Remote Sensing-based Drought Analysis using Hydrometeorological Variables

Abstract

Drought is one of the main natural causes of agricultural, economic, and environmental damage. Drought is easy to classify following a long period without precipitation events, however it is difficult to determine its onset, extent, and end. Recently, there have been many efforts to develop new drought indices and to improve existing ones for drought analysis. Most studies related to drought analysis and monitoring systems have been conducted using ground measured dataset. Ground measured dataset have limited ability to provide representative values, because the densities of the measurement networks are always sparse. To resolve such problems, numerous studies have been carried out on the development of remote sensing-based drought indices which can provide accurate spatial distributions of drought status over broad regions. This study proposed a new drought index (energy-based water deficit index (EWDI)) using Moderate Resolution Imaging Spectroradiometer (MODIS) imageries to consider hydrometeorological variables such as evapotranspiration, soil moisture, solar radiation, and vegetation activities. The EWDI is a first attempt to analyze drought events for considering water and carbon circulation. This study also compares satellite-based drought indices, the EWDI and the stand-alone MODIS-based evaporative stress index (stMOD_ESI) to assess capability of the EWDI over drought vulnerable regions such as Mongolia, Australia, and Korean peninsula from 2000 to 2010. In addition, this study found correlations among various drought indices and applicability using Receiver Operating Characteristics (ROC) analysis, which will expand our understanding of the relationships between hydrometeorological variables and drought events at global scale. Of the drought indices examined, the EWDI and stMOD_ESI were found accurate when capturing moderate drought conditions, compared to the SPI-3. These results suggest that the satellite-based drought indices (EWDI and stMOD_ESI) may be applicable on a regional scale. This study provides an insight of the evaluation of remotely sensed drought index derived from satellite imageries as well as the operational hydrometeorological investigation and water management activities.