|dc.description.abstract||Nutrient pollution is a primary cause of water quality impairment in streams in the United States and throughout the world. Regulatory approaches under the Clean Water Act, such as water quality standards and the Total Maximum Daily Load program, aim to improve water quality. In this study, novel probabilistic methods are developed to characterize vulnerability to nutrient pollution along urban streams and to assess risk of water quality impairment under varying hydrologic conditions. Vulnerability is defined as the probability that ambient conditions exceed desired water quality standards. Both EPA ecoregional and state-level targets are included in the analysis. Specifically, the study i) explores relationships between urban influences and risk to nutrient pollution; and ii) expands on the load duration curve framework to quantify vulnerability to nutrient pollution as a function of flow exceedance probability. The study objectives are examined at 20 stream locations in four ecohydrologically different regions across the United States, including Denver, CO; Phoenix, AZ; Portland, OR; and Baltimore, MD. Total phosphorus (TP) and total nitrogen (TN) water quality data collected between 1990 and 2018 with daily discharge measurements are utilized in the analysis. Indicators of urban influence include wastewater treatment capacity, urban land cover, impervious surfaces, and population density. In general, study locations exhibit vulnerability (greater than 5%) to nutrient impairment across urban gradients, including some relatively undisturbed monitoring locations. Nearly 30% of TP sites and 45% of TN sites are impaired under state level regulation. Results indicate that incorporation of more stringent EPA ecoregional targets lead to higher vulnerability estimates than those corresponding to the state-level targets. Over 70% of TP sites and 55% of TN sites with state level standards are characterized as vulnerable (greater than 5%) when EPA goals are considered. Patterns of impairment through urban gradients are more evident in arid regions with wastewater-dominated river flows, specifically in Denver and Phoenix, than humid regions. Multiple linear regressions between indicators of urban influence and vulnerability provide strong (R2 > 0.7) relationships for most monitoring locations. Inverse distance weighted annual wastewater treatment facility flow capacity and urban land cover are the most significant predictors. However, the most important nonpoint source exploratory variable differ from site to site. More monitoring locations are required to determine model significance. In addition, assessment of nutrient pollution vulnerability using the enhanced load duration approach show that higher vulnerability to impairment tends to occur under consistent hydrologic conditions within each city. For example, high vulnerability to TN and TP impairments are observed under low flow conditions at sites within and around the Denver incorporated area. Conversely, nutrient levels during high flow conditions are more likely to exceed the TN and TP standards in Phoenix, Baltimore, and Portland. Many locations are vulnerable to nutrient pollution (greater than 5%) under all possible flow scenarios, especially at downstream monitoring locations. Approximately 85% of TP sites and 70% of TN sites are vulnerable under all flow conditions assuming EPA water quality goals. The methodology developed in this study can be used to probabilistically quantify the vulnerability to water quality impairments in streams and to identify hydrologic conditions under which higher vulnerabilities prevail.