by Maithili Joshi
The relationship between fire-induced tree mortality and extreme weather remain poorly understood because it is restricted to post-fire observations of tree mortality. Studies done on the effects of forest fires and biodiversity remain understood on the patch scale, and do not consider the effects of fire on vegetation dynamics and structure. In the southeast Amazon forest, scientists established a large scale, and long term prescribed forest fire experiment in a transitional forest. Primarily, trying to determine if there are weather, and fuel, related thresholds in fire behavior associated with high levels of fire-induced tree mortality across two different fire regimes, and secondarily, what the effects of an intense forest fire are on forest structure, flammability, and aboveground live carbon stock.
Two experiments were conducted. First, three 50 hectare lots were established in a region with no signs of recent fires. The three lots were an unburned control, a lot burned every three years (2004 – 7), and a plot burned annually (2004 – 9). Second, regional analysis of weather and fire scars was studied in which burn scars in forested areas were mapped based on Landsat images.
In the 2007 drought, precipitation across the Xingu region was lower than in any other year recorded. This resulted in high-intensity fires. The difference in fire treatment is associated with fire spread rate given that “fuel composition”. More leaf litter, and other things on the ground contributed to an increase in forest fires. These fires did not self-extinguish at night as they often do in non-drought years. This is attributed to high vapor pressure deficit, low litter moisture content, and high fuel loads.
In 2007 there was a sudden spike in tree mortality compared to other years. Post-fire mortailty was pronounced along the forest edge treatment burned every three years, which mimics the rate of mortality in the region. This suggests that transitional forests are more resistant to low-intensity fires than wetter Amazonian forests because mortality rates in wetter forests tend to be considerably higher.
These results led to two drivers of fire intensity and fire induced tree mortality. First, 2007 was dryer and warmer, with higher fine fuel loads. The dry and warm weather conditions increased fuel loads, a likely cause of increased fire intensity. Second, fuel loads and mortality rates were higher along the forest edge, suggesting a difference in fire-induced tree mortality due to fine fuel loads.
Above ground biomass reduced significantly because of elevated fire-induced tree mortality. This led to reduced leaf area index causing more light to permeate, leading to more dryness in the understory. Along the edges of burned plots, grasses invaded, increasing fire intensity because grasses accumulated more fine fuel close to the ground than the trees they replaced. The grass invasion suggests that high intensity fires could promote abrupt fire-mediated transitions from forest to new stable states. These transitions are more likely to occur in fragmented forest areas, where disturbances are frequent and dry seasons are prolonged.
Another factor affecting tree mortality is deforestation, or other human disturbances. These affect mortality by reducing canopy cover and evapotranspiration. An increase in the average dry season and land-surface temperatures causes forests to become dryer and more prone to forest fires. Deforestation also creates a larger perimeter of forest fragments. Finally, tree mortality associated with logging, fire, drought, or edge effects contributes to coarse fuel loads for multiple years with twigs, branches, and standing dead trees decaying and falling to the ground. These forest fires also contribute to the climate of the Amazon, most forest burned in 2007 experienced another drought later, signifying that previous forest fires leave these areas more vulnerable to future ones.
- Brando, P. M., Balch, J. K, Nepstad, C. N.,
- Morton, D.C, et al. 2014 Abrupt Increases in Amazonian Tree Mortality Due To
- Drought–Fire Interactions PNAS 111 no. 17, 6347–6352