Fast versus slow succession in eastern North America: Why do old fields persist in the north?

Background/Question/Methods Studies of ecological succession are a foundation of community ecology.  However, there have been few attempts to integrate single-site studies into a broader-scale understanding of the processes underpinning succession patterns.  Anecdotal evidence suggests that the rate at which old-field succession occurs, specifically the rate at which woody species become dominant over initial herbaceous colonizers, is faster in more southern regions of the Eastern Deciduous Forest (EDF) than in more northern regions.   One explanation for this putative biogeographic pattern is that warmer temperatures generally increase biological processes. If this latitudinal gradient in rates of succession exists and it is due to differences in temperature, climate shifts over the next century could alter the rate at which succession occurs. This could have important implications for critical ecosystem processes (such as carbon sequestration) and community attributes (such as invasive species colonization).  However, confounding factors, such as geographic variation in soil fertility and species pools, suggest alternative hypotheses concerning the major driver of the rates of succession. 
Results/Conclusions We present results from a survey of 32 studies of the rates of old-field succession from throughout the geographic range of the EDF where we were able to estimate the rate at which woody species become dominant in old fields.  We find strong evidence for a latitudinal gradient in the rate of woody succession, agreeing with our hypothesis that woody species become dominant more quickly in southern regions of the EDF than in northern regions.  While the observational nature of these data precludes any direct tests of the underlying mechanisms, we present a general model outlining likely hypotheses to explain the variance in the balance of competition between woody and herbaceous species across a broad geographic scale and discuss the likelihood of the alternate hypotheses.  Additionally, we describe a recently started experimental network designed to directly test the most likely mechanisms driving the latitudinal gradient in the rate of woody succession.