Transit Equity Research

The Hidden Geography of Transit Access

When policymakers discuss food access, they usually mean distance to the nearest grocery store. A community is classified as a "food desert" if residents live more than one mile from a supermarket in urban areas, or ten miles in rural areas. This framework assumes that distance is the primary barrier to food access—and that shorter distances mean better access.

But distance is only meaningful if you can traverse it. For the 11% of California households without a vehicle, the relevant question isn't "How far is the nearest store?" but "How long will it take to get there?" The answer depends on transit service: frequency, coverage, and reliability. And transit service varies dramatically across California's diverse geography.

Our research reveals a phenomenon we call "hidden mobility deserts"—census tracts where grocery stores are geographically close (under half a mile) but require over 30 minutes to reach by public transit. These communities don't appear in conventional food desert maps because their stores are nearby. But for transit-dependent residents, that nearby store might as well be across town.

Measuring Transit Access at Scale

Calculating transit travel times requires detailed routing data. We use GTFS (General Transit Feed Specification) feeds from California's 72 transit agencies, representing every bus route, rail line, and ferry service in the state. Combined with OpenStreetMap walking networks, this allows us to compute multimodal travel times: walk to the bus stop, wait for the bus, ride, transfer if needed, and walk to the final destination.

Using the r5py library (based on Conveyal's R5 routing engine), we calculated 2.7 million origin-destination pairs across California. Each calculation represents the travel time from a census tract centroid to a grocery store, accounting for realistic departure times and transfer penalties. The result is a comprehensive map of transit accessibility that reveals patterns invisible to distance-based metrics.

When Proximity Doesn't Equal Access

Our analysis identified 13 census tracts across California where the nearest grocery store is less than 0.5 miles away—well under the "food desert" threshold—but requires more than 30 minutes to reach by public transit. These hidden mobility deserts share common characteristics: they're often located at the edges of transit service areas, where bus routes run infrequently or follow circuitous paths that don't connect residential areas to commercial centers.

Consider a hypothetical example: a resident lives 0.3 miles from a grocery store but across a highway overpass. By car, the trip takes two minutes. On foot, it requires a mile-long detour to the nearest safe crossing. By transit, it means walking to a bus stop, waiting (perhaps 30 minutes for an infrequent route), riding several stops, and walking the rest of the way. The store is "nearby" by every distance metric, but effectively inaccessible for transit-dependent households.

Race, Income, and Transit Disparities

Transit access isn't distributed equally across California's communities. Our research examines how accessibility varies by race and income, revealing patterns that reflect decades of land use decisions, infrastructure investments, and residential segregation.

Low-income communities and communities of color often face a paradox: they're more likely to depend on public transit (due to lower vehicle ownership rates) but may receive lower-quality service (due to historical underinvestment in transit infrastructure). Understanding these disparities requires measuring actual accessibility—not just whether transit exists, but whether it connects people to the destinations they need.

Implications for Policy

The distinction between geographic proximity and transit accessibility has direct policy implications. Programs designed to improve food access often focus on bringing stores to underserved areas—building new supermarkets, supporting corner stores, subsidizing farmers markets. These interventions assume that distance is the primary barrier.

But if the barrier is transit access rather than distance, the solution looks different. It might mean improving bus frequency on routes connecting residential areas to grocery stores. It might mean coordinating land use and transit planning so that new development locates near transit corridors. It might mean investing in pedestrian infrastructure—sidewalks, crosswalks, and safe street crossings—that makes the "last mile" actually walkable.

None of these interventions appear in conventional food access policy. But they might matter more for transit-dependent households than adding another store to an area already saturated with unconnected retail.