An Integrated Approach for Estimating Groundwater Storage Variations in Regional Aquifers, 20-R8051

Printer Friendly Version

Principal Investigators
Alexander Y. Sun
Ronald Green
Timothy Michaels

Inclusive Dates:  04/01/09 – Current

Background - Groundwater is the primary source of drinking and irrigation water supplies in many parts of the world. Groundwater aquifers are increasingly stressed by population growth and urban expansion, especially in semi-arid regions. Therefore, there is a pressing need to improve our capability to monitor and predict water availability and manage the limited water supplies in a way that ensures equitable, sustainable, and economically prudent decisions. Tracking groundwater storage variations in regional and local aquifers is challenging because groundwater levels are monitored at a limited number of locations at best. The Gravity Recovery and Climate Experiment (GRACE) satellite jointly launched by the United States and Germany provides a unique opportunity to monitor the temporal variations in terrestrial water storage (TWS), which, in turn, can be related to groundwater storage variations (GWS). Recent studies have demonstrated the potential usefulness of GRACE data for estimating GWS. However, few have considered the integration of the GRACE data for real-world water resources management. The combined use of GRACE data, in situ measurements, land surface modeling and groundwater models remains inaccessible to the majority of local water managers.

Approach - The reliability of groundwater management models may be adversely affected by uncertainties in model structure, parameter, and initial and boundary conditions. The conventional groundwater model optimization process mainly involves minimizing the discrepancy between observed and predicted groundwater pressure heads through adjusting model properties. The inverse solution is often not unique, especially when both specific yield and recharge rate are calibrated, in addition to hydraulic conductivity. The project team explored the integration of GRACE-derived GWS anomaly data as additional constraints in groundwater model optimization. The approach consists of (1) isolating GWS anomalies from GRACE data by removing other significant TWS components; (2) validating the results using in situ measurements and other sources; (3) formulating a multi-objective sequential parameter estimation strategy for integrating GRACE-derived GWS anomalies into groundwater model optimization; and (4) developing a web-based tool for seamlessly integrating and visualizing project results.

Accomplishments - Researchers chose the Edwards-Trinity Plateau aquifer (area approximately 100,000 km2) in Texas to demonstrate the technical approach. To date, researchers have successfully extracted and validated GRACE-derived GWS for the Edwards-Trinity Plateau aquifer using soil moisture data and in situ water level measurements. Results show that GRACE-derived data capture the trends in the Edwards-Trinity Plateau aquifer well. Researchers also developed an algorithm for multi-objective optimization using GRACE-derived GWS and successfully tested it via synthetic groundwater models. One conference abstract was submitted and several presentations were given to the local water resources management community.

2009 Program Home