Assessing the role of culvert sediment storage an urbanized watershed

Abstract

Watershed urbanization has profound impacts on sediment dynamics. Land development introduces novel pathways of sediment storage and remobilization to the stream environment, notably in the form of culverts. This study sought to contextualize the role of culvert sediment storage and remobilization through the lens of fluvial geomorphology and to explore the interaction of culverts with sediment dynamics in an urbanized watershed. This was performed by assessing characteristics of culvert sites that influence sediment storage and by comparing the quantity and type of sediment stored in culverts to sediment associated with other forms of storage and sediment transport. A combination of remote- and field-based techniques were used to study culvert sediment storage, natural sediment storage in channel bar deposits, channel erosion, and sediment storage and remobilization associated with high-magnitude precipitation events in the heavily urbanized Dead Run watershed near Baltimore, Maryland. Additional data sourced from a previous study in this watershed were also included to incorporate analyses of suspended sediment transport.The total mass of sediment stored throughout culvert sites in Dead Run was 749 metric tons. Regression analyses performed using sediment storage mass and culvert site characteristics indicated that storage was driven by culvert size and slope, though the predominant factor was the degree of slope change as the stream channel enters the culvert. This result suggests that the alteration of hydraulic conditions at the culvert is the primary driver of culvert sediment deposition. Comparisons of sediment size fractioned-masses from culverts to those of other geomorphic features and processes also suggested that, while culverts predominantly stored coarse sediment at approximately 50% sand and 47.5% gravel, culverts stored proportionally more sand and fine sediment compared to natural channel bars (around 0.3 tons per meter of channel length). Culvert deposits showed different responses to large precipitation events and associated high flows, with net sediment storage and remobilization occurring across different sites. However, fine sediment in culverts was observed to be in overall lower abundance throughout the watershed following the storm events, with an average loss of very fine sediment (<125 µm) at 68%. Overall, it can be inferred from this study that culvert sediment storage is primarily induced by alterations to local hydraulic conditions, though the ways in which culvert sediment is stored and remobilized vary depending on sediment size. Culvert sediment storage was observed to be quite dynamic regarding both magnitude and sediment size over the short time scale of this study Ð a trend that warrants further investigation and incorporation of this form of storage into broader-scale urban sediment studies.