VEGETATION RESPONSE TO CLIMATE CHANGE DEPENDS ON THE ABUNDANCE AND TYPE OF FIRE BREAKS

The response of vegetation to future climate change is poorly understood at the spatial scales most relevant to human activities. Paleoecological methods can reconstruct vegetation responses to past climatic changes, but data are rarely collected at sufficient spatial and temporal scales to reveal how different parts of a landscape responded to climate changes. This study compares changes in vegetation at sites on an outwash plain with pine and oak forests.

Previous research suggests that sites surrounded by abundant firebreaks responded differently to climate changes of the last 2500 years than sites with few firebreaks. Sites protected from fire fluctuate between pine and oak forests, but sites without protection remain dominated by jack/red pine. Lake-levels were ~1.5 m lower during the mid-Holocene (8000-4500 years ago), and wetlands would have been much reduced, leaving lakes as the only effective fire protection. We hypothesize that sites protected by wetlands in the recent past would develop a fire-dependent jack pine forest similar to that around modern sites with little fire protection. We compared the response of vegetation (pollen) at 3 lakes with different abundances and types of firebreaks: unprotected (Metzger, 5.5% lakes, 1.5% wetlands), protected by wetland (Ferry, 10% lakes, 15% wetlands) and protected by lakes (Lonestar, 15% lakes, 10% wetlands).

Results support the hypothesis, with pollen at Ferry dominated by jack/red pine during the mid-Holocene. Pollen assemblages at Ferry averaged 53% jack/red pine and 13% oak during the mid-Holocene. Metzger had similar values throughout the last 8000 year (avg 52% & 7% respectively). Lonestar remained more protected and fluctuated between oak and pine forest during this time (avg 27% & 31%). Our results demonstrate that paleoecological data with high spatial and temporal resolution provide a valuable way to predict landscape responses to past and future climate change.