Accurate PBL Simulation Helps Model Urban Pollution

Atmospheric transport models benefit from real-time localized data inputs.

Deteriorating air quality in urban areas is causing growing concern about public health and climate change. To gain a better understanding of air pollution and greenhouse gas (GHG) emissions, some cities and states are attempting to establish a baseline and collect data over time to track changing conditions. By comparing the performance of models to real-life data, researchers hope to isolate problem areas and produce more accurate air quality forecasts.

To test the performance of atmospheric transport models and improve estimates of urban air pollution and GHG emissions, researchers from Harvard University, Boston University and Atmospheric and Environmental Research, Inc., joined together to conduct a study in Boston, Mass. They evaluated the daytime planetary boundary layer (PBL) and the nighttime residual layer (RL) over a 15-month period using a MiniMPL sensor from the Micro Pulse LiDAR division of Hexagon. The unit was installed on the rooftop of a Boston University building to collect data.
PBL height is a key input in air quality models because the PBL defines the mixing height of air pollutants in the lowest layer of the atmosphere. The PBL responds to surface activity within an hour or less, making it very dynamic and difficult to measure. The top of the RL separates the PBL from the free atmosphere. Pollution in the RL is influenced by windborne pollutants from other geographic areas.

Simulations of the PBL height from numerical weather prediction (NWP) models are often too high or too low in comparison to data from soundings or other traditional measurement methods. If the simulated PBL is lower than the observed PBL in atmospheric transport models, simulated particles in the air are closer to the surface, biasing air pollution levels too high (and vice versa). From an air quality perspective, these biases lead to inaccurate assessments of pollution levels and effects on public health.

“Historically, PBL soundings have been taken at specific rural locations only twice a day, whereas the high temporal resolution of the MiniMPL provides a complete 24-hour record and reflects conditions in situ,” said Yanina Barrera, Harvard University. “We have developed the first long-term and continuous record of PBL and RL heights in Boston using a MiniMPL sensor.”

Micro Pulse LiDAR Supports Evaluation of Models
The MiniMPL sensor was selected for the PBL study based on its ability to measure the backscatter of light from aerosols in the atmosphere, allowing researchers to image the PBL and RL. Also, the MiniMPL unit is compact and requires little to no onsite maintenance. It can be installed on a rooftop or on the ground and be programmed to obtain continuous measurements with little oversight.

The PBL heights calculated using the MiniMPL’s normalized relative backscatter (NRB) profiles were compared to simulated heights in seven configurations of the NWP models. The comparisons revealed discrepancies in simulated PBL height from NWP models ranging from −2.5 to 4.0 km caused by a variety of systemic errors. Results from the study show that mean percent error can exceed 350% in NWP models during the morning hours of the winter season, highlighting why atmospheric transport modeling studies are often performed during the afternoon hours to reduce errors.

The RL analysis revealed emissions traveling to Boston at night originating from areas of Pennsylvania, New York, and Connecticut that contained emissions from coal mining, power plants, heavy traffic, wintertime biomass burning, and other industrial sources. As cities and states try to minimize their contribution to greenhouse gases, pinpointing emission sources and quantifying the amount of pollution being transported is key.

“Our work shows how a simple, robust MiniMPL system can help estimate GHG emissions and establish accurate baselines at urban and subnational scales,” says Barrera.

For more information about Dr. Barrera’s research, refer to Barrera, Y.; Nehrkorn, T.; Hegarty, J.; Sargent, M.; Benmergui, J.; Gottlieb, E.; Wofsy, S. C.; DeCola P.; Hutyra, L. R.; and Jones, T. Using lidar technology to asses urban air pollution and improve estimates of greenhouse gas emissions in Boston. Environmental Science and Technology. 2019, DOI:10.1021/acs.est.9b00650.