H102-05
Trends and drivers of changing stream intermittency across the United States

Thursday, 10 December 2020: 17:42
Virtual
Samuel C Zipper1, Kendra Elena Kaiser2, Sarah Godsey3, John Christopher Hammond4, C. Nathan Jones5, Joanna Blaszczak6, Margaret Shanafield7, Katie H Costigan8, Margaret A Zimmer9, Daniel C Allen10, Thibault Datry11, George H. Allen12, Kate S. Boersma13, Walter K Dodds14, Stephanie K Kampf15, Corey Krabbenhoft16, Meryl Mims17, Julian D Olden18 and Adam N Price9, (1)University of Kansas, Kansas Geological Survey, Lawrence, United States, (2)Boise State University, Boise, ID, United States, (3)Idaho State University, Department of Geosciences, Idaho Falls, ID, United States, (4)U.S. Geological Survey, Maryland-Delaware-D.C. Water Science Center, Catonsville, MD, United States, (5)University of Alabama, Biological Sciences, Tuscaloosa, AL, United States, (6)University of Nevada Reno, Natural Resources and Environmental Sciences, Reno, NV, United States, (7)Flinders University, Science and Engineering, Bedford Park, SA, Australia, (8)University of Louisiana at Lafayette, Lafayette, LA, United States, (9)University of California Santa Cruz, Santa Cruz, CA, United States, (10)University of Oklahoma Norman Campus, Biology, Norman, OK, United States, (11)INRAE Lyon, Villeurbanne Cedex, France, (12)Texas A&M University, Department of Geography, College Station, TX, United States, (13)University of San Diego, Biology, San Diego, United States, (14)Kansas State University, Manhattan, KS, United States, (15)Colorado State University, Department of Ecosystem Science and Sustainability, Fort Collins, CO, United States, (16)University at Buffalo, College of Arts and Sciences and RENEW Institute, Buffalo, NY, United States, (17)Virginia Tech University, Department of Biological Sciences, Blacksburg, VA, United States, (18)University of Washington, School of Aquatic & Fishery Sciences, Seattle, WA, United States
Abstract:
Non-perennial streams and rivers account for over 50% of global river network length in recent estimates, but are understudied relative to their perennial counterparts. Here, we quantify trends in intermittency and potential drivers of observed changes across the contiguous United States (CONUS) from 1980 - 2018. Specifically, we evaluate intermittency signatures describing the no-flow duration, timing of the onset of no-flow conditions, and drying rate of 540 non-perennial streams. We find that the no-flow duration is increasing at up to 4% per year and the onset of no-flow conditions is getting earlier in the year on average across CONUS. However, there is great inter- and intra-regional variability in the observed trends. For example, the date of first no flow is getting later in the year in the Northern Great Plains region but earlier in the other five ecoregions investigated. A subset of gages have noticeably nonlinear changes in intermittency through time, including both shifts from perennial to non-perennial and non-perennial to perennial conditions. We develop random forest models to quantify the drivers of spatial and temporal variability in each of the three intermittency signatures. We find that the ratio of precipitation to potential evapotranspiration in a given climate year and/or the preceding year are the dominant controls over each of these three signatures, but basin physiographic characteristics such as elevation, drainage area, and soil permeability also have a significant effect. Overall, our results suggest there are significant and widespread changes in stream intermittency across CONUS, and that the impact of climate change is influenced by basin characteristics and anthropogenic influences at local to regional scales.