Prediction of Large-Scale Wildfires With the Canopy Stress Index Derived from Soil Moisture Active Passive

Abstract

Land surface variables such as surface soil moisture are recognized as important wildfire indicators. However, quantifying wildfire fuel combustibility through only satellite-retrieved soil moisture remains challenging, because soil moisture does not provide information about vegetation (fuel) moisture and water availability to plants is complicated by another factor of soil properties. Thus, to enhance the wildfire prediction ability of the soil moisture active passive (SMAP) mission, this study examines a canopy stress index (CSI) retrieved from 1 km SMAP level 2 products. The strong relationship between prefire CSI and fire severity is demonstrated over two large-scale (greater than 1000 ha) wildfires in Gang-won Province, South Korea. CSI can effectively predict the severity of large-scale wildfires one week before fire events, differentiating dry soils from wildfire hazards. SMAP L2 data with a temporal resolution of 5–7 d over the study sites are suitable for supporting aerial firefighting activities and reducing false fire warnings.

Keywords

• Canopy stress index (CSI)
• canopy temperature
• large-scale wildfire prediction
• soil moisture active passive (SMAP) brightness temperature