful conditions. Indeed, empirical evidence has been shown for both increased sensitivity in drier systems (Huxman et al. 2004) and ecosystem attributes moderating responses to altered precipitation (Wilcox et al. 2015), yet methodological differences among studies obscure the relative strengths of these drivers in controlling ecosystem responses to climate change.