EXPLORING TEMPORAL PATTERNS OF SELECTED SOLUTES DURING HYDROLOGIC EVENTS IN AN URBAN SETTING TO UNDERSTAND PATHWAYS

Understanding the impact of human activities and climate change on the health of streams relates to knowledge of the various pathways for solutes to enter the streams. Over his career, Wm. Berry Lyons has been a leader in studies of this type, and this presentation is dedicated to those efforts. The study of sodium and chloride in the environment has a long history, and there has been a particular focus on road salting in urban areas. Often temporal measurements are coarse (e.g., monthly), and thus our understanding of the hydrogeochemical dynamics is limited. Through sets of temporal measurements from the Red Cedar River (Michigan, U.S.A.), we explore the temporal pattern of dissolved Na, Cl, NO₃, SIO₂, K, E coli, and organic carbon concentrations over hydrologic events capturing both snowmelt and rain through salting seasons, and use concentration – discharge hysteresis loops to gain insight into pathways. Results show that 1) changes in Cl (and Na) concentrations during salting periods follow a characteristic “urban” pattern of first flush, followed by dilution and the recovery, 2) there is no first flush for Cl and Na in the non-salting periods and often Na is not diluted, 3) Na and Cl patterns can become decoupled (different environmental behaviors, other salt sources) for making the use of Cl/Na ratios (which are typically above 1) to identify a halite source complicated, and 4) comparison of the hysteresis loops to three component models shows the possible relative importance of groundwater, soil water and surface-event water pathways and that these pathways vary amongst individual solutes and can change for a solute over season. These results show the complexity of understanding and modeling the solute chemistry of rivers. Hysteresis loops may give insights to understanding this complexity and with more study (e.g., function of season, geo-location, climate) they may be used to inform quantitative models.