Investigating the effect of urbanization-induced precipitation variability on hydrological response using mesoscale meteorological model output linked to SWMM5

Shannon Reynolds, Steven J. Burian, J. Marshall Shepherd and Michael Manyin


This study explores the relative effect of urban-induced precipitation variability on the hydrological response of a large urban watershed. The influence of dynamic and aerosol processes on convective activity in a regime dominated by urban-induced and sea-breeze processes were studied using land surface-atmosphere modeling. The July 25, 2001 case in Houston, Texas selected for simulation offered an ideal case day to examine the factors leading to urban-induced convection. Results indicated vertical motion induced by surface convergence and sensible heat flux are enhanced over urban areas relative to non-urban areas.

An investigation of the effect of the perturbed precipitation variability on the hydrological response was completed to ascertain the significance of the urban-induced variability for stormwater management, flood control, and other urban hydrology objectives. Simulated rainfall fields from the mesoscale meteorological model (MM5) for land surface conditions representing non-urban characteristics and conditions representing present-day Houston served as input to a calibrated SWMM5 watershed model for the Greens Bayou watershed located north of downtown Houston, Texas. After gage adjustment, the two simulated rainfall datasets were ingested into the SWMM5 hydrologic model representing both urban and pre-developed land surface conditions.

Results of this study show increasingly localized, higher-intensity storms without increasing rainfall volumes over the watershed area. Shifting of the storm downwind of the urban area was also evident. The hydrologic response for urban rainfall simulations within SWMM5 representing present-day urban conditions show higher peak discharges and decreased tails as compared to pre-developed simulated rainfall over the same SWMM5 land surface conditions. The study results imply high resolution rainfall data combined with local urban area characteristics provide necessary elements for urban stormwater management design to more accurately simulate urban environments.

Permanent link: