Improving prediction and assessment of global fires using multilayer neural networks

Fires determine vegetation patterns, impact human societies, and are a part of complex feedbacks into the global climate system. Empirical and process-based models differ in their scale and mechanistic assumptions, giving divergent predictions of fire drivers and extent. Although humans have historically used and managed fires, the current role of anthropogenic drivers of fires remains less quantified. Whereas patterns in fire–climate interactions are consistent across the globe, fire–human–vegetation relationships vary strongly by region. Taking a data-driven approach, we use an artificial neural network to learn region-specific relationships between fire and its socio-environmental drivers across the globe. As a result, our models achieve higher predictability as compared to many state-of-the-art fire models, with global spatial correlation of 0.92, monthly temporal correlation of 0.76, interannual correlation of 0.69, and grid-cell level correlation of 0.60, between predicted and observed burned area. Given the current socio-anthropogenic conditions, Equatorial Asia, southern Africa, and Australia show a strong sensitivity of burned area to temperature whereas northern Africa shows a strong negative sensitivity. Overall, forests and shrublands show a stronger sensitivity of burned area to temperature compared to savannas, potentially weakening their status as carbon sinks under future climate-change scenarios.

Improving prediction and assessment of global wildfires using neural networks

ABSTRACTFires determine vegetation patterns, impact human societies, and provide complex feedbacks into the global climate system. Empirical and process-based models differ in their scale and mechanistic assumptions, giving divergent predictions of fire drivers and extent. Especially, the role of anthropogenic drivers remains less understood. Taking a data-driven approach, we use an artificial neural network to learn region-specific relationships between fire and its socio-environmental drivers across the globe. As a result, our models achieve higher predictability than previously reported, with global spatial correlation of 0.92, temporal correlation of 0.76, interannual correlation of 0.69, and grid-cell level correlation of 0.6, between predicted and observed burned area. Our analysis reveals universal global patterns in fire-climate interactions, coupled with strong regional differences in fire-human relationships. Given the current socio-anthropogenic conditions, Equatorial Asia, southern Africa, and Australia show a strong sensitivity of fire extent to temperature whereas northern Africa shows a strong negative sensitivity. Overall, forests and shrublands, show a stronger sensitivity of burned area to temperature compared to savannas, potentially weakening their status as carbon sinks under future climate-change scenarios.

Fire - climate interactions in a warming world

<p>Elevated fire activity in 2019 across the arctic, Amazon, Australia, and other regions sparked a discussion about the role of climate change for the recent rise in biomass burning.  Given that drivers of fire vary widely between different fire types and regions, interpreting trends requires a regional breakdown of the global pattern. Our Global Fire Emissions Database (GFED) now provides nearly 25 years of consistent data and offers important insights into changing fire activity. The GFED record captures a global decline in burned area, driven mostly by reductions in savanna fires from fragmentation and land use change. The global declining trend is therefore driven by areas with relatively low fuel loads where fire often decreases during drought.  Here, we report on increasing fire trends in several other regions, which become even more apparent when proxy data from before the satellite era are included. Increasing trends are concentrated in areas with higher fuel loads that burn more easily under drought conditions, and where warming leads to increasing vapor pressure deficits that contribute to more extreme fire weather and higher combustion completeness values. Therefore, the rate of decline in fire emissions is less pronounced than that in burned area, and emissions of several reduced gases have actually increased over time. The historic time series provides important context for trends and drivers of regions that burned extensively in 2019, and moving beyond burned area to estimate fire emissions of greenhouse gases and aerosols is critical to assess how these events may feed back on climate change if trends continue.     </p>