The question of how biological diversity influences the functioning of ecosystems has been of interest for decades and represents one of the grand challenge questions in ecology. In terrestrial ecosystems, most of the work on this topic has come from grasslands and other systems dominated by low stature vegetation that can be experimentally manipulated. Mature forests present a challenge because the size and lifespans of trees make it difficult to conduct manipulative diversity experiments. Although some studies have focused on previously established plantation forests, these opportunities are limited and often don’t coincide with measurements of whole-ecosystem function. The accumulation of data from eddy covariance networks provides a unique opportunity in that the growing temporal coverage over a large number of sites should eventually make it feasible to examine the influence of diversity using statistical, as opposed to experimental, approaches. Here, we present early results from an effort to examine forest canopy diversity in relation to ecosystem fluxes of carbon and water. We combine traditional metrics of species diversity with measures of spectral diversity from aircraft remote sensing and compare these with indices of ecosystem function related to carbon assimilation, water use efficiency and resistance to climate variability. Results are presented with respect to theories of diversity and ecosystem function and we discuss challenges that need to be overcome to further the relevance of ecological research networks to this important area of ecological theory.