The set of users, sources, or contributors that can reach a node, jointly defined by the node, a friction-weighted medium, and a tolerance horizon beyond which use is dropped. Coverage is not a primitive property of the node but a quantity computed from the three, so changing any one changes the set.
Think about an ice cream truck parked on a corner. The kids who can get to it before their ice cream melts are its 'catchment' — everyone close enough to come buy. If the truck moves, a different bunch of kids can reach it. So 'who can come' depends on where the truck is and how far kids are willing to walk.
The Reachable Crowd
An Access Catchment is the whole group of people who can reach a place, given how hard it is to get there and how much effort they're willing to spend. Picture a playground: its catchment is every kid who lives close enough to walk there before they get tired of walking. Three things decide it together — the spot itself, the paths to it (are they easy or full of busy roads?), and how far people will bother to go. Change any one and the group changes: move the playground, build a safe crossing, or kids willing to walk farther, and suddenly more or fewer can reach it. This turns a single dot on a map ('where it is') into a crowd ('who can use it'), and that crowd is how you judge whether the spot covers enough people.
The Coverage Set
An Access Catchment is the set of users who can reach a node, given a friction-weighted network connecting candidates to it and a tolerance horizon — time, distance, cost, effort — beyond which they give up. The catchment is defined jointly by the node, the medium, and the tolerance; change any one and the catchment changes. It converts a point (where the resource is) into a set (who can use it), and that set becomes the coverage metric the design is judged by. Structurally it's the demand-side dual to operational reach: the same friction-plus-tolerance machinery that describes how far a supplier can project outward describes how broad a group can be drawn in. That duality gives one shared toolkit for improving things: move the node, densify the network to cut friction, raise or lower the tolerance horizon, improve crossings, or add more nodes. The key insight is that coverage isn't a built-in property of a resource — it's computed from medium, friction, and tolerance, so you can improve it by changing any of the three.
An access catchment is the set of users, sources, or contributors that can reach a node, given a friction-weighted network or medium connecting candidate users to the node and a tolerance horizon — time, distance, cost, latency, effort — beyond which use is dropped. The catchment is jointly defined by the node, the medium, and the tolerance: change any one and the catchment changes. The construct converts a point (where the resource is) into a set (who can use it), and that set becomes the coverage metric by which the design is judged. The structural commitment is that the catchment is the demand-side dual to operational reach: the same friction-field-plus-tolerance machinery that describes how far a supplier can project — the supply-side projection from a point of action — describes how broad a group can be drawn in around a point of attraction. The dual framing licenses a single intervention vocabulary across both polarities: move the node, densify the network to reduce friction, raise or lower the tolerance horizon, change the friction field through better crossings or higher speeds, or add nodes for multi-coverage. Naming the catchment as a derived quantity is what makes coverage analyzable: coverage is not a primitive property of a resource but is computed from the medium, the friction, and the tolerance, so it can be improved by intervening on any of the three. A point becomes a set, the set becomes a metric, and the metric becomes the object the planner optimizes.
Urban planning: the walkshed — the population within a 5- or 10-minute walk of a transit stop on the actual street network — is the routine instrument of transit-equity analysis.
Retail: a store's trade area is the catchment defined by travel-cost-weighted attractiveness, and chains site stores to maximise non-overlapping catchment.
Public health and education: hospital service areas, ambulance-response polygons, and school-attendance zones are defined over the network, not Euclidean radius.
Communications and logistics: broadcast range, cellular coverage, and last-mile delivery zones are catchments in propagation- or transport-friction media.
Hydrology and ecology: the watershed is the dual case — the upstream set flowing into a node — and home ranges are catchments around a nest or flower.
Computing: CDN edge-region assignment maps users to the nearest cache under a latency tolerance over the routing fabric.
Separates Euclidean radius, network distance, friction-weighted cost, and the catchment, and names the tolerance horizon as a chosen design parameter rather than a fact of nature.
Collapses coverage gaps, equity comparisons, and access disparities into one diagnostic: given node, friction, and horizon, what is the catchment and how does it intersect the intended population?
Supports coverage-gap and catchment-overlap reasoning, counterfactual node-placement, and tolerance-horizon sensitivity — turning "who can get there" into computable questions about a derived quantity.
Urban transit to CDN architecture: the walkshed apparatus transfers to edge placement by re-keying the friction field from street network to routing latency.
Hydrology back to computing: the watershed — the upstream contributing set — transfers to reverse-traceability, e.g. which content sources feed a recommender cluster, proving nothing depends on friction being physical.
A transit stop's walkshed — the population within a 5-minute walk on the actual street network — can have its coverage gap closed by adding a pedestrian crossing, not building a new stop, because coverage is derived from friction.
Access Catchment is not System Slack because system slack is spare reserve buffering against shocks whereas access catchment is the reachable set of users around a node.
Access Catchment is not Boundary because a boundary is a drawn or defended line whereas the catchment's edge is derived and moves automatically when node, friction, or horizon changes.
Access Catchment is not Network Effect because a network effect is value rising with adoption whereas the catchment is the who-can-reach set, indifferent to value-with-scale.