Abstract Urban noise pollution is a widespread environmental stressor that adversely affects residents’ quality of life. Although environmental factors such as urban greenery are known to influence noise perception, existing studies largely focus on static spatial patterns or specific noise types, with limited attention to the spatiotemporal dynamics of multiple determinants and their evolution during major societal disruptions (e.g., the COVID-19 pandemic). To address these gaps, this study introduces a novel approach that integrates a rolling time-window mechanism with Geographically Weighted Poisson Regression to investigate environmental factors influencing urban noise and their spatiotemporal dynamics. Leveraging New York City’s 311 complaint records alongside multi-source urban data, we systematically analyze how built environment, urban functional composition, and morphological characteristics collectively shape the spatiotemporal patterns of various noise complaint types. Our findings demonstrate siginificant spatiotemporal variations in how these factors shape noise complaints, with environmental features exerting a markedly stronger influence on noise perception than functional or morphological attributes. Notably, green indicators exhibited divergent effects on residential noise complaints. Green view index (GVl) consistently attenuated complaints, with each 1% increase linked to a 2.02% drop under normal conditions and a 4.69% drop during the COVID-19 pandemic. In contrast, the Normalized Difference Vegetation Index (NDVI) showed a weaker association and, in some contexts, was even linked to higher complaint rates, indicating that different dimensions of urban greenness may exert heterogeneous, context-dependent effects on perceived noise. Building on these findings, spatial clusters of noise response patterns were identified, and tailored management strategies were developed. Overall, this study offers scientific insights and policy guidance for the refined management of urban noise environments.
Gong et al. (Wed,) studied this question.