HOW DO SEDIMENT AND NUTRIENT LOADING FROM BANK EROSION VARY WITHIN A WATERSHED?

Streambank erosion can be a significant or even dominant source of sediment and phosphorus (P) leading to water quality impairment in some watersheds. Identifying and protecting some streambanks from erosion could therefore contribute to sediment and nutrient reduction and consequent water quality improvement. However, if resources are limited, where in a watershed should those resources be targeted to maximize reduction of sediment and P inputs? In headwater reaches which are numerous and often assumed to have net erosion, or in higher-order downstream reaches with higher lateral migration rates? A few studies have identified spatial patterns of bank erosion within watersheds, but there remains no framework for systematically evaluating whether these case studies are typical or anomalous.

In this study, we use the Horton-Strahler laws of drainage network composition to arrive at a generalized scaling relationship between stream order and channel length. We then define an erosion rate per unit channel length that varies functionally with stream order and with adjustable parameters describing eroding length fraction, average bank height, and compensating deposition. The product of channel length and the erosion rate expression for each order yields a function that expresses the expected yield of bank-derived sediment to the watershed outlet by stream order. Parameter values for the Horton-Strahler laws are drawn from the literature and from selected watersheds in Iowa. These values are then used to constrain trends in the distribution of erosion across stream orders that would lead to net streambank sources increasing or decreasing with order. Comparison with measurements from watersheds in Iowa suggests that sediment and P loads may often be dominated by either first-order channels or high-order reaches, depending strongly on drainage network morphometry.