Drought Resilience JAWRA Article

Home Browse Search

MILLER, DALE, BRUSH, VICUNA, KADIR, DOGRUL, AND CHUNG coincident with the Little Ice Age cooling period. Samples from sediments, tree rings, and tree stumps, combined with isotope dating analysis have been used to reconstruct these naturally occurring droughts that lasted 50 to more than 100 years (Stine, 1994; Herweijer et al., 2006; Cook et al., 2007). Indeed, two epic drought periods, one lasting from approximately 900 -1100, and the second lasting from about 1200- 1350, contributed to the decline and disappearance of the Anasazi people, a culture that relied on irrigated agriculture to support its population. Drought is also seen as a contributing factor in the failure of Euro- pean colonies in South Carolina and North Carolina in the 1500s. More recently, four droughts in the wes- tern U.S. centered on AD 1710, 1770, 1850, and 1930 have been associated with the Pacific Decadal Oscilla- tion (PDO), and indicate drought recurrence intervals of 60 -80 years (Benson et al., 2003) and a linkage to large -scale climatic phenomena. During the last 150 years, California has been in a slightly above average wet regime, with an annual average precipitation of 58 cm (23 inches), and at least 11 significant drought periods (Ingram et al., 1996; Cook et al., 2004). At the same time, California Central Valley agriculture has expanded over most of the Valley floor, and includes a system of managed irrigation and water conveyance that has assumed climatically stationary conditions for conveyance sys- tem development and planning. The 1929 -1934 drought has traditionally been the benchmark event used for designing storage capacity and yield of large California reservoirs. The stationarity principle may no longer be valid, as substantial anthropogenic changes in Earth's climate are altering the means and extremes of precipitation, evapotranspiration, and rates of discharge to rivers (Milly et al., 2008). Changes in the temperature regime in California associated with projected future climate are expected to result in reduced winter snowpack and increased winter runoff (Miller et al., 1999; Hayhoe et al., 2004; Maurer and Duffy, 2005). In addition, the population of California's Central Valley has increased from less than 3 million people in 1970 to more than 6 million in 2002, and is projected to increase to 15 million peo- ple by 2050 (U.S. Bureau of the Census 1982; Califor- nia Department of Finance 2007). Since the 1970s, as the urban area of the Central Valley has increased, agricultural acreage has remained relatively constant by expanding into previously uncultivated land. The increase in population coupled with constant agricul- tural acreage has resulted in steadily increasing water demands. Approximately 35% of the water demand in the Central Valley is currently met with ground water (California Department of Water Resources 2003), with pumping rates increasing in years of reduced surface water availability. Flow deficits associated with future climate scenarios, cou- pled with present and future levels of water demand, may inflict significant stress on Central Valley aqui- fers. In light of these challenges, the California Department of Water Resources (CDWR) and other water agencies have begun to evaluate new approaches for incorporating the changing climate into water resources planning and management (CDWR, 2006; Anderson et al., 2008). The goals of this study are to quantify the impacts of long -term hydrologic droughts — a first -order approximation of an analogue for climate change related snowpack reduction — on water storage, and to illustrate the potential for surface and subsurface storage to limit the adverse impacts of drought and snowpack reduction on water supply and hydropower generation. This includes how ground -water pumping compensates for reductions in surface water inflow; the extent to which the water table is reduced; and how, when, and if this system recovers or reaches a new equilibrium. In the next section, we provide details on our approach for simulating persistent droughts in the California Central Valley. This is followed by the results and discussion section, then our summary and conclusions. APPROACH This analysis of the impacts of sustained droughts in the California Central Valley is based on a series of specified reductions in net surface water inflows observed during the 1923 -1972 period. The reductions considered in the study represent a 30% (below aver- age), 50% (dry), and 70% (critically dry) effective reduction for periods ranging from 10 to 60 years, and were applied to the CDWR's California Central Valley Ground water- Surface Water Simulation Model (C2VSIM). The methodology used here is part of a series of analyses that allow for the decomposition and response term by term, allowing for a reduction - ist evaluation of the impacts of decreases in net sur- face flow from reservoirs and Central Valley precipitation. Previous studies of California's future water supply were based on downscaled climate model projections with hydrologic model simulations and permutations of the 1922 -1993 unimpaired streamflows (Miller et al., 2003) with an operating criterion of maximizing statewide water supply net benefits (e.g., Lund et al., 2003; Zhu et al., 2005; Tanaka et al., 2006; Medellin - Azuara et al., 2008). However, these studies are unable to pin down the term -by -term isolated response to droughts, present day or future. With that in mind, it was deemed JAWRA 858 JOURNAL OF THE AMERICAN WATER RESOURCES ASSOCIATION