Water quality impairments at urban beaches can be challenging to manage due to a wide range of possible pollutant sources, including both regional and local sources. In this study, we used high-resolution time series data as the input into a 3D hydrodynamic model to predict Escherichia coli (E. coli) and human Bacteroides marker concentrations at nearby beaches. Model results predicted that both organisms reach beaches following select wet weather events and during combined sewer overflows (CSOs), but in 50% of these events, the E. coli model predictions were <235 CFU/100 mL when human sources were present. Local pollutant sources were also evaluated during a variety of conditions using fecal source markers, including human, gull, and dog bacterial markers. Gull waste was frequently detected at beaches. In contrast, human markers were rarely detected, but when present, detection generally coincided with CSOs and low levels of E. coli, suggesting dilute regional contamination. Together, the results suggest that regional fecal pollution containing human sources does not coincide with elevated E. coli levels, and localized contamination likely drives E. coli levels at beaches. This work highlights the need to consider risk-based metrics rather than relying on a single easily measured indicator at beaches.