Analyzing the sensitivity of drought recovery forecasts to land surface initial conditions

Droughts are complex hydro-meteorological phenomena, which are challenging to both quantify and predict. From the perspective of drought quantification, knowledge of the land surface conditions is vital for determining the impacts a drought event is having on both the environment and society. Although such land surface information is essential for quantifying drought in real-time, the precise effect of land surface moisture deficits on future drought conditions is unknown. Forecasting of recovery from drought events is undoubtedly reliant on its intensity, yet the lead time at which a drought can be expected to recover is poorly understood. Due to this gap in knowledge, this study attempts to quantify the expected lead time for drought recovery, and the rate of drought recovery, by examining the loss of sensitivity to initial conditions within a climatological forecast. From this perspective, the expected recovery time from a specific drought event is quantified, based on a case study in the Upper Colorado River Basin in Southwestern USA for two initialization dates in years 2003 through 2008. This study has ramifications for understanding the time of drought recovery, and highlights the importance of accurate land surface state estimation. With respect to recent studies, the experiments presented here suggest that forecasts can be sensitive to initial conditions at greater lead-times, and therefore drought conditions are potentially more persistent than previously thought. (C) 2014 Elsevier B.V. All rights reserved.