Characterizing turbulent exchange over a heterogeneous urban landscape

Much of the world’s population now resides within cities where altered energy use, building distributions, transportation networks, and surface characteristics influence land-atmosphere interactions, energy and water budgets, and carbon cycles, relative to rural areas. Knowledge of the surface properties that affect exchanges of energy and mass, as well as how exchanges change over time, is critical for accurate local weather and climate forecasting, and pollution dispersion modelling. One way of measuring flows of energy and mass over cities is through the use of eddy covariance (EC). This stationary approach has been implemented in many cities globally, and has contributed greatly to knowledge of the exchange between the urban surface and the urban atmosphere. However, EC was developed for ecosystems like forests, where source/sink distributions are horizontally homogeneous; This surface uniformity does not usually pertain to cities, where sources of heat, water, momentum, and trace gases exhibit spatial heterogeneity. Eight years (2008 - 2016) of continuous EC flux measurements over a residential neighbourhood in Vancouver, BC, Canada, were used to characterize the relationship between surface source/sink heterogeneity and the efficiency of turbulent exchange of heat, water vapour, momentum, and carbon dioxide (CO2) (represented by the correlation coefficients r_wT, r_wh, r_uw, and r_wc, respectively). Using a combination of remotely-sensed satellite and light detection and ranging (LiDAR) imagery, geospatially-referenced land cover data, traffic densities, and source area modelling, exchange efficiencies were examined seasonally, diurnally, as a function of atmospheric stability, and in terms of distinct, spatially-variable surface cover attributes. Transport of momentum and scalars exhibited varied dependencies that resulted in dissimilar exchange efficiencies; r_wT was primarily moderated by stability, time of day and year, and surface patchiness, r_uw was mostly affected by stability and surface roughness, and r_wh and r_wc were mostly affected by surface patchiness. As source/sink heterogeneity increased, exchange became less efficient. Competing sources and sinks acting simultaneously on a turbulent entity resulted in an exchange efficiency closer to zero. Under stable conditions, r_wT, r_wh, r_uw, and r_wc depended mostly on stability, while surface heterogeneity contributed more to dissimilarities between momentum and scalar exchange efficiencies under unstable conditions.Arts, Faculty ofGeography, Department ofGraduat