Regional Assessment
AERMOD, RLINE, and RLINEXT Case Analyses in Atlanta, Georiga
This research assessed the impact of USEPA’s AERMOD dispersion model (version of 19191) source types on predicted pollutant concentrations via a case study for the I-75/I-575 Northwest Corridor (NWC) in Atlanta, GA. Using MOVES-Matrix for MOVES 2014b, carbon monoxide (CO) emissions rates for every hour of a one-year study period were generated using traffic volumes and speed from Atlanta Regional Commission’s Activity-Based Model (ABM) 2020 and AERMET meteorological profiles provided by the state environmental agency (EPD). To develop concentration profiles to assess prediction differences across source types, consistent input datasets for hourly emissions by ABM link, hourly AERMET data, and gridded receptor placement (standard 20-meter grids and variable grids after link-screening) were run with the AERMOD source types: AREAPOLY (manually created and automatically generated), LINE, VOLUME, RLINE, and RLINEXT (with and without noise barriers). The team processed more than one trillion source-receptor pairs the study area using the cyberinfrastructure resources provided by the Partnership for an Advanced Computing Environment (PACE) at Georgia Tech. The results indicate that predictions from AREAPOLY and LINE are essentially identical. Predictions from RLINE and RLINEXT are almost the same, but these predictions are higher in most cases than any other source type. The VOLUME source type always yields the lowest concentrations and is less sensitive to wind directions and speed, due to the embedded wind meander dispersion parameters implemented only for VOLUME sources. Machine learning results indicate that wind speed, receptor ID (which accounts for adjacent roads and their and their mass flux emission rates in grams/meter2/second), and wind direction influence the results much more than source type selection. Introducing noise barriers to RLINEXT lowered concentration as expected, but modeling barrier effects was challenging due to the restrictive assumptions in AERMOD. Sensitivity analysis for RLINEXT suggests that barrier height, distance to the roadway, wind speed, and wind direction all affect morel predictions. Modelers need to exercise care in matching barriers to roadway link segments (i.e., barrier edge effects were observed).
Download the final report by Georgia Tech.
Access the animated CO concentration at the I-75/I-575 Northwest Corridor.
Modeling Framework of Population Exposure to Traffic-related PM2.5 and Environmental Equity Analysis: Case Study in Atlanta, Georgia (Ongoing)
The exposure to fine particular matter of 2.5 μm or smaller (PM2.5) has been widely connected with adverse impacts on human health, but exposure estimation is often limited by a lack of high-resolution modeling framework, which is of crucial importance for quantifying inhaled particulate mass and in conducting environmental equity analysis. The existing models of population exposure are mostly based on coarse input data (area-wide concentration, for instance), and are not sensitive to reflect the change of travel paths. Although there are a few exposure models that provide detailed outputs, these models have not been comprehensively integrated to a complete modeling framework, quantifying both off- and in-vehicle inhalation.
In this work, a modeling framework of population exposure to PM2.5 with high spatiotemporal resolution is proposed and applied to the region of Atlanta, GA for environmental equity analysis. The research is conducted with the following steps: First, the output database of Atlanta Regional Commission’s (ARC) Activity-Based Model 2015 (ABM15) is retrieved, and a refined path retention algorithm is proposed based on previous study (Zhao, et al., 2019) to generate individual travel paths and population temporospatial distribution. The high-resolution concentration profiles by AERMOD is also retrieved from the previous study (Kim, et al., 2019). Second, the travel activity data is integrated with MOVES-Matrix (Guensler et al., 2017) to obtain the individual PM2.5 emission for comparison with the inhalation. Third, the exposure modeling will be conducted based upon the quantification of both off-vehicle and in-vehicle inhaled mass, derived from detailed spatial and temporal attributions of concentration and travel activity input. Analysis will also include an assessment of temporospatial uncertainty analysis. The outputs will be aggregated into household/demographic group levels and environmental equity will be evaluated across the demographic groups for exposure to traffic-related PM2.5. The products of this research provide exposure/emission analysis at an individual level, with an appropriate number of input data. The proposed modeling framework can be used to link concentration profiles and population exposure for analysis such as environmental equity issues, and the output dataset also benefits the decision-making in the Atlanta region.
Corridor Analysis
Investigation of MOVES Project-Level Uncertainty: Impact of Temporal- and Spatial- Aggregation of Onroad Operating Conditions on Emission Rate Estimates
On-road traffic operations data are an important input to modeling vehicle emissions and energy consumption. A small change in vehicle activity inputs can yield disproportionately large impact on model outputs, due to the nonlinear relationship between traffic conditions, fuel use, and emission rates. Previous studies have shown that temporal aggregation of traffic inputs can result in uncertainties in emissions estimates as large as 13% for PM2.5. However, the impact of spatial aggregation and fleet composition is not fully understood.
In this research, on-road operating speeds, traffic volumes, and fleet compositions are temporally and spatially aggregated to various levels to assess the resulting uncertainty in emissions estimates from a selected roadway section. In this study, three temporal averaging periods (5-minute, 15-minute, and 1-hour) and three spatial levels (lane-by-lane, two inside lanes vs. three outside lanes, and across all lanes) were evaluated resulting in nine independent scenarios The most disaggregated scenario (5-minute and lane-by-lane) was considered to be the “baseline” case, and the other eight were considered the test scenarios. Energy and emission rates are pulled from MOVES-Matrix (a matrix of 90 billion MOVES energy and emission rates) based upon the inputs of each test scenario. Scenario uncertainties are assessed using Monte Carlo and Bootstrap methods. The results of the case study indicate that emissions estimates can vary by as much as +7.7% to -20.4% for CO, and +10.3% to -18.6% for PM2.5 at the studied freeway segment. Significant variation in emissions estimates (overestimation under some circumstances and underestimation under others) were also observed by time-of-day, and by pollutant type. The emissions estimate variability (uncertainties) were found to be associated with aggregation of fleet composition, especially spatially.
Access Lu, H., Liu, H., Xia, T., Angshuman Guin, Rodgers, M.O., Randall Guensler, 2021. Investigation of MOVES Project-Level Uncertainty: Impact of Temporal- and Spatial- Aggregation of Onroad Operating Conditions on Emission Rate Estimates. 100th Annual Meeting of the Transportation Research Board. Washington, D.C., U.S.
Individual Vehicle Modeling
Combined Effect of Changes in Transit Service and Changes in Occupancy on Per-Passenger Energy Consumption
Many transit providers changed their schedules and route configurations during the COVID-19 pandemic, providing more frequent bus service on major routes and curtailing other routes, to reduce the risk of COVID-19 exposure. This research first assessed the changes in Metropolitan Atlanta Rapid Transit Authority (MARTA) service configurations by reviewing the pre-pandemic versus during-pandemic General Transit Feed Specification (GTFS) files. Energy use per route for a typical week was calculated for pre-pandemic, during-closure, and post-closure periods by integrating GTFS data with MOVES-Matrix transit energy and emission rates (MOVES signifying MOtor Vehicle Emission Simulator). MARTA automated passenger counter data were appended to the routes, and energy use per passenger-mile was compared across routes for the three periods. The results showed that the coupled effect of transit frequency shift and ridership decrease from 2019 to 2020 increased route-level energy use for over 87% of the routes and per-passenger-mile energy use for over 98% of the routes. In 2021, although MARTA service had largely returned to pre-pandemic conditions, ridership remained in an early stage of recovery. Total energy use decreased to about pre-pandemic levels, but per-passenger energy use remained higher for more than 91% of routes. The results confirm that while total energy use is more closely associated with trip schedules and routes, per-passenger energy use depends on both trip service and ridership. The results also indicate a need for data-based transit planning, to help avoid inefficiency associated with over-provision of service or inadequate social distancing protection caused by under-provision of service.
