The state of South Carolina has abundant water resources to support its rapidly growing population, booming industry, and flourishing agriculture. Nonetheless, the state periodically experiences short duration droughts that typically occur when water is most needed—especially by farmers who practice dryland farming—thus threatening both local and state economy. These droughts are often driven by large scale climate teleconnection patterns, such as the El Niño Southern Oscillation (ENSO), the Atlantic Multidecadal Oscillation (AMO), etc. Understanding the probability of occurrence and risks associated with these teleconnection patterns is crucial for efficient water demand management. In this study, we first examined the association between ENSO and AMO with streamflow in agricultural watersheds. We note that the magnitude and frequency of the effects of these teleconnections vary across the state and across events (weak or strong events). We also found that the strength of association varies across seasons, indicating that flow in some season may not be influenced by climate variability. We then developed a multivariate probabilistic model to assess the frequency and return period of teleconnection driven drought in these watersheds. Conditional probability models are developed, and univariate and multivariate return periods of drought, based on the state and phase of the teleconnection patterns, are computed.
