Date of Award

2016

Type

Thesis

Major

Master of Science

Department

Earth & Space Science

Abstract

Drought is a prolonged reduction in precipitation which may result in significant effects on local ecosystems, watersheds, and water sources, with the potential to have negative impacts on human populations. Normalized indexes can be used to characterize drought magnitudes and frequencies using historical data records on temperature, rainfall, snowpack, soil moisture, streamflow, and/or other water supply indicators. This study examined the applicability of two drought indices: Standardized Precipitation Index (SPI) and Streamflow Drought Index (SDI). The SPI and SDI estimate the magnitude of observed rainfall (SPI) and discharge (SDI) deviations from the longterm average. In this study, SPI and SDI were used to determine droughts from 1970-2000 in 7 North Georgia catchments. Because other studies have shown the importance of time on index performance, SPI and SDI were calculated on data aggregated for 1, 3, 6, 9, 12, and 24 month intervals. The performance of the indices was evaluated using the strength of the relationship between SPI and observed discharge and between SPI and SDI using linear regression. The goodness of fit was measured by a statistic referred to as the coefficient of determination (r2). The r2 is the amount of variation in the dependent variable explained by the variation in the independent variable. Twoway Analysis of Variance (ANOVA) models were conducted to determine how watershed and temporal scale (independent variables) influenced the SPI-discharge and the SPI-SDI regression relationships. The dependent variable in these models was the coefficients of determination estimated using the linear regressions. Hydrological drought severity was analyzed on an annual basis and revealed the following trends for SPI and SDI: 1) The timing of severe drought events varied slightly across watersheds; 2) the most extreme drought events occurred in 1978, 1981, 1986-1988, and 1999-2000; and 3) 1986- 1988 events were similar in severity to the 1999-2000 drought events. Watersheds varied temporally in their hydrologic response to drought, which may be a function of localized recharge or anthropogenic factors. Based on i^s analysis, discharge and SPI had a more robust relationship with longer time steps (9 - 24 months); SPI and SDI relationships revealed an analogous pattern. Regions deficient in discharge data can use SPI as a viable alternative to SDI; whereas, SDI can VI function as a proxy for SPI in catchments lacking precipitation data at timescales exceeding six months.

Download
DAISY

Share

COinS