Local jurisdictions are putting forward solutions to the global problem of climate change through Climate Action Plans. These documents strive to first quantify the greenhouse gas emissions within a city's scope of influence and then put forward strategies to reduce future emissions. The Bay Area Quality Management District (Air District) has provided formal guidance to assist jurisdictions in developing these plans. A key component of local jurisdictions' greenhouse gas footprint is passenger vehicle travel. There is, however, no obvious approach to allocating the responsibility of the GHG from vehicle travel to individual jurisdictions. Several entities have put forward suggestions for how to approach this allocation, some of which the Air District notes in the guidance referenced above. The purpose of this data portal is to quickly and easily deliver consistent, jurisdiction-specific estimates of vehicle miles traveled. Importantly, this data is consistent with the travel outcomes associated with the Metropolitan Transportation Commission's long-range transportation plan, which meets a state-mandated GHG reduction target.

The Metropolitan Transportation Commission, or MTC, is the transportation planning, financing and coordinating agency for the nine-county San Francisco Bay Area. As part of their work MTC deploys and maintains a large-scale simulation model of daily travel behavior to support its regional transportation planning efforts, most notably in the regional transit document Plan Bay Area.

The model simulates transportation activities that the average household currently living in (for simulations of historical years), or expected to be living in (for simulations of future years), in the nine-county Bay Area. The model is documented in great detail on this webpage and the implementation scripts are shared and maintained on GitHub. The Data Portal summaries are drawn from the same model output that informs MTC's regional transportation plan and sustainable communities strategy. Therefore, the data is consistent with the travel data used in preparing regional plans to address transportation and land use, air quality, and greenhouse gases. The simulation involves the prediction of numerous travel-related outcomes, including:

• where the household is located;
• the location of employment of workers in the household and where the students go to school;
• the time during which workers work, students go to school;
• the time and location in which individuals engage in other, non-work and non-school, activities;
• the means by which travelers get around (e.g., walk, drive, take the bus);
• and, the routes travelers take to get around (e.g., I-80 to Mission Boulevard, the #1 bus).

Two important notes are as follows:

1. The model attempts to simulate a "typical" weekday, defined as a weekday when school is in session, the weather is good, and no major accidents or construction activities are taking place on the roadway.
2. The travel model operates an abstracted spatial system: rather than moving between two precise locations, travelers move between what we call "travel analysis zones". In MTC's current travel model, the travel analysis zones are similar in size to Census tracts. We assume that a traveler going to a travel analysis zone could be going to any precise location within the travel analysis zone's boundaries.

One of the outputs of the simulation is a detailed record of every trip made by every simulated traveler. We define a "trip" as a movement between two locations where travelers engage in activities, e.g. from home to work, from school to the grocery store. The trip records include the trip's means of travel and travel distance as well as the traveler's home and work locations. This output can be used to create the data presented in the Climate Action Plan Vehicle Miles Traveled Data Portal. This is done via the following steps:

• define the subject geography, i.e. the city or county;
• identify the simulated travelers that live in the subject geography;
• identify the simulated travelers that work in the subject geography;
• compute the VMT for each simulated trip, which is simply the distance traveled in a distance divided by the number of travelers in the vehicle (if three travelers are in the vehicle, the movement is listed three times in the trip file);
• identify trips that begin in the subject geography;
• identify trips that end in the subject geography;
• and, sum the vehicle miles traveled by the six live/work market segments and the three start/end travel segments shown in the Data Portal summaries.

These data summaries include only non-commercial travel. Meaning, it does not include travel made by taxi drivers, plumbers, etc., or for moving goods/freight.

Three key caveats are as follows:

1. Our simulation model explicitly assumes that every worker living in the nine-county Bay Area also works in the nine-county Bay Area. This is, of course, not always true, particularly in areas at the boundary of our planning region (e.g., Dixon). This caveat should be kept in mind when examining the results for jurisdictions near the boundary of our planning area.
2. The data is simulated: we are using models to predict where people live and how they get around. The models are based on observations collected at different points in time and are routinely reviewed by State and Federal entities. We believe the data is robust and useful for informing Climate Action Plans — but it is simulated, not observed.
3. The land use allocations used by MTC likely differ from those used by the county Congestion Management Agencies. Specifically, when doing sub-regional planning, MTC — in the Congestion Management Plan context — allows counties to re-allocate land use within their counties to different locations. There may, therefore, be minor inconsistencies between regional outcomes of the types presented in the Data Portal estimated by MTC and county outcomes estimated by Congestion Management Agencies. The Congestion Management Agencies may also use different travel modeling tools.