Soil moisture data collection and water supply forecasting: the sequel

The strong relationship between snow water equivalent and seasonal water supply has long since been demonstrated. Theses statistical relationships vary in strength depending on a host of factors such as latitude, elevation, and others. Part of the error associated with these relationships is due to unknown future meteorological conditions that impact the timing and volume of the runoff.  Another part of the error associated with statistical relationships is due to the variable state of soil moisture.  Soil moisture conditions across a watershed are generally presumed to influence seasonal water supplies from snowpack (Wetzel and Woodward, 1987).  If extremely dry conditions are prevalent in the fall prior to the seasonal snowpack, then it is presumed that these soils have additional capacity to absorb and retain some greater than normal amount of snowmelt, leaving a reduced amount to generate seasonal streamflow. Conversely, if the soils are saturated prior to the onset of snowmelt, it is presumed that, since the soils have less capacity for infiltration and certainly less storage, that the majority of snowmelt should contribute directly to seasonal streamflow.  The total potential snowmelt loss to soil moisture recharge can be significant, assuming a 24 inch soil depth, 8 to 10 inches of snowmelt or more, could be infiltrated depending on soil type and condition.  Some portion of this would eventually contribute to runoff and some portion would be lost from the immediate contributing system through either evapotranspiration or to deeper groundwater.  The Natural Resources Conservation Service now has soil moisture sensors at five sites in Utah (Julander and Cleary, 2001) with plans to install many more.  Preliminary data indicate great potential value from these data in reducing water supply forecast error. Interesting relationships between summer/fall precipitation and soil moisture are analyzed.  Snowpack, soil moisture and runoff are compared.