Abstract ID: 726

Towards understanding the tipping point of a transitional forest in Mato Grosso: results of a large-scale fire experiment.

Global climate change associated alterations of temperature and precipitation over the Amazon region are predicted to cause widespread forest dieback by the end of the twenty-first century, but with positive feedbacks between drought and fire, dieback may be much faster than predicted. To improve our understanding of these possible positive feedbacks, we conducted a long-term, large- scale fire experiment in a transitional forest in Mato Grosso. The experiment consisted of three 50-ha plots: a control plot (CP); a plot that burned in 2004 and in 2007 (B2); and a plot that burned annually from 2004 to 2007 (B4). In each plot we measured leaf area index (LAI), tree growth rates, and annual litterfall. We also conducted phenological observations on the 10 most common tree species. Overall, the forest was little affected by the first experimental fire. LAI was ~19% lower in B2 and B4 than in CP; tree growth rates in burned plots was 34% higher than in CP; litterfall did not differ among treatments; and, more trees fruited and flowered in B1 and B4 than in CP. Although two subsequent burns increased differences between CP and B4, substantial changes in forest structure/metabolism occurred only after the 2007 fires. During the 2007 fires, high ambient temperature, low humidity, and scant prior precipitation created perfect conditions for intense fires, particularly where fuel loads were elevated due to tree and branch mortality (B2). These results reinforce the notion that the interaction among forest fires, past forest disturbances, and droughts may have multiplicative effects on forest structure and dynamics in the Xingu region.

Session:  Fire - Fire, drought, and changes in vegetation structure and composition

Presentation Type:  Oral