We incorporated traffic data from the Highway Performance Monitoring System (HPMS), which provides annual average daily traffic (AADT) by vehicle class for 14.3 million road segments representing highways, collector roads, and arterial roads in the contiguous U.S. At the time of this study’s completion, only HPMS data through 2020 were publicly available. Because traffic in 2020 was affected by the COVID-19 pandemic, we paired the 2019 HPMS dataset with the other data inputs from 2021, acknowledging that traffic volume and patterns may have changed between 2019 and 2021. We formed two different measures of AADT: total AADT and truck AADT (i.e., AADT from vehicle classes 4–13, which include single-unit trucks, buses, and semi- and multi-trailers).
HPMS synthesizes data from individual U.S. states’ departments of transportation, leading to variations in the spatial completeness of the data. To address this limitation, each state’s AADT was scaled using the ratio of total activity reported by HPMS to the total activity reported by the nationally consistent Federal Highway Administration database55. We then transformed total and truck AADT on individual road segments to vehicle kilometers traveled (VKT = AADT * road length) due to the variable lengths of road segments and thereafter summed total and truck VKT over all road segments contained within 0.01° x 0.01° grid cells over the contiguous U.S. (Figure S14). For every warehouse, we sampled the 0.01° x 0.01° traffic grid at the grid cell nearest to each warehouse and ± 7 grid cells in each direction and summed total and truck VKT over this area to represent near warehouse traffic, similar to our treatment of TROPOMI data.
Do you have any questions about this protocol?
Post your question to gather feedback from the community. We will also invite the authors of this article to respond.