Structure and Origins of Trends in Hydrologic Measures over the Western US

This study investigates, at fine scale resolution, the spatial structure of observed trends in some key hydrologic measures over the western United States, and whether these trends are statistically significantly different from trends associated with natural climate variations. Measures of interest include wet day average temperatures, JFM runoff fractions of water year totals, numbers of wet day flood events, ratios of 1st April Snow Water Equivalent to cool season precipitation (SWE/Pcool), and ratios of summer soil moisture to annual precipitation (SM/Pann). The wet-day temperatures analyzed for trends were from a set of gridded daily observations usually used as inputs to the Variable Infiltration Capacity (VIC) hydrological model on a 12 km grid from 1950-1999. Those gridded observations were compared to simulated naturally occurring fluctuations from a 850 year control run of the CCSM3 FV climate model, as downscaled to the same grid. The observed and simulated meteorologies were used to drive the VIC hydrological simulations to obtain the remaining variables analyzed. Trends as large as the observed trends in several of the hydrometeorological variables--including wet day average temperatures and SM/Pann--occur less than 10% of the time in the long control run over large parts of the Western US. Although much of the trends in JFM runoff fractions can be attributed to large scale trends in cool season precipitation, there is also an elevation and temperature component to these changes that can be attributed to winter and spring warming over the western US. These runoff changes were consistent with reductions in SWE/Pcool that occurred preferentially in the same elevation and mean temperature zones, i.e., those that are most sensitive to winter and spring warming. The observed trends are different from long term fluctuations found in the natural variations in the control run and are more like those encountered in simulations forced with recent greenhouse gas increases. If these trends continue into future decades, they will have serious implications for the hydrological cycle and water supplies of the Western United States.