Event-based analysis of concentration-discharge relationship during winter road salt application

It has been widely established that many freshwater systems within the northern United States are exposed to significantly high chloride concentrations during winter months. The exposure of these freshwater systems is primarily due to use of road salt as deicing agents to aid in the safe transport of people and goods during icy conditions, which migrate from impervious surfaces to affect soil and water quality and damage infrastructure. Although increasing chloride concentration is of vast importance to ecosystem health and function, limited studies seek to understand the methods by which event-based application of road salt propagates through stream and river networks. Many studies focus on longer time series and low-frequency data to explain the increased chloride concentration in streams, ignoring the dynamics of environmental variables on an event scale. Here, we present an event-based, concentration-discharge analysis during two winter seasons (2017-18 and 2018-19, between November and March) across six urban streams within the Saint Louis, Missouri, metropolitan region. We characterize events relative to chloride concentrations exceeding the EPA chronic threshold for chloride (230 mg Cl-/L) and those lasting more than 12 hours using a peak-trough analysis to identify when an event occurs in our time series. We then characterize each event using a concentration-discharge (cQ) relationship, including hysteretic behaviors, to understand how discharge conditions control freshwater systems' exposure during winter months. Lastly, we performed a classification and regression tree analysis between cQ relationship parameters and explanatory variables. Our results showed no statistically significant difference (p > 0.05) between either the cQ slopes or the intercepts relative to the individual sites but that minimum air temperature regulated cQ dynamics with temperature above 1.9 ℃ having the greatest number of chemostatic events while temperature below -2.2 ℃ had more dilution events; mobilization was mostly associated with temperatures above -1.9 ℃. Our analysis of rain on snow (ROS) events shows that most of the winter events observed across the study sites consisted of ROS events which also accounted for very high variability in concentration response to discharge. Together, our results reamplifies the complexity of solute transport to streams during colder months.

The files includes the raw data for chloride, an excel file with the USGS stream gauges, the raw cQ data used for our analysis and a readme file. Results from this data have been submitted with Dorley & Haake (2023) to Frontiers in Water