Long‐term variability in precipitation and streamflow in Iceland and relations to atmospheric circulation

Abstract

Abstract How the variability of the atmospheric circulation affects precipitation in Iceland is not completely understood. Also, the sea surface temperature (SST) has a strong influence on the temperature over the country, and thereby, snow and glacial melt. This study, therefore, aims at explaining how atmospheric circulation and sea surface temperature influence seasonal and annual precipitation, and, consequently, runoff in Iceland. Empirical orthogonal function (EOF) analysis is performed on annual and seasonal time series of precipitation and discharge to identify their key modes of variability during the period 1966–2004. The correlation between the time series of each EOF mode with individual time series of sea level pressure (SLP), air temperature and SST was then evaluated. The analyses evidenced how large‐scale climate variables are connected to the regional precipitation and runoff in Iceland. They showed that the strength of the polar vortex may be, at least, as important for the precipitation in some areas of Iceland as the strength of the Icelandic Low (IL). Moreover, the location of the semi‐permanent IL often defines the predominant wind direction over the country and, as such, the regions of preferred precipitation. Since the watersheds act as large precipitation gauges with response patterns depending on the geology and glaciers, the variability of the annual discharge closely resembles the variability of precipitation, except for the glacial rivers. Glacial melt is highly correlated to air temperature and SST, and the spring discharge is affected by winter and spring temperatures. The results also revealed that Icelandic hydrological conditions in the spring can be forecasted by precipitation and temperature of the autumn and winter seasons, as well as by the general prevalent circulation patterns. Additionally, a potential for seasonal forecast of precipitation, and river discharge in other seasons was identified, particularly if seasonal forecast of SLP is available. Copyright © 2008 Royal Meteorological Society