Integrated Modeling of Drought-Impacted Areas using Remote Sensing and Microenvironmental Data in California

Abstract:

California’s ongoing drought condition shriveled not just the agricultural sector, but also the natural resources sector including forestry, wildlife, and fisheries. As future predictions of drought and fire severity become more real in California, there is an increased awareness to pursue innovative and cost-effective solutions that are based on silvicultural treatments and controlled burns to improve forest health and reduce the risk of high-severity wildfires. The main goal of this study is to develop a GIS map of the drought-impacted region of northern and central California using remote sensing data for the summer period of 2014. Specifically, Landsat/NAIP imagery will be analyzed using a combination of object-oriented classification and spectral indices such as the Modified Perpendicular Drought Index (MPDI). This spectral index basically scales the line perpendicular to the soil line defined in the Red-NIR feature space in conjunction with added information about vegetative fraction derived using NDVI. The resulting output will be correlated with USGS-produced estimates of climatic water deficit (CWD) data to characterize the severity of the drought. The CWD is simulated based on hydrological tool, Basin Characterization Model (BCM) that ingests historical climate data in conjunction with soils, topography, and geological data to predict other monthly hydrological outputs including runoff, recharge, and snowpack. In addition to field data, data collected by state agencies including USFS, calforests.org will be used in the classification and accuracy assessment procedures. Visual assessment using high-resolution imagery such as NAIP will be used to further refine the spatial maps. The drought severity maps produced will greatly facilitate site-specific planning efforts aimed at implementing resource management decisions.