Subsidence due to Excessive Groundwater Withdrawal in the San Joaquin Valley, California

Francis Corbett¹, Thomas Harter¹ and Michelle Sneed^{2 1}Department of Land Air and Water Resources, University of California, Davis. ²U.S. Geological Survey Western Remote Sensing and Visualization Center, Sacramento. Abstract: Groundwater development within the Central Valley of California began approximately a century ago. Water was needed to supplement limited surface water supplies for the burgeoning population and agricultural industries, especially within the arid but fertile San Joaquin Valley. Groundwater levels have recovered only partially during wet years from drought-induced lows creating long-term groundwater storage overdraft. Surface water deliveries from Federal and State sources led to a partial alleviation of these pressure head declines from the late 1960s. However, in recent decades, surface water deliveries have declined owing to increasing environmental pressures, whilst water demands have remained steady. Today, a large portion of the San Joaquin Valley population, and especially agriculture, rely upon groundwater. Groundwater levels are again rapidly declining except in wet years. There is significant concern that subsidence due to groundwater withdrawal, first observed at a large scale in the middle 20th century, will resume as groundwater resources continue to be depleted. Previous subsidence has led to problems such as infrastructure damage and flooding. To provide a support tool for groundwater management on a naval air station in the southern San Joaquin Valley (Tulare Lake Basin), a one-dimensional MODFLOW subsidence model covering the period 1925 to 2010 was developed incorporating extensive reconstruction of historical subsidence and water level data from various sources. The stratigraphy used for model input was interpreted from geophysical logs and well completion reports. Gaining good quality data proved problematic, and often values needed to be estimated. In part, this was due to the historical lack of awareness/understanding of subsidence drivers. The model is calibrated to both measured and extrapolated subsidence data. Sensitivity analyses are implemented and several future scenarios evaluated: reduced pumping, 'business as usual' pumping, and increased pumping demand. We show that water level decline, beginning in the 1950s and ending in the early 1970s, is followed closely by subsidence. Also, recent groundwater pumping is shown to drive renewed subsidence. An evaluation of agricultural water use, the main driver of groundwater level decline, shows that deficit irrigation, switching to crops with significantly lower consumptive water use, and active recharge programs are key to addressing long-term groundwater overdraft in light of limited surface water resources.