Estimating Aerodynamic Parameters in Heterogenous Urban Environments Using High-Resolution Land Surface Data

The roughness length (z₀) and displacement height (z_d) are essential aerodynamic parameters in numerical weather and climate models. They describe the surface-layer mean-wind profile and affect the characteristic length scales of turbulent momentum, heat, and mass exchanges between the surface and atmosphere. In heterogeneous urban environments, the values of z₀ and z_d vary considerably across space. The dependence of z₀ and z_d on roughness element morphologies is primarily based on wind tunnel or numerical modeling data. Few studies have tested such empirical relationships against measurements collected in urban environments. In this work, we evaluate methods of estimating z₀ and z_d using high-resolution land surface data and measurements collected on two eddy-covariance flux towers (AmeriFlux US-INc and US-INg) associated with the ongoing Indianapolis Flux Experiment (INFLUX) in Indianapolis, IN. These two towers within the city represent different forms of urbanization. US-INg sits between a major highway and a vegetated suburban neighborhood. In contrast, US-INc sits over a network of intersecting highways surrounded by buildings of various shapes and sizes. We quantify the roughness elements (i.e., vegetation and built structures) in a four-square-kilometer domain surrounding each tower using square-meter resolution LIDAR data made publicly available by the United States Geological Survey. We use the Flux Footprint Prediction (FFP) from Kljun et al. (2015) to estimate the upwind area of influence for each half-hour observation. Element geometries within a footprint area are used to estimate z₀ and z_d. We refer to this as the morphometric method for estimating these parameters. We also estimate z₀ and z_d using two anemometric techniques and compare these to the morphometric estimates. Results show large discrepancies between estimated aerodynamic parameters depending on the morphometric methodology and highlight the importance of including vegetation in urban areas where vegetation is abundant. The large discrepancies between anemometric and morphometric methods can be partially attributed to uncertainties associated with anemometric methods estimating aerodynamic parameters during convective regimes. Finally, we identify relationships between the roughness element morphologies and turbulent fluxes measured at the tower. Results from this study will guide future research under the INFLUX and Baltimore Social and Environmental Collaborative (BSEC) projects to develop improved morphometric methods for models such as WRF and E3SM with the goal of better simulating urban environments.