Nitrogen (N) is one of the most important pollutants on urban road surfaces. Understanding the N deposition forms, load characteristics, and influential factors can help to provide management and control strategies for road stormwater runoff pollution. This study focuses on a highly urbanized area in Guangzhou, China, and presents the characteristics of both dissolved and particulate N deposition forms as well as their correlations with land-use types and traffic factors. In addition, an artificial neural network (ANN) based classification model is utilized to estimate N pollution hotspot area and total nitrogen (TN) flux from road to receiving water bodies. The results showed that N on urban road surfaces mainly existed in the form of particulate organic nitrogen. Land use types dominated by residential area (RA) and urban village (UV) have higher TN build-up loads. Geodetector analysis indicated that land use has a greater impact on nitrogen build-up loads than traffic factors. Through classification and estimation using the ANN model, RA, and UV were classified as hotspot areas, and the TN flux from roads in the study area was calculated to be 3.35 × 105 g. Furthermore, it was estimated that the annual TN flux from roads in Guangzhou accounts for 19 % of the city's total urban domestic discharge. These findings are expected to contribute to the pollution control of stormwater runoff from urban road surfaces and provide valuable guidance for enhancing the ecological health of urban water environments.
Read full abstract