Date of Award
Master of Science
A great deal of attention has been given to global climate change by the hydrologic community. Temperature, precipitation and streamflow trend analysis, on different spatial and temporal scales, is important in understanding the impact of climate change. Midwest region is the heartland of agriculture production in U.S., and change in hydrologic variables may affect both quantity and quality of production. In the study, mean, maximum and minimum temperature along with mean precipitation for 106 climate divisions in the Midwest were analyzed to test the existence of monotonic trend and shift changes in the seasonal hydrological time series. In addition to that, trend and shift in 88 streamflow stations in the Midwest and its relation with temperature and precipitation were analyzed. Mann Kendall test with and without considering lag-1 auto-correlation were employed to analyze the trend. Non-parametric Pettitt test was used to analyze the shift; Sen’s slope estimator was used to identify the magnitude of hydrological trend. Discrete Wavelet analysis was done to analyze the effect of periodicities on trends and shifts in hydrological variables. In addition, association between the occurrence of shifts and phases of climate indices, such as El Nino Southern Oscillation (ENSO) and Pacific Decadal Oscillation (PDO), was investigated. The results indicated significant increasing trend in mean and minimum temperature for majority of the climate divisions in all the seasons studied. While, increase in maximum temperature in winter and spring was observed, majority of the climate divisions showed decreasing trend in summer and fall. Increasing trend in precipitation was detected mostly in spring, summer and fall as compared to winter. Persistence was mostly observed for all the variables during the summer season and when accounted for, trend remained for most of the climate divisions. Spatially prevalent shifts were noticed, which were in agreement with gradual trend for most of the hydrologic variables. The results of the wavelet analysis indicated D2 (dyadic scale of 4 years) and D3 (dyadic scale of 8 years) to be the most effective periodic component in detecting trend in winter, spring and summer. D1 (dyadic scale of 2 years) and D3 proved to be the most effective in detecting trend in temperature data in fall. Likewise, precipitation and streamflow showed the dominance of D3 component in detecting real trend in the data. Majority of shift changes coincided with PDO and ENSO phases. The use of wavelet helped in detecting the typical timescale of ENSO and the effect of coupled climate indices on hydrologic variables. A possible linkage between streamflow, temperature and precipitation trend across some regions were detected clearly corroborating the importance of exploring the synergism between meteorological, climatic and hydrologic factors to assess the changing character of the variables. The contribution from this research include a better understanding of the changes in the hydrology of the Midwest that can help in better water management decisions.
This thesis is only available for download to the SIUC community. Others should
contact the interlibrary loan department of your local library.