# Site02 — People, Transport and Flood Screening

Open [the people and transport atlas](Site02_People_Transport_and_Flood.html), also linked from [the main research atlas](Site02_Water_and_Ecology_Explorer.html). Both HTML files work offline; map data are embedded. External source links require internet.

## Reading the map

1. Use Close context for local flood overlap, or Wider context for the site plus its 2 km surroundings. This is not the full river catchment.
2. Choose a reference offset (+0, +2, +4, +6 or +8 m). This is not inundation depth, AHD or an annual probability.
3. Play the assumed weekday/weekend activity pattern or set an hour. These weights are scenarios, not footfall observations; the selected flood footprint remains fixed.
4. Switch to Census background for 2021 usual-resident density. Select a block to read its whole-block values. Block counts must not be treated as population in the clipped study area or flood footprint.
5. Search the affected-location table. Choose existing transport, activity locations, proposals, or roads. Locate any result to inspect its geometry and source. Export the current filtered list as CSV.
6. Concept proposals are anchored at existing facilities. Optional 400 m circles are straight-line proximity, not verified walking catchments or evacuation areas.

## Evidence

- ABS (2022), *Census mesh block counts, 2021*, released 28 June 2022. [Official page](https://www.abs.gov.au/census/guide-census-data/mesh-block-counts/2021). Counts are usual residents and dwellings, not people currently present. Retrieved 21 September 2026. Original workbook URL is recorded in `Metadata/Mobility_Sources/retrieval.json`; the matching 415 rows are preserved in `ABS_Census_2021_Filtered.csv`.
- ABS, ASGS 2021 full mesh-block boundaries. [Official GIS service](https://geo.abs.gov.au/arcgis/rest/services/ASGS2021/MB/MapServer/0). Bounding-box query returns 415 blocks, with 371 intersecting the final 2 km study area. All queried codes matched the count workbook. Display is clipped, but downloadable Census features retain full boundaries and full-block denominators.
- OpenStreetMap contributors (2026), mapped facilities and roads, retrieved through [Overpass](https://overpass-api.de/api/interpreter), 21 September 2026. Raw response, query and retrieval record are retained under `Metadata/Mobility_Sources`. [Copyright](https://www.openstreetmap.org/copyright), [ODbL 1.0](https://opendatacommons.org/licenses/odbl/1-0/). OSM-derived facility/road databases and their attributed adaptations are distributed under ODbL; ABS data remain separately attributed. Mapped features may be incomplete or describe different parts of one establishment. No service timetables or frequency analysis is included.
- Existing package flood footprints: ACT 2015 1 m DEM, connected static river-stage sensitivity model. Vertical datum unverified; constructed reference from the river elevation profile. See the main atlas and existing methods for source and hydrological limitations.

## Activity assumptions

Each category receives one 24-hour weight profile for a weekday and a weekend. Weights lie between 0 and 1; transport has illustrative commute peaks, schools have weekday arrival/departure peaks, recreation has afternoon/weekend activity, and civic/retail uses have daytime activity. No validation against actual arrivals, travel speeds, passenger demand or pedestrian counts was possible.

The map interpolates these weights between hours; a radial symbol fades to zero within a fixed 180 m radius. This is a location-based activity potential display, not a calibrated spatial density estimate. Overlapping features can produce stronger colour. The profile chart is the mean across mapped non-proposal features, not a person count. No walking or vehicle trajectories are simulated. Assumptions are inspectable in `Tables/Assumed_Activity_Profiles.csv`.

## Flood screening definitions

- `intersects`: mapped point/footprint intersects the selected connected flood footprint. A polygon may intersect at its boundary only; this does not establish that all facilities inside it are flooded.
- `not_intersected`: mapped geometry is wholly within valid DEM coverage and does not intersect that scenario. This is not evidence of flood safety.
- `not_assessed`: geometry has no overlap with the selected footprint but is partly or wholly outside valid DEM coverage. Unknown terrain is never classified as safe.
- `bridge`: road/bridge alignment intersects the footprint; deck elevation, structural conditions, hydraulic clearance and passability are unassessed.
- `coverage=partial` can coexist with an established overlap. The uncovered remainder stays unknown.

The valid-DEM polygon is reconstructed from the exact grid cells (1 m, EPSG:28355). Source GeoJSON is projected to the same CRS; invalid geometry is repaired in derived analysis copies only. Facility polygons use their mapped footprint; point features use their mapped location; relations without assembled footprints use a representative point. Candidate proposals use only a point anchor, not the final footprint or access route. Road lengths are geometric intersections, not measured closed-road lengths. GeoJSON downloads use WGS84 longitude/latitude.

## Concept improvement anchors

Four candidates investigate the mapped Queanbeyan Interchange, Queanbeyan rail station, City Council Library and Queanbeyan Park. They suggest accessibility, cycle parking, information and connecting-route studies. Positions are approximate existing-feature anchors, not surveyed new facility footprints. Suitability, ownership, access, demand, heritage/ecology and engineering conditions require further investigation. No proposal is designated a refuge or evacuation point. A park footprint may intersect a scenario even when its proposed representative anchor does not.

## Files for GIS and audit

- `GIS/GeoJSON/Mobility/Facilities_Activity_and_Proposals.geojson`: mapped facility geometry, source URLs, proposal flag and per-scenario result.
- `GIS/GeoJSON/Mobility/Roads_and_Flood_Screening.geojson`: mapped alignments, per-scenario intersection lengths and unknown-terrain lengths.
- `GIS/GeoJSON/Mobility/Intersecting_Road_Segments_XXm.geojson`: only the overlapping geometry for each offset.
- `GIS/GeoJSON/Mobility/Census_2021_Full_Blocks.geojson`: full mesh-block boundaries, counts and density denominator.
- `GIS/GeoJSON/Mobility/Valid_DEM_Coverage.geojson`: exact valid terrain domain.
- `Tables/Mobility_Flood_Screening.csv`: all facilities/proposals, coordinates and screening results.
- `Metadata/mobility_data_qa.json`, `Metadata/mobility_ui_qa.json`: join, geometry, link and interaction-logic checks.

No observed pedestrian counts, current transit schedule, site-calibrated hydraulic model or verified bridge-deck elevations were available. The original APRX and existing hydrology/ecology GIS sources are unchanged. The new mobility layers are supplied as GeoJSON, not inserted into the APRX. Live browser visual/click validation was blocked by the local-file browser security policy; source, interaction logic and exported map visuals were checked separately.