NCSU Institutional Repository >
NC State Theses and Dissertations >
Please use this identifier to cite or link to this item:
|Title: ||Urban Land-surface Impacts on Convective Thunderstorm and Precipitation Characteristics|
|Authors: ||Pyle, Patrick Chase|
|Advisors: ||Dr. Dev Niyogi, Committee Co-Chair|
Dr. S. Pal Arya, Committee Co-Chair
Dr. Viney Aneja, Committee Member
|Keywords: ||surface atmosphere interactions|
urban land surface
urban heat island
Mesoscale Model Five (MM5)
Weather Research and Forecasting (WRF) model
|Issue Date: ||1-Nov-2006|
|Discipline: ||Marine, Earth and Atmospheric Sciences|
|Abstract: ||A six year storm climatology in the Indianapolis, Indiana region is investigated using base reflectivity radar data. Severe storm reports provided by the Storm Prediction Center (SPC) are used to select relevant thunderstorm cases to study. Storm composition change was noted in urban and rural environments and further statistical analysis is performed to view the overall effect of the urban region on storm characteristics. A specific thunderstorm case on 13 June 2005 is further examined using the fifth-generation NCAR/Penn State Mesoscale Model (MM5) V3.7.2. The Noah Land Surface Model (LSM) is used to represent the urban environment with a sensitivity simulation removing the urban region and replacing it with the dominant rural land use category. Results show a drastic change in surface energy balance characteristics as well as model derived radar reflectivity patterns when the urban region is removed in the NOURBAN simulation.
The same 13 June 2005 case is also investigated using the Weather Research and Forecasting (WRF) V.2.1.1 model. Identical control and nourban simulations are performed similar to the MM5. Results between the two mesoscale models are difficult to compare due to the domain structure as well as the different physics options and data initialization techniques used within the two models; however, few comparisons can be made. Several land use sensitivity simulations are also performed in an effort to better understand the underlying effects of the urban region on the case of interest. Simulations include a nourban case where the urban region is completely removed and replaced by the dominate surrounding rural land use, similar to the MM5 study. Other simulations performed are used to set up a statistical factor separation experiment. Land use variables that are manipulated include albedo, surface roughness length (z0), and urban sprawl. All variables are manipulated to increase the overall effectiveness of a larger, more 'robust' city (Indianapolis) to resemble the current and projected increase in urbanization. Model comparisons within these simulations were compared using only the nest 3 (4 km) resolution due to the poor representation of the event on the finest 1.33 km grid spacing nest. Surface energy balance parameters were not altered a great deal in the simulations; however, the overall precipitation patterns and storm characteristics show a wide range of variability. When the urban region is removed, the storm of interest does not propagate through the urban region. Accumulated precipitation totals through the time of the event are dramatically less around and downwind (northeast) of the urban region. The factorial experiment suggested that a more robust city acted to place the maximum precipitation amounts on the northern lateral edge of the city that may be caused by the urban 'building barrier effect.'
Observations from the Joint Urban Project 2003 held in Oklahoma City are used to compare to the Coupled Ocean-Atmosphere Mesoscale (COAMPS) model sensitivity simulations of a Mesoscale Convective System (MCS) event that occurred on 30 July 2003. Two distinct simulations were used to compare the control simulation. The OKC urban region was enhanced by using a coupled urban canopy model (UCM) approach to attempt to resolve the finer scale heating and drag features of the urban environment (e.g., roof, wall, street, and anthropogenic heating contributions) to the urban heat island (UHI). Also, the Gas Exchange Photosynthesis Based model (GEM) was coupled to the Noah Land Surface Model (LSM) to enhance the overall heterogeneity of the urban-rural interface. Results indicate that the simulation with urban canopy parameterizations (UCP) enhanced the overall intensity of the event, while the GEM proved to show more heterogeneity by placing the convection away from the downtown area, similar to the observed.|
|Appears in Collections:||Theses|
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.