Evaluating Uncertainty of Regional Water Budgets in the Central Valley, CA.


Dissertation

View 2017 UCOWR Presentation

View 2015 AGU Poster


Accurate estimation of groundwater budgets and effective management of agricultural groundwater pumping in California’s Central Valley is a priority for achieving new, legally-mandated groundwater management goals. Comprehensive measurements of agricultural groundwater pumpage in the Central Valley are uncommon, despite that this is typically the largest component of the groundwater budget. Without these measurements, accurate estimation of groundwater budgets remains a challenge in much of the Central Valley. CVHM and C2VSim are two regional-scale hydrologic models that couple groundwater and agricultural water budget models to provide historical and current estimates of distributed groundwater pumping, changes in groundwater storage, and evapotranspiration (ET) in the Central Valley. However, both models estimate these water budget components using conceptually different representations of soil-moisture conditions, estimations of ET requirements, and prioritizations of water allocation.

Conceptual models of the C2VSim soil zone 'control volume' methodology (left) and the the CVHM soil zone 'control interface' methodology (right).

Conceptual models of the C2VSim soil zone 'control volume' methodology (left) and the the CVHM soil zone 'control interface' methodology (right).

The uncertainties related to these conceptual differences have not been adequately investigated. Here, we evaluate differences in distributed agricultural groundwater pumping, groundwater change-in-storage, and ET estimates for both models at regional and sub-regional scales.

Central Valley CA 'subregions' (left) and regional aquifer systems (right) for which both CVHM and C2VSim were compared.

Central Valley CA 'subregions' (left) and regional aquifer systems (right) for which both CVHM and C2VSim were compared.

Results show wide-ranging, but typically large differences in the magnitude of simulated distributed agricultural groundwater pumping and in temporal groundwater storage trends, both at the regional and sub-regional scale. In general, model agreement is poor at the sub-regional scale that is important for water management.  Because many of the input data are the same for both models, these findings suggest that estimates of these important water budget components are sensitive to conceptual differences between methods, especially at the sub-regional scale.