PhD Defense: Barry Baker
Location
Physics : 401
Date & Time
October 17, 2014, 9:00 am – 11:00 am
Description
TITLE: Physical Processes in the Atmospheric Boundary Layer with Implications for Air Quality
ABSTRACT: Ozone (O3) is a secondary pollutant dependent on complex photochemical reactions of nitrogen oxides (NOx) and volatile organic compounds (VOCs) and is sensitive to meteorological conditions that govern solar radiation, temperature, and wind speed/direction (Stockwell, 2011). The U.S. Environmental Protection Agency (USEPA) has classified O3 as one of the six criteria pollutants that is considered harmful to both plants and human health (Hollingsworth, 2007). Recently, the USEPA suggested that the standard for O3 be tightened from 75 ppbv to 60-70 ppbv. The ability to model the local concentrations of pollutions are critical to make informed decisions for policy and a major difficulty is the modeling of atmosphere is turbulence within the boundary layer. The atmospheric boundary layer (BL) is the lower layer of the atmosphere that is influenced by the surface with a time scale of an hour or less. Dispersion and transport of pollutants, momentum and heat from the surface are controlled through the vertical turbulence found in the BL. Turbulent intensity in the BL is determined by the diurnal evolution of the sun, surface water, surface temperature t and heat flux exchange into the atmosphere. The BL can be described with a length scale called the BL height, which the turbulent intensity is dependent upon. During the daytime, warm surface temperatures create a large amount of turbulence and a BL height of ~1-2km is typical. When the sun begins to set the surface cools and creates a more stable condition with cool, dense air near the surface and warm air above. This causes the length scale of turbulence to decrease and therefore the BL height decreases to 100-500m above the ground during nocturnal hours. The BL height plays a crucial role in determining the vertical profile of the lower atmosphere in both Mesoscale weather prediction and air quality models. The current study investigates the influence of the diurnal cycle of the BL on the surface air pollutants. Specifically, it examines how the nighttime and transition period turbulence impacts the concentration of ozone (O3) at the surface and in the atmospheric column during the following day. In order to complete the study, a series of models including the Weather Research and Forecasting (WRF) model and the Comprehensive Air Quality model with Extensions (CAMx) are used to simulate atmospheric conditions in the Maryland and Texas areas. First, a six-month simulation using the Blackadar BL scheme is completed to quantify model errors and to understand the impact emissions from surrounding regions has on Maryland air quality. Several sensitivity tests involving BL processes such as the minimum diffusivity or BL parameterization within CAMx are completed. The Blackadar scheme was found to artificially suppress the nocturnal BL height and is thought to be major cause of the large model bias (15-20 ppbv) compared to observations found over the six months. An experiment is then conducted where a minimum BL height of 160 m and an improvement of 5 ppbv or 7% in the median model bias are found in the Maryland area. With Maryland being in a NOx-limited regime, the same process of setting a minimum height is tested in the Texas area using the YSU scheme but minimal differences produced miniscule changes in the vertical diffusivity. The Maryland case is reran using the YSU scheme, which more accurately represents the diurnal cycle of the atmospheric BL, by using a static critical bulk Richardson number (Ri). The use of the Ri number improved the model bias by another 5 ppbv or 13% over using the Blackadar BL scheme. A new algorithm to predict the nocturnal BL depth is implemented for the first time in a mesoscale numerical weather prediction model, which scales the critical bulk Ri with the Obukhov length adding a dependency on near surface properties of the flow. The algorithm is tested using meteorological surface measurements and tower measurements, along with pollutant measurements in the Texas and Maryland areas. In the Maryland area, the median model bias improved by 1 ppbv or 10% compared to using a static critical bulk Ri number and 15% compared to the Blackadar BL scheme.
ABSTRACT: Ozone (O3) is a secondary pollutant dependent on complex photochemical reactions of nitrogen oxides (NOx) and volatile organic compounds (VOCs) and is sensitive to meteorological conditions that govern solar radiation, temperature, and wind speed/direction (Stockwell, 2011). The U.S. Environmental Protection Agency (USEPA) has classified O3 as one of the six criteria pollutants that is considered harmful to both plants and human health (Hollingsworth, 2007). Recently, the USEPA suggested that the standard for O3 be tightened from 75 ppbv to 60-70 ppbv. The ability to model the local concentrations of pollutions are critical to make informed decisions for policy and a major difficulty is the modeling of atmosphere is turbulence within the boundary layer. The atmospheric boundary layer (BL) is the lower layer of the atmosphere that is influenced by the surface with a time scale of an hour or less. Dispersion and transport of pollutants, momentum and heat from the surface are controlled through the vertical turbulence found in the BL. Turbulent intensity in the BL is determined by the diurnal evolution of the sun, surface water, surface temperature t and heat flux exchange into the atmosphere. The BL can be described with a length scale called the BL height, which the turbulent intensity is dependent upon. During the daytime, warm surface temperatures create a large amount of turbulence and a BL height of ~1-2km is typical. When the sun begins to set the surface cools and creates a more stable condition with cool, dense air near the surface and warm air above. This causes the length scale of turbulence to decrease and therefore the BL height decreases to 100-500m above the ground during nocturnal hours. The BL height plays a crucial role in determining the vertical profile of the lower atmosphere in both Mesoscale weather prediction and air quality models. The current study investigates the influence of the diurnal cycle of the BL on the surface air pollutants. Specifically, it examines how the nighttime and transition period turbulence impacts the concentration of ozone (O3) at the surface and in the atmospheric column during the following day. In order to complete the study, a series of models including the Weather Research and Forecasting (WRF) model and the Comprehensive Air Quality model with Extensions (CAMx) are used to simulate atmospheric conditions in the Maryland and Texas areas. First, a six-month simulation using the Blackadar BL scheme is completed to quantify model errors and to understand the impact emissions from surrounding regions has on Maryland air quality. Several sensitivity tests involving BL processes such as the minimum diffusivity or BL parameterization within CAMx are completed. The Blackadar scheme was found to artificially suppress the nocturnal BL height and is thought to be major cause of the large model bias (15-20 ppbv) compared to observations found over the six months. An experiment is then conducted where a minimum BL height of 160 m and an improvement of 5 ppbv or 7% in the median model bias are found in the Maryland area. With Maryland being in a NOx-limited regime, the same process of setting a minimum height is tested in the Texas area using the YSU scheme but minimal differences produced miniscule changes in the vertical diffusivity. The Maryland case is reran using the YSU scheme, which more accurately represents the diurnal cycle of the atmospheric BL, by using a static critical bulk Richardson number (Ri). The use of the Ri number improved the model bias by another 5 ppbv or 13% over using the Blackadar BL scheme. A new algorithm to predict the nocturnal BL depth is implemented for the first time in a mesoscale numerical weather prediction model, which scales the critical bulk Ri with the Obukhov length adding a dependency on near surface properties of the flow. The algorithm is tested using meteorological surface measurements and tower measurements, along with pollutant measurements in the Texas and Maryland areas. In the Maryland area, the median model bias improved by 1 ppbv or 10% compared to using a static critical bulk Ri number and 15% compared to the Blackadar BL scheme.
