Shifting fire regimes in Africa

In West Africa, fire regimes range from savannas that experience annual, low-intensity fires to wet tropical forests that rarely burn (Dwomoh and Wimberly 2017). Patterns of satellite active fire detections and burned area are mainly influenced by vegetation, whereas the seasonality and intensity of fires are more strongly associated with climate. Between 2003-2015, fire activity generally decreased in the savanna ecoregions and increased in the forested ecoregions. Forest fires were widespread during the regional drought of the 1980’s and resurged again in Ghana during an ENSO-associated drought event in 2016. (Dwomoh and Wimberly 2019). Forest degradation caused by fire and other disturbances increases the risk of subsequent fire (Dwomoh et al. 2017, Dwomoh and Wimberly 2019), raising concerns that increasing droughts resulting from climate change will catalyze new cycles of wildfire and forest loss. We are continuing to study the determinants and consequences fire in African ecosystems by using satellite remote sensing to characterize burned areas, vegetation moisture status, and forest structure.


Global change and fire in the western U.S.

Wildfires in western North America have increased in size and intensity over the past several decades because of climate change and landscape alterations that have affected ignitions and fuels. The EcoGRAPH group has studied the effects of climate, vegetation, and forest management on wildfire in the western United States (Liu and Wimberly 2015) and developed models to project the effects of climate change on future fire regimes (Liu and Wimberly 2016). These studies projected increased burned area and showed that the indirect effects of shifting vegetation types in response to climate change will be as important as the direct effects of climate change on fire weather and fuel moisture. We also projected the combined effects of climate change and wildland-urban interface (WUI) expansion on fire risk in the Colorado Front Range (Liu et al. 2014). The combined effect of these two factors was non-linear, with climate change exacerbating the risks from expansion of low-density homes into fire-prone vegetation types.

Fuel treatments, such as thinning and prescribed burning, are a potentially valuable tool for adapting to changing fire regimes. We have used satellite remote sensing data and fire behavior models to analyze the effects of different types of fuel treatments on wildfire size and severity (Wimberly et al. 2009, Cochrane et al. 2012). In general treatments that combined reduction of canopy fuels by thinning with reduction of surface fuels by prescribed burning or mastication were most effective at reducing fire severities and sizes. We also explored the influences of forest roads on fire ignitions, the locations of fire boundaries, and fire severity (Narayanaraj and Wimberly 2011, 2012, 2013). Both human and lightning ignitions were more frequent close to roads, but fires also had lower fire severities and were more likely to stop spreading close to roads.


Historical fire regimes and landscapes in the Pacific Northwest

Old-growth forests in the Pacific Northwest region of the United States are characterized by large tree sizes, accumulations of dead wood, and complex horizonal and vertical structure. The extent of old growth forests was greatly reduced by logging during the 20th century (Wimberly and Ohmann 2004). However, large wildfires occurred in the region prior to European settlement, raising the question how current levels of old growth compare to those under pre-settlement fire regimes. To answer this question, we developed the LAndscape Dynamics Simulator (LADS) to simulate wildfires and the dynamics of old growth forests in historical landscapes (Wimberly et al. 2000, Wimberly 2002). Despite the occurrence of large wildfires, we found that old growth was historically the most prevalent age class, covering 29-52% of the total forest area. The current area of old- growth forest is considerably lower than this historical range of variability (Wimberly et al. 2004). More recent versions of LADS have been modified to better model the dynamics of mixed-severity fire regimes (Wimberly and Kennedy 2008, Kennedy and Wimberly 2009). In the future, a warmer and drier climate will likely result in increased fire frequencies and severities that will push landscapes even further outside the range of historical variability (Wimberly and Liu 2014).