Isobenefit Urbanism

Scenario library

Case studies with prepared data and precomputed runs

Pick a scenario, then move between the precomputed runs to see how each parameter changes the outcome. Every panel is a complete simulation (growth, then post-processing) at a coarsened preview resolution, with a fixed seed and downloadable parameters; formal runs happen in QGIS at the scenario's own resolution using the downloads below. Built land is coloured by density tier (yellow to brown), mixed-use centres in reds, existing fabric muted, nature green, water and barrier corridors blue, and land excluded for slope tan.

These runs are hypothetical scenarios built on automatically extracted OpenStreetMap layers. OSM coverage varies from place to place, so a mapped gap may be developed land, or a centre may go unrecorded. The layers download as ordinary editable files: updating them to match what is on the ground (developed land, planned sites, protected areas, known centres) improves the simulation accordingly.

Scenario

Precomputed run

Download this scenario (ZIP) Download this run's parameters How to use these in QGIS

Served (walk to both)	Avg walk to centre	Avg walk to green	New population	Centre m²/person	Green m²/person

Reproduce this run

1. Download the scenario data (ZIP): the extents, prepared OSM layers and parameter presets (also browsable on GitHub).
2. Download this run's parameters (a plugin params.json).
3. In QGIS, add the layers from the ZIP, open the run dialog, press Load parameters and pick the downloaded file, then choose the layers in the Input layers group (extents, built, green, unbuildable, centres, stops, stations).
4. Choose an output folder and run name, then press Run. The preview above used a coarsened grid; the loaded parameters run at the scenario's own resolution, so expect finer and slower results of the same kind.

Exact settings used by this run

The scenarios in detail

Each scenario is a real place with a worked parameter set: local density norms translated into the plugin’s controls, prepared input layers, and a population target to grow toward. Anyone can rerun a scenario with the same data and settings.

Every scenario lives in a folder under scenarios/ in the repository, containing:

• extents*.geojson, the formal simulation boundary (or boundaries), in the scenario’s metric CRS;
• params.json, the full parameter set in the plugin’s format. The run dialog’s Load parameters button reads this file directly, and every plugin run writes the same format back as a *_params.json sidecar next to its output, so any past run can be reloaded too;
• the OSM input layers (built, green, centres, unbuildable, streets, stops, stations, railways, industrial), pre-fetched with the plugin’s own extraction rules by scripts/fetch_scenario.py. The download window is the convex hull of the extents features, so one fetch covers every pilot area; the hull is kept as osm_download_extent.geojson;
• steep.geojson, terrain slope bands (15° / 20° / 25° / 30°) from the Copernicus GLO-30 elevation model. This ships as a separate, editable layer so local knowledge can trim or extend it; the scenario’s slope_max_deg parameter specifies which bands preclude development. Where it applies, review the layer, then merge the selected bands into the unbuildable layer for a run.

Every scenario downloads as a single ZIP (extents, all input layers including the editable steep.geojson, and the parameter presets), or browse the folders on GitHub:

#	Scenario	Theme	Download
1	Cambourne, UK	New-settlement growth (the reference demo)	ZIP
2	Dnipro, Ukraine	Regeneration and edge growth	ZIP
3	Crews Hill, London	Green-belt release at the metropolitan edge	ZIP
4	Celina, Texas	US suburbia at the metropolitan fringe	ZIP
5	Kigali, Rwanda	Plan-guided rapid urbanisation	ZIP
6	Medellín, Colombia	Planned hillside expansion on steep terrain	ZIP
7	Freiburg, Germany	Validation against built walkable districts	ZIP

1. Cambourne, UK: the reference demo

Cambourne is a fast-growing Cambridgeshire new settlement, and the worked example used throughout the overview page. The scenario folder is the demo project (scenarios/cambourne/, with cambourne.qgz): a 30,000-person target across the demo extents, 50 m cells, EPSG:27700, walks of 800 m to a centre and 400 m to green, and tiers of 6,000 / 3,000 / 1,500 people/km² at shares 0.2 / 0.3 / 0.5. The overview page’s own demonstrators use a smaller 4.2 km window with a 12,000-person target.

As the reference demo, Cambourne also illustrates the input layers every scenario shares:

Layer	Plugin role	Geometry
extents	The simulation boundary; one or more polygons	Polygons
built	Existing built fabric, frozen as context	Polygons
green	Green space to preserve	Polygons
unbuildable	Water, floodplain, slopes and other exclusions	Polygons
centres	Existing or planned urban centres	Points or polygons
streets	Map context; motorway/rail corridors become barriers	Lines
stops, stations	Public transport; stations anchor centres	Points
railways, industrial	Carved as barriers / unbuildable	Lines / polygons

2. Dnipro, Ukraine: regeneration and edge growth

One window covers two growth areas on either side of the river (scenarios/dnipro/): the central right bank (regeneration and infill) and the left-bank edge, where the Samara floodplain is a hard unbuildable limit. The extents layer holds both boundaries as drafts to adjust in QGIS, and a single run grows both areas together. Density tiers and the population target follow the national residential norms and load from params.json; urban centres are supplied as a plain point layer. 25 m grid, EPSG:32636.

3. Crews Hill, London: a green-belt release

The Crews Hill area of Enfield, at London’s northern edge inside the M25, is one of the largest green-belt releases proposed in an emerging London local plan, at about 5,500 homes around an existing rail station. The scenario examines whether a release can deliver a walkable settlement rather than car-led sprawl, and which green network survives. Folder: scenarios/london_crews_hill/.

• Tiers (indicative UK design-code values, household 2.36): 9,500 / 15,500 / 26,000 people/km² (40 / 65 / 110 dwellings/ha), shares 0.3 / 0.5 / 0.2 (low / medium / high).
• Target: about 13,000 new residents (5,500 homes). Walks: 800 m to a centre, 400 m to green. Grid: 25 m, EPSG:27700. Dispersal: off (a contiguous urban extension).
• The boundary and numbers are indicative, to be confirmed against the local plan.

4. Celina, Texas: US suburbia at the fringe

Celina, on the Dallas–Fort Worth northern fringe, has repeatedly been the fastest-growing city in the United States, converting ranchland into master-planned subdivisions at speed. The scenario examines what the walkable-access rules change where the low density tier dominates and the street grid is coarse. Folder: scenarios/celina_tx/.

• Tiers (indicative suburban values, household 2.5): 1,500 / 7,500 / 18,500 people/km² (2.5 / 12 / 30 dwellings per acre), shares 0.6 / 0.3 / 0.1 (low / medium / high).
• Target: about 50,000 new residents, of the order of the city’s own growth projections. Walks: 800 m to a centre, 400 m to green. Grid: 30 m, EPSG:32614. Dispersal: moderate (leapfrog growth is characteristic).
• The boundary and numbers are indicative, to be confirmed against city projections.

5. Kigali, Rwanda: plan-guided rapid urbanisation

Kigali manages rapid urbanisation through a city-wide master plan that steers growth into designated expansion zones while protecting a network of green corridors and wetlands. The model works on the same principle, a small set of rules guiding new growth under green protection, so the scenario applies it to one designated expansion direction on the eastern fringe, toward Ndera and Masaka. Folder: scenarios/kigali_east/.

• Tiers (indicative values, household 4.3): 10,000 / 20,000 / 35,000 people/km² (roughly 25 / 45 / 80 dwellings/ha), shares 0.4 / 0.4 / 0.2 (low / medium / high).
• Target: about 60,000 new residents for one expansion area. Walks: 800 m to a centre, 400 m to green. Grid: 25 m, EPSG:32736. Dispersal: moderate.
• Terrain: slopes over 15° preclude development (slope_max_deg: 15, matching Rwanda’s percent-slope planning limits; a few percent of the window). Bands in the editable steep.geojson, from Copernicus GLO-30.
• The boundary and numbers are indicative, to be confirmed against the Kigali master plan zoning.

6. Medellín, Colombia: planned hillside expansion

Pajarito and Ciudadela Nuevo Occidente on Medellín’s northwestern slopes are a planned expansion of high-rise social housing served by the Metrocable. The scenario tests the growth rules where topography is the binding constraint: steep terrain sits in the unbuildable layer, and the green network and short walking distances have to work around it. Folder: scenarios/medellin_pajarito/.

• Tiers (indicative values, household 3.1): 8,000 / 18,000 / 35,000 people/km², shares 0.3 / 0.4 / 0.3 (low / medium / high); the high tier reflects the area’s housing towers.
• Target: about 40,000 new residents. Walks: 800 m to a centre, 400 m to green. Grid: 25 m, EPSG:32618. Dispersal: moderate.
• Terrain: slopes over 20° preclude development (slope_max_deg: 20; about 30% of the study window). The bands are in the editable steep.geojson, from Copernicus GLO-30.
• The boundary and numbers are indicative, to be confirmed against the POT zoning.

7. Freiburg, Germany: a validation scenario

Rieselfeld and Vauban in western Freiburg are two widely studied walkable districts, planned in the 1990s and often cited as models of the form this plugin aims for. The scenario runs the model where a good answer already exists: delete the two districts from the built layer (keeping a reference copy), let the model regrow the same land toward the districts’ real population, and compare the result against what was built. The comparison shows which behaviours the growth rules capture and which they miss. Folder: scenarios/freiburg_rieselfeld/.

• Tiers (indicative values, household 2.0, centred on the districts’ real densities): 8,000 / 14,000 / 22,000 people/km², shares 0.2 / 0.6 / 0.2 (low / medium / high).
• Target: about 16,000 residents (the two districts’ combined population). Walks: 800 m to a centre, 400 m to green. Grid: 25 m, EPSG:25832. Dispersal: off (both districts are contiguous extensions).
• The boundary is indicative; the validation protocol (remove, regrow, compare) is described in the folder’s params.json notes.

Adding a scenario

Scenario contributions happen in the repository; the steps are described in scenarios/README.md.

The scenario ZIPs contain map data © OpenStreetMap contributors, available under the Open Database License (ODbL), and slope bands derived from the Copernicus GLO-30 digital elevation model: produced using Copernicus WorldDEM-30 © DLR e.V. 2010–2014 and © Airbus Defence and Space GmbH, provided under COPERNICUS by the European Union and ESA; all rights reserved.