The potential transient dynamics of forests in New England under historical and projected future climate change

Projections of vegetation distribution that incorporate the transient responses of vegetation to climate change are likely to be more efficacious than those that assume an equilibrium between climate and vegetation. We examine the non-equilibrium dynamics of a temperate forest region under historic and projected future climate change using the dynamic ecosystem model LPJ-GUESS. We parameterized LPJ-GUESS for the New England region of the United Sates utilizing eight forest cover types that comprise the regionally dominant species. We developed a set of climate data at a monthly-step and a 30-arc second spatial resolution to run the model. These datasets consist of past climate observations for the period 1901-2006 and three general circulation model projections for the period 2007-2099. Our baseline (1971-2000) simulation reproduces the distribution of forest types in our study region as compared to the National Land Cover Data 2001 (Kappa statistic = 0.54). Under historic and nine future climate change scenarios, maple-beech-basswood, oaks and aspen-birch were modeled to move upslope at an estimated rate of 0.2, 0.3 and 0.5 m yr(-1) from 1901 to 2006, and continued this trend at an accelerated rate of around 0.5, 0.9 and 1.7 m yr(-1) from 2007 to 2099. Spruce-fir and white pine-cedar were modeled to contract to mountain ranges and cooler regions of our study region under projected future climate change scenarios. By the end of the 21(st) century, 60% of New England is projected to be dominated by oaks relative to 21% at the beginning of the 21(st) century, while northern New England is modeled to be dominated by aspen-birch. In mid and central New England, maple-beech-basswood, yellow birch-elm and hickories co-occur and form novel species associations. In addition to warming-induced northward and upslope shifts, climate change causes more complex changes in our simulations, such as reversed conversions between forest types that currently share similar bioclimatic ranges. These results underline the importance of considering community interactions and transient dynamics in modeling studies of climate change impacts on forest ecosystems