Large-scale climate, precipitation and British river flows: Identifying hydroclimatological connections and dynamics

• Published in: Journal of Hydrology
• Main Authors: Lavers, David, Prudhomme, Christel, Hannah, David M.
• Format: Article in Journal/Newspaper
• Language: unknown
• Published: 2010
• Subjects: Meteorology and Climatology; Hydrology; Atmospheric Sciences; North Atlantic; North Atlantic oscillation
• Online Access: http://nora.nerc.ac.uk/id/eprint/13450/; https://doi.org/10.1016/j.jhydrol.2010.10.036

Investigation of links between large-scale climate and river basin-scale precipitation and discharge is necessary to improve understanding of the hydroclimatological process chain. This paper aims to identify the large-scale circulation variables that yield significant statistical relationships with precipitation and river flow for ten British basins, and evaluate the dependence of the strength and location of centres of climate–hydrology correlation on basin situation. Using quasi-geostrophic theory as a framework to select appropriate explanatory variables, this study undertakes ranked correlation analysis between gridded ERA-40 atmospheric data and precipitation and river flow observations at a monthly resolution over a 26 year period. Relationships between large-scale atmospheric circulation and precipitation and river flow are found to vary spatially and seasonally across Britain, with strongest associations evident in western Britain and in winter. Monthly precipitation totals and river flow demonstrate highly significant negative (positive) correlation with mean sea level pressure MSLP (Zonal Wind U), but the location of high correlation centres vary for different basins. There are weaker atmospheric links with river flow compared to precipitation, reflecting the non-linearity of the rainfall–runoff transformation and the importance of basin properties as a modifier of climate inputs. The North Atlantic Oscillation Index (NAOI) cannot capture the seasonal movement of climate–hydrology connections due to the fixed station locations used for the index calculation, and thus has weaker monthly correlations with precipitation and river flow compared to gridded ERA-40 circulation variables (and in particular the comparable MSLP).