Search

Anthropogenic macro debris, otherwise known as trash, in urban streams is a persistent environmental problem throughout the world. The factors influencing trash accumulation within US urban stream channels has not been thoroughly investigated, with previous studies focusing largely on surrounding land use. I hypothesized that fluvial geomorphology is a better predictor of trash concentration than GIS-derived variables and that increased geomorphic complexity of stream reaches is associated with higher trash concentrations in urban streams. To test these hypotheses, I examined the relationships between in-stream trash concentrations and different geomorphologic characteristics, catchment characteristics, and catchment and riparian land cover in Cleveland (Ohio) and Charlotte (North Carolina) streams. I collected data on trash (>5 cm), large wood, cross-section and longitudinal profiles, and sediment sizes in 24 urban stream reaches, and calculated 19 geomorphic complexity metrics. Catchment characteristics and land cover variables were derived using ArcGIS Pro. Single and multiple regression approach have been used to find relationships between trash concentration and predictor variables. Trash concentrations ranged from 0.18 to 4.7 pieces/m bankfull width, with an average of 1.55 pieces/m. 71.8% of the collected trash was plastic. More variation in trash concentrations across stream reaches was explained through geomorphologic variables as compared to GIS variables when city-specific multiple linear regression models were made. When data from the two cities were combined, the opposite was true. Cross-section characteristics, especially bankfull width, were amongst the strongest predictors of trash concentration in both cities. Longitudinal profile and sediment characteristics were not predictive of trash concentration for Cleveland streams, and these data were not collected in Charlotte. Therefore, my hypothesis that more trash will be found in streams with more geomorphic complexity was not supported; instead, cross section variables were found to be more important in predicting trash concentrations. Specifically, our analysis suggests that roughness associated with stream banks plays an important role in trash accumulation. Our findings also suggested that there is no generalizable relationship across US cities between trash concentrations and specific predictors, possibly as a result of cross-city variation in development patterns and geologic setting.