Seasonal streamflow extremes in Texas river basins: Uncertainty, trends, and teleconnections

This study uses long-term daily streamflow records from selected river basins in the state of Texas. Four kinds of seasonal streamflow extremes based on scales (1 and 7 day) and 95th and 97.5th percentile threshold levels were analyzed for pre-industrial period (1925-1964) and postindustrial period (1965-2003) in order to assess possible changes in uncertainty, trends, and their teleconnection to large-scale climate indices. Marginal entropy was used to investigate the uncertainty and its relation to 1 day seasonal streamflow extremes where more uncertainty was observed in the postindustrial period. Wavelet transform was performed on seasonal streamflow extremes to evaluate changes in their periodicity by increasing the scale length of extremes as well as threshold extremes. Significant power changes from one season to another, based on wavelet analysis of seasonal streamflow and also variability, were observed within similar river basins. Cross-correlation analysis and cross-wavelet transform were employed between seasonal climate indices based on El Nino-Southern Oscillation index, Pacific Decadal Oscillation index, and Southern Oscillation Index with respect to seasonal streamflow extremes. The maximum change in correlation strength based on regional scale was observed in the 97.5th percentile winter extreme with winter Nino 3.4 index. Further, trend analysis of seasonal streamflow extremes was carried out using the Mann-Kendall test for pre-industrial and postindustrial periods to determine possible increasing or decreasing patterns, and the results suggested that significant positive trends were observed at some stations only in the postindustrial period.