If you have any problems related to the accessibility of any content (or if you want to request that a specific publication be accessible), please contact us at email@example.com.
Assessing Groundwater Pumping and Landsat Satellite Crop Evapotranspiration Estimates in Diamond Valley, Nevada for Improved Water Resources Management
AuthorOtt, Thomas Joseph
AltmetricsView Usage Statistics
Groundwater pumping for irrigation has resulted in overdraft for many aquifers across the western United States. As a result, regulators and producers are developing and implementing groundwater management plans to monitor and reduce groundwater pumping in many areas. Groundwater management plans often require quarterly and annual reporting of pumped groundwater which can be difficult for water managers to quality assure and quality control (QAQC). In this study, it is hypothesized that satellite-based evapotranspiration (ET) estimates will be useful to QAQC and ultimately estimate groundwater pumping at field and basin scales. To test this hypothesis, the objectives of this study were as follows: first, to pair an ensemble of satellite-based ET model estimates of Net ET (ET less precipitation) with metered pumping data from Diamond Valley, Nevada, second, to develop a regression model to QAQC metered pumping data, and third, to ultimately predict pumping and assesses uncertainty using nonparametric bootstrapping. Predictions of pumping volume were summarized at field and basin scales and compared with pumping volumes reported by the Nevada Division of Water Resources (NDWR). While previous studies have used satellite-based ET to estimate groundwater pumping volumes, those studies have had little to no metered pumping data to validate their estimates. In this study, Landsat satellite-based ET estimates of daily and monthly ET and fraction of reference ET (ETf) were derived from METRIC, SSEBop, and SIMS ET models. Next, the ensemble of ET results was spatially averaged over each field and paired with water right places of use (POU) which resulted in over 100 pumping-ET pairs. While the regression of Net ET and metered pumping had a relatively low coefficient of determination (R2 of 0.45), it proved to be useful for identifying outliers and errors in the metered pumping database. Predicted pumping volume for all metered wells was on average within 5.6% of metered pumping volume at the basin scale but had high variability at the field scale. Total basin pumping volume estimated based on constant irrigation water requirement were on average 35% more than reported pumping volume by NDWR or predicted pumping volume using the regression model.Results from this study illustrate the utility in assessing and estimating pumping volumes using satellite-based ET models. While there are many uncertainties with both ET estimates and metered pumping data, it is argued that satellite-based ET estimates of pumping are more accurate than estimating pumping based on a constant irrigation water requirement. Furthermore, satellite-based estimates of pumping can be used to QAQC reported pumping data and has the potential to be used to estimate historical pumping volumes.