Replica’s seasonal mobility model datasets in Studio and related applications are based on origin–destination (O-D) pairs derived from upstream data sources that include only trips with both origins and destinations located within the United States. As a result, trips that originate or end outside the U.S. are not directly represented as cross-border O-D pairs in Studio. Trips originating outside the country are therefore not simulated, and trips made by U.S. residents that might involve leaving and re-entering the country are instead represented as domestic trips within the U.S. The underlying road network does include roads outside of the U.S. (e.g., Mexico and Canada), and if the routing engine determines that a U.S.-to-U.S. trip near the border would be faster by briefly traveling outside the country, it may route the trip along those roads. However, the model does not simulate trips that originate outside the U.S., and it does not impose an explicit border-crossing delay or penalty. Because the upstream O-D inputs exclude international endpoints, Studio should not be used to measure or estimate cross-border travel demand between the U.S. and other countries.
The models powering Network Explorer and related applications estimate traffic volumes on roadway segments using upstream network data sources and modeling approaches that do not rely on specific O-D pairs. Because of this, volumes on corridors approaching international border crossings may capture some of the overall traffic activity occurring on those facilities, including traffic associated with cross-border movement. However, these volumes should not be interpreted as reliable estimates of cross-border trip counts. Upstream network data sources often have limitations near border crossings, where long delays and queuing conditions are common. These conditions can cause trips to be segmented into multiple shorter traces for privacy reasons, meaning a single cross-border journey may appear as several disconnected trip fragments in the underlying data. As a result, while modeled corridor volumes may provide a general sense of traffic levels near border facilities, they are not well suited for estimating the number of trips that actually cross the border.