Sources of variability of evapotranspiration in California

The variability (1990-2002) of potential evapotranspiration estimates (ETo) and related meteorological variables from a set of stations from the California Irrigation Management System (CIMIS) is studied. Data from the National Climatic Data Center (NCDC) and from the Department of Energy from 1950 to 2001 were used to validate the results. The objective is to determine the characteristics of climatological ETo and to identify factors controlling its variability (including associated atmospheric circulations). Daily ETo anomalies are strongly correlated with net radiation (R-n) anomalies, relative humidity (RH), and cloud cover, and less with average daily temperature (T-avg). The highest intraseasonal variability of ETo daily anomalies occurs during the spring, mainly caused by anomalies below the high ETo seasonal values during cloudy days. A characteristic circulation pattern is associated with anomalies of ETo and its driving meteorological inputs, R-n, RH, and T-avg, at daily to seasonal time scales. This circulation pattern is dominated by 700-hPa geopotential height (Z(700)) anomalies over a region off the west coast of North America, approximately between 32degrees and 44degrees latitude, referred to as the California Pressure Anomaly (CPA). High cloudiness and lower than normal ETo are associated with the low-height (pressure) phase of the CPA pattern. Higher than normal ETo anomalies are associated with clear skies maintained through anomalously high Z(700) anomalies offshore of the North American coast. Spring CPA, cloudiness, maximum temperature (T-max), pan evaporation (E-pan), and ETo conditions have not trended significantly or consistently during the second half of the twentieth century in California. Because it is not known how cloud cover and humidity will respond to climate change, the response of ETo in California to increased greenhouse-gas concentrations is essentially unknown; however to retain the levels of ETo in the current climate, a decline of R-n by about 6% would be required to compensate for a warming of +3degreesC.