Adapting Backward Lagrangian Stochastic Footprint Modeling to Diverse Urban Scenarios
Yu Ding is currently a Ph.D. student in the School of Meteorology at the University of Oklahoma (OU), advised by Prof. Chenghao Wang. Before coming to OU, she completed her master’s degree in Hydrology and Water Resources at Hohai University, China. Her previous research focused on improving the accuracy of satellite precipitation data and integrating bias correction and machine learning algorithms to enhance data precision. Yu has an interest in utilizing remote sensing techniques and hydrological modeling. She is working on on developing an integrated high-resolution pollutant dispersion model over complex terrain (e.g., urban environments).
Abstract
Accurate determination of the source area associated with atmospheric measurements is essential for interpreting urban observations and improving model performance. This study applies a backward Lagrangian stochastic footprint model to a range of urban scenarios, highlighting its adaptability across different atmospheric stabilities, urban morphologies, and sensor locations. Within the canyon, the wind field is prescribed using an analytical vortex formulation that captures both streamwise and vertical velocity variations. This formulation offers a more realistic depiction of within-canyon turbulence compared to conventional approaches and aligns well with flow characteristics revealed by large-eddy simulation (LES) studies. Results show that footprint contributions vary significantly with canyon aspect ratios, stability conditions, and sensor height. Roofs dominate source areas for measurements above the canopy, while walls and roads contribute more prominently when sensors are located within street canyons. The model demonstrates strong potential for application to diverse urban measurement contexts and can be readily extended for dispersion analysis, making it a valuable tool for advancing urban environmental modeling.