Delineating Fire-Hazardous Areas and Fire-Induced Patterns Based on Visible Infrared Imaging Radiometer Suite (VIIRS) Active Fires in Northeast China

Abstract

(1) Background: Fire affects global agricultural and/or forest ecosystems with high biomass accumulation. However, the delineation of fire-hazardous areas based on satellite-derived active fire intensity is not well-studied. Therefore, examining the characteristics of fire occurrence and development plays an important role in zoning fire-hazardous areas and promoting fire management. (2) Methods: A fire intensity (FI) index was developed with Visible Infrared Imaging Radiometer Suite (VIIRS) active fires and then applied to identify fire-hazardous areas in Northeast China. Combined with terrain, land cover and net primary productivity (NPP), the spatial and temporal characteristics of active fire occurrence were consistently analyzed. Next, a conceptual decision tree model was constructed for delineating fire-induced patterns impacted by varied factors in Northeast China. (3) Results: The accidental, frequent, prone and high-incidence areas of active fires defined by the FI index accounted for 31.62%, 30.97%, 26.23% and 11.18%, respectively. More than 90% of active fires occurred in areas with altitude 2. Similarly, about 75% occurred in cropland and forest. Then, four fire-induced conceptual patterns driven by different factors were classified, including the agricultural and forest active fire-induced patterns (i.e., the Agri-pattern and FRST pattern) with NPP ranging 2500–5000 kg·C/m2, and two others related to settlements and unused land with an altitude <350 m asl. The Agri-pattern dominates in Northeast China because of agricultural straw burning. (4) Conclusions: Despite the national bans of open burning of straws, active fires due to agricultural production have occurred frequently in Northeast China in the last decade, followed by small and sporadic forest fires. The approach for defining fire-hazardous areas and varied fire occurrence patterns is of significance for fire management and risk prediction at continental to global scales.

Keywords

• VIIRS active fire
• fire-hazardous area
• fire-induced patterns
• spatial recognition
• Northeast China