Applying Improved Surface Urban Hydrological Processes in the New Single Layer Urban Canopy Model over Houston

The accuracy of the single layer urban canopy model (SLUCM) within the Advanced Research Weather (ARW) Weather and Research Forecasting Model (WRF) is evaluated over Houston. Initially, five simulations were run from August 24-26, 2000, and the first four of these five simulations were run using the SLUCM in WRF Version 3.6. Sensitivity tests were conducted using the following planetary boundary layer (PBL) schemes: the YonSei University (YSU) scheme, the Mellor-Yamada-Janjic (MYJ) scheme, the Mellor-Yamada-Nakanishi-Niino 2.5 (MYNN 2.5) scheme, and the Boujeault-Lacarrere (BouLac) scheme. A fifth simulation was run that incorporated the BULK urban scheme and the YSU scheme. The accuracy of the SLUCM and BULK schemes was evaluated by comparing the model output with TCEQ (Texas Commission on Environmental Quality) air monitoring station observations. The results showed that while the model successfully captured the diurnal pattern of temperature and winds, it over predicted the near surface winds and temperatures over the city during the nocturnal time period. This suggests that the model was over exaggerating the urban heat island (UHI) effect. An analysis of the surface sensible and latent heat fluxes at each station suggests that these errors were occurring because not enough cooling was occurring in the model due to suppressed latent heat fluxes. This suppression of latent heat fluxes was most pronounced at the stations located in the city. This appears to be happening because of the oversimplification of urban hydrological processes within the SLUCM.

Recently, an improved SLUCM has been developed for WRF Version 3.7; this new SLUCM includes five new urban hydrological processes. Several new simulations were run using this new SLUCM. These simulations were run for the same time period as before, and the same four PBL schemes were used. The first four simulations utilized the new SLUCM with all of the urban hydrological processes set to off, and then four more simulations were conducted with four of the five urban hydrological processes set to on. In addition to this, four simulations were conducted using the BULK scheme in WRF. A final set of simulations were run from August 10-16, 2006 to validate the model's surface heat fluxes with observations taken during the Texas Air Quality Study (TexAQS) 2006 field program. A final evaluation will be made where the output of these simulations will be compared to TCEQ observations and wind profiler data to determine if the new SLUCM improves the WRF's predictions. It is expected that with the new urban hydrological processes, there should be more evaporation and cooling occurring over the urban surfaces in the city. As a result, the WRF's UHI effect should be closer to observations, and the error in the simulations should be less than the error that occurred when using the previous SLUCM. Lastly, an analysis will be made to determine which of the four PBL schemes makes the most accurate forecasts when used in conjunction with the new SLUCM.