Long-term predictability of soil moisture dynamics at the global scale: Persistence versus large-scale drivers

Abstract

Here we investigate factors that influence the long lead time predictability of soil moisture variability using standard statistical methods. As predictors we first consider soil moisture persistence only, using two independent global soil moisture data sets. In a second step we include three teleconnection indices indicative of the main northern, tropical, and southern atmospheric modes, i.e., the North Atlantic Oscillation (NAO), the Southern Oscillation Index (SOI), and the Antarctic Oscillation (AAO). For many regions results show significant skill in predicting soil moisture variability with lead times up to 5 months. Soil moisture persistence plays a key role at monthly to subseasonal time scales. With increasing lead times large-scale atmospheric drivers become more important, and areas influenced by teleconnection indices show higher predictability. This long lead time predictability of soil moisture may help to improve early warning systems for important natural hazards, such as heat waves, droughts, wildfires, and floods.