Exploring Hydroclimatic Variability on the Red River of the North, the Least Stationary River of the United States

Abstract

Nation-wide surveys of hydrological trends have described the Red River of the North as the least stationary river in the continental United States. In this study, I examine first, how the Red River has varied across the spectrum of high, median, and low flows and second, how mean climatology across the watershed and the broader region has behaved before and after an abrupt increase in streamflow. For those gauges with more than 50 years of daily observations, minimum flows on the Red River increased between 3% and 28% per year, median flows from 3% to 10% per year, and maximum flows from 1% to 5%. For most time periods tested (315/420), discharge on the Red River had a positive and statistically significant (p≤0.05) increase. There were no significant trends for tests with start years after 1980. Change-point analysis revealed a cluster of upwards shifts in basin-wide streamflow in the early 1990s. After 1992, daily flows on the Red River have been higher throughout the year and the river has exhibited novel secondary peaks in June and November. Since the onset of this new regime, the northeastern Great Plains have been both warmer and wetter. These changes have been most prominent in the summer and fall, with temperatures 0.6°C to 1.0°C warmer and rainfall increasing between 39 to 85 mm. I expected the observed increase in regional temperature might have caused the spring snowmelt pulse to happen earlier, but the daily flow data did not show any significant changes in the timing of peak flow on the river. The coincidence between higher flows in the Red River and warmer and wetter conditions after the early 1990s suggests the strong non-stationary behavior of the river could be due, at least in part, to climate change and variability.