Paper presented at Data Driven Approaches to Drought (DDAD) 2011.


Assessment and understanding of past climate is an important step for drought mitigation and water resources planning. In this study, streamflow simulation obtained from the variable infiltration capacity (VIC) model was used for drought characterization, and subsequently regionalization was done based on the annual severity level, for the Brazos basin in Texas over a time span of 1949-2000. It is important to study drought characteristics within a regional context. Hence, identification of homogenous drought regions is a prerequisite, so that the drought characteristics can be studied within each of these regions. In this study, the concept of entropy was used for formation of homogenous regions based on drought severity. A standardized version of mutual information known as directional information transfer was used for station grouping. Results obtained were compared with the conventional k-means clustering method. The regions obtained were similar in both cases.


Drought Regionalization, Entropy, Directional information transfer, K-means clustering

Date of this Version