Abstract (from http://link.springer.com/article/10.1007%2Fs10980-015-0294-1):
Context
Tree species distribution and abundance are affected by forces operating at multiple scales. Niche and biophysical process models have been commonly used to predict climate change effects at regional scales, however, these models have limited capability to include site-scale population dynamics and landscape-scale disturbance and dispersal. We applied a landscape modeling approach that incorporated three levels of spatial hierarchy (pixel, landtype, and ecological subsection) to model regional-scale shifts in forest composition under climate change.
Objective
To determine (1) how importance value of individual species will change under the PCM B1 and GFDL A1FI modeling scenarios and (2) how overall forest composition at different spatial scales will change under these climate change scenarios in the short, medium, and long term in the Central Hardwood Forest Region (CHFR).
Methods
We used LANDIS PRO to predict forest composition changes from 2000 to 2300 accounting for climate change, population dynamics, dispersal, and harvest in the CHFR. We analyzed forest composition shifts under alternative climate scenarios and at multiple spatial scales.
Results
Shifts in forest composition were greater under the GFDL A1FI than the PCM B1 modeling scenarios and were greatest at the scale of ecological sections followed by forest sub-regions and the whole CHFR. Forest composition shifted toward more southern and xeric species and lesser northern and mesic species.
Conclusions
We suggest it is essential to include site- and landscape-scale processes in models and to evaluate changes at multiple spatial and temporal scales when evaluating changes in species composition due to climate change and disturbance.
