Catch and Release: A dense, longitudinal array of water quality sondes reveals spatial and temporal complexities in suspended sediment flux

Excessive suspended sediment in streams is one of the most common causes for industrial, ecological and recreational stream impairment in the US. Identifying the primary geomorphic or anthropogenic sources of sediment is a key step in the effective mitigation of impairment. This study seeks to identify sources of suspended sediment in an agriculturally impaired watershed, Marsh Creek, in southeast Idaho. We employ thirteen multi-parameter water quality sensors to simultaneously measure stage, turbidity, temperature and conductivity every 15 minutes over a full calendar year. Examined at both the event and annual scale, these data enable mass balance calculations for mainstem and tributary contributions. Revealed in this monitoring is an approximately eight-fold longitudinal increase in sediment flux over 74 km that is largely augmented by eroding mainstem banks in reaches with higher stream power in the lower 30 km, with less than 20% contributed from tributaries. Independent data confirming the bank source were acquired through cost-effective sediment fingerprinting using 15N and C:N signatures from potential soil endmembers. Additionally, Google Street View-type longitudinal imagery of banks was collected via a kayak survey to confirm the spatial extent and magnitude of bank erosion along Marsh Creek. These data converge on bank erosion as the primary source of fine sediment. Sediment load at various hierarchical temporal and spatial scales is impacted by in-stream storage and remobilization, especially over shorter timescales ranging from daily to seasonal periods. Once averaged over the annual scale, local, temporary in-channel storage is overcome and these data reveal source reaches that can be prioritized for restoration and mitigation projects.