Hydrodynamic Simulations for Trash Loading in Southern California’s Dense Urbanized Watersheds

Abstract

Waterways and water bodies worldwide are impacted by anthropogenic litter (hereafter “litter” or “trash”), generated from nonpoint sources. This study analyzes litter loads across various land uses within two Los Angeles County watersheds: the Ballona Creek and the Los Angeles River. Our objective is to present a methodology to develop buildup and washoff parameters for densely populated urban areas, such as the coastal metropolitan area of Los Angeles, California. An assessment is also made to test how these model parameters perform when applied to another urbanized watershed with similar rainfall and climate patterns (i.e., the San Diego River Watershed in California). Using extensive litter collection data from small drainages (approximately 572 locations, each draining 0.05–8.5 ha), we estimate buildup and washoff model parameters. These parameters are used to simulate the buildup of litter and subsequent washoff (load) of litter in our selected watersheds. Simulated results are validated against observed data from different years in all three watersheds. To date, no study has used litter washoff data to develop these parameters for stormwater models and applied them on a regional scale. We compared linear and nonlinear power functions of litter buildup during interstorm periods. Although there were limited data to develop washoff parameters, an exponential washoff function was used and calibrated to the observed washoff. Generally, storm events with the greatest antecedent dry days had the largest litter loading. Buildup rates varied among land uses, and key calibration parameters included the maximum buildup, buildup rate constant, washoff exponent, and washoff coefficient. A parameter sensitivity analysis was conducted to reveal the washoff exponent as the most sensitive parameter. Annual litter loading estimates were 4.60 kg/ha for the Ballona Creek and 11.50 kg/ha for the Los Angeles River. Litter loading estimates for Ballona Creek were validated and calibrated to align with observed data from the Ballona Creek Trash Interceptor, resulting in an annual washoff load of 2.1 kg/ha. The estimated annual litter loading for the Lower San Diego River was 2.15 kg/ha falling between the mean (0.77 kg/ha) and the maximum (3.56 kg/ha) observed values. When applying model parameters from one watershed to another, models require extensive calibration and validation data, as extrapolating data between watersheds can introduce errors. This research can inform stormwater modelers about the process for developing parameters to estimate water quality in numerical modeling software. We demonstrate how data from field sampling can be directly used to develop model parameters. Results show that with proper validation data, stormwater models can accurately simulate annual litter loads in modeled watersheds. Future studies can test model parameters in other urbanized watersheds to validate the potential for standardized buildup values. Field studies can be done in conjunction with the results from this survey to investigate and further validate litter loads mobilized by stormwater. This study highlights the need for long-term monitoring to improve and update models in the future. Although there is a risk of inaccurate results when extrapolating model parameters from one watershed to another, this research indicates the potential for developing buildup and washoff parameters that can be used as a starting point for parameter selection across urban watersheds.