Live Fuel Moisture Estimation Using SMAP Soil Moisture and MODIS Vegetation Indices in Southern California, USA

Live fuel moisture (LFM) is a key fire risk indicator and is commonly used in fire behavior models. We compared eight widely used vegetation indices derived from MODIS reflectance and SMAP soil moisture on their capability to estimate LFM at 38 sites in Southern California, USA, a fire-prone Mediterranean ecosystem. Seasonality, inter-annual variability, or both were considered in our multivariate regression model construction. We found that VARI, a vegetation index calculated utilizing the reflectance from blue, green, and red bands performed the best among all remotely sensed moisture level indicators. SMAP soil moisture did not yield a good model performance mostly due to the great number of missing soil moisture retrievals in the data series. Frequency of LFM sampling also affected the model performance. The time lead of SMAP soil moisture to vegetation response can be a potential remotely-sensed predictor of LFM or an input to biophysical process model to retrieve LFM.