Evaluation of moisture and heat transport in the building-resolving urban transport code QUIC EnviSim

Evaluation of moisture and heat transport in the building-resolving urban transport code QUIC EnviSim

Kevin Briggs, Matthew Overby, Daniel Alexander, Rob Stoll, Pete Willemsen, Eric R. Pardyjak

QUIC EnviSim is an extension to the Quick Urban Industrial Complex (QUIC) Dispersion Modeling System. In addition to computing the dispersion of pollutants and contaminants, QUIC EnviSim is designed to compute the moisture, heat, and momentum fields within the urban canopy. The turbulence and momentum fields are provided by the QUIC Dispersion Modeling System, which may be driven by inputs from WRF data and experimental or user-defined velocity profiles. This transport model is coupled to a Land Surface Model (LSM), which is a coupled energy and mass balance system that uses Monin-Obukhov Similarity Theory (MOST) to compute turbulent fluxes at the land atmosphere interface, and a multilevel 1D column approximation for diffusion into all surfaces of the urban form. The driving component of the LSM is a radiation module, which uses the NVIDIAŽ OptiX high performance ray-tracing engine to compute the incoming solar heat flux and radiative heat exchange between urban surfaces. A primary objective of the project is to efficiently provide better than real time results. To accomplish this we utilize the parallel computing capabilities of graphics processing units (GPUs) in each component of QUIC EnviSim. In this presentation we will focus on the evaluation of the component of the modeling system that computes the heat and moisture transport in the urban canopy through comparing our results against experimental work. First, we will present an evaluation of advection across a step discontinuity in moisture and temperature over a flat surface by comparing our results to a classical experiment of a tarmac adjacent to an irrigated grassy patch. Next, we test the model's ability to compute the transport of heat from an isolated three-dimensional cube as well as an array of idealized buildings. Finally, the model's complete capabilities are evaluated against available full-scale observations from the Oklahoma City Joint Urban 2003 field campaign.