550-Year Reconstruction of Streamflow Variability in Spring Valley, Nevada

摘要

Spring Valley, Nevada, is one of several areas proposed for the pumping and export of ground-water to Las Vegas by the Southern Nevada Water Authority. Long-term annual-to-decadal variability of water supply in the region is not well understood, so tree-ring records were used to develop a longer baseline of variability in streamflow and drought episodes. Long-lived (up to about 600 years), climatically sensitive single-leaf pinyon pine (Pinus monophylla) trees within the Cleve Creek watershed provided a 550-year (1458-2007) tree-ring chronology that was compared to water-year mean monthly runoff from USGS gauge 10243700. Using a proxy record from within the watershed under study increased confidence in the statistical relationships used for streamflow reconstruction. Linear correlation between the tree-ring chronology and the streamflow record over 34 years of overlap was 0.73, explaining 53% of the instrumental variance. After comparison with multiple linear regression and linear regression with transformed data, the line of organic correlation (LOC) method was used to develop a streamflow reconstruction with water-year resolution from 1458 to 2007. During these 550 years, a total of 257 wet and dry episodes were quantified according to their duration, magnitude, and peak. The longest episode was 1848-1855 (an 8-year wet spell); the greatest magnitude belonged to the drought of the mid-1600s (1652-1655); the three highest peaks all corresponded to dry episodes, 1506-1508, 1590, and 1933-1936. Using a numerical scoring rule, the 1930s drought (1933-1936) was in eighth position, making it one of the most remarkable episodes in the past half millennium. This result is not entirely consistent with recent dendroclimatic reconstructions for the eastern Sierra, suggesting that regional drought severity varies by locality within the Great Basin. Evaluating the responses of trees at multiple elevations to various local climate and hydrological parameters through in situ monitoring will help refine tree-ring reconstructions of past ecohydrological conditions.