Assessing and quantifying changes in precipitation patterns using event-driven analysis

Authors: GUZMAN, JORGE - University Of Oklahoma; CHU, MARIA - University Of Illinois; Steiner, Jean; Starks, Patrick

Submitted to: Journal of Hydrology
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 11/16/2017
Publication Date: 2/1/2018
Citation: Guzman, J.A., Chu, M.L., Steiner, J.L., Starks, P.J. 2018. Assessing and quantifying changes in precipitation patterns using event-driven analysis. Journal of Hydrology. 15: 1-15. https://doi.org/10.1016/j.ejrh.2017.11.006.
DOI: https://doi.org/10.1016/j.ejrh.2017.11.006

Interpretive Summary: Studies have claimed that climate change may adversely affect precipitation patterns by increasing the occurrence of extreme events. The effects of climate change on precipitation is expected to take place over a long period of time and will require long-term data to demonstrate. Frequency analysis of precipitation event characteristics can help to quantify changes in precipitation patterns that are commonly masked by data aggregation when using annual time series. The objective of this study was to assess and quantify changes in precipitation patterns using 55-years of 5-minute precipitation data in Oklahoma. Precipitation datasets collected since 1961 by USDA-ARS Grazinglands Research Laboratory in the 620 km2 Little Washita River Experimental Watershed (LWREW) were used in this study. Precipitation events were characterized for more than 20 sites in the watershed. Time series analysis was conducted on total annual precipitation while frequency analysis was performed on parameters derived from 5-minute precipitation events. Statistical tests were used to evaluate the consistency and homogeneity of the time series. Monthly empirical cumulated frequency distributions were used to quantify changes in precipitation patterns and changes were tested by the Kolmogorov-Smirnov test. Although the total precipitation per event remained within the expected variability, a decrease in the number of precipitation events per site in the last 20 years was observed. There was no indication of increases in event precipitation intensity greater than 16 mm in 5-minute. However, 5-minute intensities in the range of 1 to 8 mm were more frequent while between 0.25 to 1 mm less frequent during events resulting in an annual average of 27% increase in maximum precipitation intensity. Also, events became on average 5.5 minutes shorter. A decrease in the number of precipitation events, as well as precipitation intensification can have important implications in crop production, exacerbate the fate and transport of constituents, increase soil erosion, and result in increased volume of groundwater extractions.

Technical Abstract: Studies have claimed that climate change may adversely affect precipitation patterns by increasing the occurrence of extreme events. The effects of climate change on precipitation is expected to take place over a long period of time and will require long-term data to demonstrate. Frequency analysis of precipitation event characteristics can help to quantify changes in precipitation patterns that are commonly masked by data aggregation when using annual time series. The objective of this study was to assess and quantify changes in precipitation patterns using 55-years of 5-minute precipitation data in Oklahoma. Precipitation datasets collected since 1961 by USDA-ARS Grazinglands Research Laboratory in the 620 km2 Little Washita River Experimental Watershed (LWREW) were used in this study. Precipitation events were characterized for more than 20 sites in the watershed. Time series analysis was conducted on total annual precipitation while frequency analysis was performed on parameters derived from 5-minute precipitation events. Statistical tests were used to evaluate the consistency and homogeneity of the time series. Monthly empirical cumulated frequency distributions were used to quantify changes in precipitation patterns and changes were tested by the Kolmogorov-Smirnov test. Although the total precipitation per event remained within the expected variability, a decrease in the number of precipitation events per site in the last 20 years was observed. There was no indication of increases in event precipitation intensity greater than 16 mm in 5-minute. However, 5-minute intensities in the range of 1 to 8 mm were more frequent while between 0.25 to 1 mm less frequent during events resulting in an annual average of 27% increase in maximum precipitation intensity. Also, events became on average 5.5 minutes shorter. A decrease in the number of precipitation events, as well as precipitation intensification can have important implications in crop production, exacerbate the fate and transport of constituents, increase soil erosion, and result in increased volume of groundwater extractions.