Abstract. Droughts can result in enormous impacts for environment, societies, and economy. In arid or semiarid regions with bordering high mountains, snow is the major source of water supply due to its role as natural water storage. The goal of this study is to examine the influence of snow water equivalent (SWE) on droughts in the United States and find large-scale climatic predictors for SWE and drought. For this, a Maximum Covariance Analysis (MCA), also known as Singular Value Decomposition, is performed with snow data from the ERA–Interim reanalysis and the self-calibrating Palmer Drought Severity Index (sc–PDSI) as drought index. Furthermore, the relationship of resulting principal components and original data with atmospheric patterns is investigated.

The leading mode shows the spatial connection between SWE and drought via downstream water/moisture transport. Especially the Rocky Mountains in Colorado (CR) play a key role for the central and western South, but the Sierra Nevada and even the Appalachian Mountains are relevant, too. The temperature and precipitation based sc–PDSI is able to capture this link because increased soil moisture results in higher evapotranspiration with lower sensible heat and vice versa. A time shifted MCA indicates a prediction skill for drought conditions in spring and early summer for the downstream regions of CR on the basis of SWE in March. Furthermore, the phase of the El Niño–Southern Oscillation is a good predictor for drought in the southern US and SWE around Colorado. The influence of the North Atlantic Oscillation and Pacific North American Pattern is not that clear.