B044-03
An Empirical Approach for Statistically Delineating Shared Global Fire Regimes

Wednesday, 9 December 2020: 20:38
Virtual
William Walter Hargrove, USDA Forest Service Southern Research Station, Eastern Forest Environmental Threat Assessment Center, Hot Springs National, AR, United States, Jitendra Kumar, Oak Ridge National Laboratory, Oak Ridge, TN, United States and Steven P Norman, USDA Forest Service, Asheville, NC, United States
Abstract:
Occurrence and behavior of fires across the globe are governed by climatic conditions, type and availability of vegetative fuels, and ignitions. Understanding the seasonality, frequency, intensity and severity of fires is critical for land managers to plan the landscape, and for researchers to understand the Earth system.

Unsupervised multivariate geographic clustering was used to produce world maps of Global Fire Regimes (GFRs) having similar fire intensity and timing characteristics. More than 83M MODIS “hotspot” thermal detections during 002-2019 were grouped into 10km cells, and 21 variables describing fire intensity, frequency, and seasonality characteristics were analyzed. A seasonal date transform allows the commonality of, e.g., “summer fires”, to be grouped together, even if they occur during different calendar months in the northern and southern hemispheres. Thus, the same GFR can be shared across geographic distances, if the 21 aggregated fire characteristics are sufficiently similar.

The majority of GFRs show wide spatial distribution across the world. An annual timeline for each GFR shows the seasonal timing of the top six bi-weeks of the year having the greatest frequency of fires. Most GFRs had two separate annual date modes of maximum fire frequency (1,374, 45.8%), often at the beginning and end of the growing or rainy season, followed by GFRs having three annual date modes (823, 27.4%) and those having a single annual maximum (466, 15.5%). GFRs having >3 high-frequency fire modes indicate swidden agriculture or deforestation.

GFRs can be selected from particular geographic “training” locations, and occurrences of the same GFRs elsewhere in the world will identify other global areas having similar fire characteristics. Developed for fire phenology, the Seasonally Synchronized date transform can be used for cross-hemisphere comparison of other types of phenological biotic or climatic events. Because locations in a single GFR or Location Group have similar fire/fuel systems, GFRs can facilitate the directed transfer of fire ecology knowledge and management. Managers can expect that experiences and understanding gained by other managers within the same GFR or Location Group might also apply to their own location, encouraging the exchange of successful fire management strategies worldwide.