Climate change and inertia: A long-term perspective from pine and oak-dominated communities in northwestern Wisconsin

Background/Question/Methods

Ecological theory, models, and limited empirical data predict that forest communities will vary in their responsiveness to climate change, depending on factors such as site quality, disturbance regimes, and competitive interactions.  Paleoecological records from a dense network of sites in northwestern Wisconsin documenting changes in plant communities and fire regimes during the past 3000 years allowed us to test predictions about factors that stabilize vegetation during periods of centennial-scale climate change.  In particular, we tested whether on poor quality sites jack pine (P. banksiana) communities would be more stable because of strong feedbacks between fire and jack-pine and lack of competition from other tree species.  At sites on finer-grained sand we expected vegetation would change more because changes in climate (and possibly fire regimes) would alter the outcome of competitive interactions among species and provide the opportunity for a new community type to become established.  We used fossil pollen data from 5 lake sediment cores to measure the amount of change in plant communities both as the rate of change in nonmetric multidimensional scaling (NMS) axis scores and as changes in the vegetation types assigned to fossil pollen assemblages using a classification scheme based on reference samples. 

Results/Conclusions

Our results show that plant communities on poor-quality sites (coarse sands) retained jack pine vegetation throughout the past 3000 yrs, while plant communities on finer-grained soils showed 2-3 transitions in vegetation types during the same period.  However, the average rate of change in NMS scores/100yr was not related to site quality. In fact, the highest average rate of change occurred in a record from a poor quality site that had jack pine vegetation in 86% of the samples, with only brief interludes of other vegetation types.  This suggests that frequent stand-destroying fires created high rates of vegetation change, but that jack pine vegetation quickly re-established.  In comparison, records from three sites on finer-grained sand had only moderate average rates of change per century, but over the past 3000 yrs each site supported 3-4 different vegetation types, each of which persisted for decades to centuries.  The results of this study demonstrate that context is important to predicting how vegetation will respond to climate change, and that communities in which disturbance overwhelms the effects of competition can have greater inertia during periods of climate change.