Trends in Quality Controlled Precipitation Indicators in the United States Midwest and Great Lakes Region

Abstract

Changes in precipitation can have broad and significant societal impacts. A number of previous studies that analyzed changes in precipitation across the Great Lakes and Midwest for a variety of time periods and using a range of quality-control standards and methods observed increased precipitation rates and totals, although there was considerable site-to-site variability, even for sites in close physical proximity. Biases and discontinuities in precipitation observations may contribute to this variability. This study identifies and examines changes in precipitation utilizing a unique approach to observation series screening over a region encompassing the Great Lakes and broader Midwestern region of the United States for the period 1951–2019. A multiple tier procedure was utilized to identify high quality input data series from the Global Historical Climatology Network-Daily dataset. Annual and seasonal time series of precipitation indicators were calculated and subjected to breakpoint analysis as further quality control. Trends were analyzed across a broad range of related indicators, from totals and frequencies of threshold events to event duration and potential linkages with total precipitable water. Results indicate that annual precipitation has generally increased across the region in terms of totals, although there is substantial variation across the study domain in the significance and magnitude of annual trends by indicator. Annual trends were spatially most consistent across eastern areas of the study domain while relatively greater station-to-station variability in trend significance and magnitude was observed across northern and western portions. Significant trends were generally fewer in number for seasonal precipitation indicators and less spatially coherent. The greatest number of significant trends occurred in fall with the fewest in spring. Correlation of indicator trends with trends of mean total precipitable water suggests weak correlations annually and moderate correlations at the seasonal scale. The trends of the precipitation indicators in our study exhibited more coherent spatial patterns when compared with studies with different quality control criteria, illustrating the importance of quality control of observations in climatic studies and highlighting the complexity of the changing character of precipitation.