# Wanuskewin geography and visual-source research

Verified 27 September 2026 UTC (26 September in Vancouver). Study bbox: west −106.610, south 52.214, east −106.580, north 52.233. Primary sources only. No external contact, purchases, account creation, restricted-content access, or full national/regional raster downloads performed.

## Result

**A practical terrain upgrade is available now:** NRCan's 2019 Saskatoon LiDAR-derived DTM supplies 1 m cells across the visitor-centre area and much of the valley/river corridor. It covers only part of the requested rectangle. Preserve an explicit no-data mask and visibly identify any lower-resolution fallback. It is not evidence for a complete, current photorealistic reconstruction.

**No openly licensed, park-wide photogrammetric mesh, interior scan, or current architectural model was confirmed.** Official park maps, architect photographs, and 2018 design fly-throughs are useful references. They do not carry a confirmed general asset-reuse licence. City aerial imagery is discoverable, but its item-level terms are blank; treat publication/texture reuse as unresolved.

## 1. Confirmed NRCan terrain

Live [STAC bbox query](https://datacube.services.geo.ca/stac/api/search?collections=hrdem-lidar&bbox=-106.610%2C52.214%2C-106.580%2C52.233&limit=10) returned one acquisition project, `SK-Saskatoon_2019-1m`. The [collection metadata](https://datacube.services.geo.ca/stac/api/collections/hrdem-lidar) identifies the licence as `OGL-Canada-2.0`, the vertical datum as CGVD2013, and the reprojection as EPSG:3979. Different acquisition projects are not vertically edge-matched. Spatial cell size is not a guarantee of 1 m positional or vertical accuracy.

| Field | Verified value |
|---|---|
| Item | `SK-Saskatoon_2019-1m` |
| STAC datetime | `2019-05-30T12:00:00Z` |
| Updated metadata | `2026-01-20T18:06:07.322317Z` |
| Cell size | 1 m |
| CRS | NAD83(CSRS) / Canada Atlas Lambert, EPSG:3979 |
| Vertical datum | CGVD2013 |
| Raster shape | 30,880 rows × 34,042 columns |
| Transform | `[-803956, 1, 0, 432286, 0, -1]` |
| DTM download | HEAD 200; 405,686,580 bytes; `Accept-Ranges: bytes` |
| Detailed coverage file | Successfully read, 1,814,528 bytes |

Direct assets:

- [DTM COG — bare-earth terrain](https://canelevation-dem.s3.ca-central-1.amazonaws.com/hrdem-lidar/SK-Saskatoon_2019-1m-dtm.tif)
- [DSM COG — surface including vegetation/buildings](https://canelevation-dem.s3.ca-central-1.amazonaws.com/hrdem-lidar/SK-Saskatoon_2019-1m-dsm.tif)
- [Simplified acquisition extent GeoJSON](https://canelevation-dem.s3.ca-central-1.amazonaws.com/hrdem-lidar/SK-Saskatoon_2019-1m-extent.geojson)
- [Detailed DTM/DSM coverage GeoPackage](https://canelevation-dem.s3.ca-central-1.amazonaws.com/hrdem-lidar/SK-Saskatoon_2019-1m-coverage.gpkg)
- [DTM VRT](https://canelevation-dem.s3.ca-central-1.amazonaws.com/hrdem-lidar/SK-Saskatoon_2019-1m-dtm.vrt)
- [DSM VRT](https://canelevation-dem.s3.ca-central-1.amazonaws.com/hrdem-lidar/SK-Saskatoon_2019-1m-dsm.vrt)
- [Federal product catalogue](https://open.canada.ca/data/en/dataset/957782bf-847c-4644-a757-e383c0057995)
- [Open Government Licence — Canada](https://open.canada.ca/en/open-government-licence-canada)

### Actual coverage, not merely catalogue overlap

The simplified extent crosses the study bbox diagonally. Its local northwest boundary runs from `[-106.611902,52.217266]` toward `[-106.578514,52.236821]`. The southeast side is covered. The complete polygon is at the GeoJSON link above; do not substitute the item's rectangular bbox for it.

The detailed GeoPackage contains `dtm` and `dsm` layers, each one identical 53,043-vertex polygon in EPSG:3979. Schema: `fid INTEGER PRIMARY KEY`, `geom POLYGON`, `DN MEDIUMINT` (value 1). Its bounds are `[-794033,410730,-772946,429272]` in EPSG:3979.

A direct WKB polygon test, with Lambert Conformal Conic projection using the GPKG CRS definition, confirmed:

| Test coordinate (longitude, latitude) | Detailed coverage |
|---|---|
| Visitor-centre reference `−106.596,52.224` | Inside; projected `−779164.9204,427770.9357` |
| `−106.595,52.220` | Inside |
| `−106.600,52.226` | Outside |
| Southwest, southeast, northeast bbox corners | Inside |
| Northwest bbox corner | Outside |

A 50 × 50 regular point sample estimates **68.16% of the geographic bbox** inside the detailed polygon. This is a coarse coverage estimate, not a survey measurement. The parent implementation separately read a native 2,382 × 2,433 COG window and reported 64.5% valid raster cells, minimum 466.49 m and maximum 501.88 m. These percentages use different extents: the rectangular projected raster window includes margins outside the original geographic bbox. Parent-verified samples: visitor centre 492.98608 m, a Path of the People sample 466.8745 m, and a bison-trail sample 484.7878 m. These are DTM samples, not independent ground-survey checks.

### Suitable implementation

Read only the study window using COG byte ranges, preserve `nodata`, convert horizontal coordinates correctly, and document vertical datum before blending with Terrarium. Use the DTM for ground and retain its coverage mask. DSM minus DTM can inform canopy/building-height investigation within covered pixels, but the 2019 acquisition predates the completed 2020 building renewal: it cannot certify the present building roof or current landscaping.

The same bbox also returns mosaic items `4_3-mosaic-1m` and `4_3-mosaic-2m`; their large 500 km tile bounds do not prove full local coverage. Their current item date is a mosaic metadata date, not a fresh park survey.

NRCan's [COG/STAC guidance](https://nrcan.github.io/CanElevation/stac-dem-mosaics/) explicitly supports windowed reads. Its [February 2025 access guide](https://download-telecharger.services.geo.ca/pub/elevation/dem_mne/HRDEMmosaic_mosaiqueMNEHR/HRDEM_Mosaic_WCS-WMS_instructions_EN.pdf) also documents WCS at `https://datacube.services.geo.ca/ows/elevation`, but says WCS will be phased out and COG access is preferred. A tiny WCS probe in this research returned HTTP 308; no WCS raster result was relied on.

### Raw LiDAR point clouds: lead only

[NRCan's official point-cloud AOI guide](https://nrcan.github.io/CanElevation/pointclouds/projects-tiles-by-aoi/) links national project and tile indexes. The [project index ZIP](https://canelevation-lidar-point-clouds.s3.ca-central-1.amazonaws.com/pointclouds_nuagespoints/Index_LiDARprojects_projetslidar.zip) responds 200 and is 4,127,732 bytes, last modified 17 September 2026. **No Wanuskewin raw point-cloud tile was spatially verified in this pass.** The available DTM/DSM must not be described as an obtained point cloud or mesh.

## 2. Saskatchewan and City of Saskatoon sources

| Source | What was verified | Reuse/coverage decision |
|---|---|---|
| [WSA LiDAR Read Me item](https://www.arcgis.com/home/item.html?id=cbabfddeb5054cfc9f6d91e5cecfe872) | Official Water Security Agency item says WSA LiDAR access is through the federal HRDEM catalogue. Its [PDF download](https://www.arcgis.com/sharing/rest/content/items/cbabfddeb5054cfc9f6d91e5cecfe872/data) responds 200, 156,257 bytes. | Confirms an access route, not a second independent Wanuskewin acquisition. |
| [Saskatchewan All_LiDAR_BareEarth ImageServer](https://gis.saskatchewan.ca/imagery/rest/services/All_LiDAR_BareEarth/ImageServer) | Service describes Consul/Arena and Great Sand Hills projects, 0.5 m cells. | **Not Wanuskewin coverage.** Do not infer coverage from the service's provincial name. |
| [WSA Westside Irrigation project](https://www.arcgis.com/home/item.html?id=ded1f1e4349f4deabf5bdd3a82404886) | Catalogued extent ends at longitude −106.645. | West of the study bbox; no overlap. |
| [COS Current Imagery web map](https://www.arcgis.com/home/item.html?id=8d403bbee7b04c78941023d6cca003c5) | CitySaskatoon.Admin, public-authoritative item; explicitly described as 2021 aerial imagery. | Useful current-conditions lead. Full park pixel coverage not verified. |
| [COS NW imagery item](https://www.arcgis.com/home/item.html?id=a27b621ae09648fe9b344b2b002d7360) and [tile service](https://tiles.arcgis.com/tiles/eJz9754Ox6TaFSC2/arcgis/rest/services/COS_Basemap_Imagery_WM_NW/MapServer) | Extent encloses study bbox; public item metadata; EPSG:3857 tile pyramid to level 21; `licenseInfo` and copyright text blank. | **Licence unresolved.** Tile level/resolution is not proof of source ground-sampling distance. No imagery copied into the viewer. |
| [Natural Area Inventory 2019](https://www.arcgis.com/home/item.html?id=99d1cd922a6943368b12fb0385b13137) | City/Meewasin green-network polygons, digitized from City 2017 7.5 cm imagery. Extent intersects study bbox. | Item terms require City and Meewasin approval for external use/publication. Not freely reusable solely because public metadata can be queried. |
| [FlySask](https://www.flysask2.ca/) | SGIC says public viewing is available and announces migration to [FlySask3](https://web.flysask3.ca/) on 30 September 2026; extra high-resolution data is member-only. | Viewing access is not confirmed texture redistribution permission. No protected access attempted. |
| [ISC SGIC Ortho Imagery](https://www.saskregistries.ca/MapsandPhotos/Imagery/SGICOrthoImagery) | Official distributor; listed product specification includes 0.6 m pixels, NAD83 CSRS98/UTM 13, GeoTIFF/TIFF/JPG, and an end-user licence. SGIC-2 availability is location-dependent. | Commercial/licensed acquisition lead. Exact Wanuskewin tile/date and allowed derivative use not verified; nothing purchased. |

The [City's general open-data page](https://www.saskatoon.ca/services-residents/connect-your-city/open-data) describes freely reusable datasets, but that broad statement does not establish that every public ArcGIS item is part of the open-data offering or override an explicit item restriction.

## 3. Official maps, architecture and existing virtual material

- [Official trails page](https://wanuskewin.com/experience/trails/) links the [park trail-map JPEG](https://wanuskewin.com/isl/uploads/2023/02/Wanuskewin-Trail-Map-scaled-1.jpg). It gives names, colours and lengths: Discovery 0.8 km/red; People 2.1 km/dark blue; Harmony 1.3 km/light blue; Bison 1.6 km/yellow; Bison Viewing 0.4 km/purple, plus Meewasin/green. This is a reference map, not a georeferenced surveyed trail dataset. The page also states no drones on site.
- [Architect aodbt's project page](https://aodbt.com/projects/wanuskewin/) identifies original completion in 1992 and renewal in 2020, total size 59,909 sq. ft., cedar cladding and bison-related massing. It explains the later gallery and gathering-space additions and carries credited project photography. Use to resolve roof/massing/material questions; no downloadable CAD/BIM or reproduction licence was found.
- [Landscape designer Crosby Hanna's project page](https://www.crosbyhanna.ca/projects/wanuskewin-heritage-park) documents its trail, creek-crossing, plaza, bison-viewing and landscape work, including later renewal. It is a first-party design reference rather than an as-built survey.
- [Wanuskewin's 29 August 2018 design announcement](https://wanuskewin.com/wanuskewin-releases-design-plans/) links official concept fly-throughs: [facility](https://youtu.be/XXxHnuqzvhc), [exhibits](https://youtu.be/nAA7O3jex0E), [playground](https://youtu.be/5edLH7F_Va4). Links were verified on the official page; the videos were not visually reviewed here. Label them **2018 design proposals**, not a current scanned tour.
- [Official kiskeyihtamowin course](https://learn2.wanuskewin.com/courses/kiskeyihtamowin/) advertises a virtual educational video tour with archaeology, plants, bison and other modules; its public page says not enrolled/contact for price. Only the public landing page was read. It is not an open 3D-asset library.

[Wanuskewin's terms](https://wanuskewin.com/terms-of-use/) reserve copyright in site materials and describe case-by-case permissions. These public pages support research and links; they do not establish permission to republish photos, videos, maps, artworks or interpretive content as application assets. The separate rights/reproduction page presented a verification interstitial and was not bypassed.

## 4. Public ecology and conservation references

- [Wanuskewin land page](https://wanuskewin.com/our-story/land/) explains restoring agricultural fields to native grasslands with the conservation herd. Suitable context for grassland/riparian habitat zones, not a measured vegetation inventory.
- [Wanuskewin bison page](https://wanuskewin.com/our-story/bison/) dates the bison return to December 2019 and describes conservation, ecological restoration and cultural relationships. Animals should be labelled illustrative unless positions/counts come from a dated observation.
- [Wanuskewin 2024–25 annual report](https://wanuskewin.com/isl/uploads/2025/07/Wanuskewin-2024-25-Annual-Report-1.pdf) reports grassland monitoring, northern leopard frog and beaver surveys, trail work, invasive-species control, prescribed burning, and hawthorn/buffaloberry planting. This supports a varied landscape and seasonal interpretation; it does not locate each plant or animal.
- [Parks Canada: bison transfer to Wanuskewin](https://www.parks.canada.ca/pn-np/sk/grasslands/nature/faune-wildlife/bison/wanuskewin) is a primary partner source for the conservation relationship and includes a public video/transcript.
- [Official self-guided activities](https://wanuskewin.com/experience/tours/self-guided-activities/) supplies educational context for the bison viewing platform, playground, gardens and exhibits. It places the discussed petroglyph in the exhibit hall, but provides no verified downloadable 3D scan.

## Priorities for a more accurate recreation

1. Use the obtained 1 m DTM window and an explicit coverage mask. Display acquisition year and identify fallback terrain.
2. Cross-check every rendered trail against the official map and measured terrain, retaining uncertainty where source geometry differs. A source map should not be claimed as surveyed geometry.
3. Rebuild the visitor centre only to the detail actually established by current permitted references. Separate real footprint/massing evidence from estimated heights and illustrative roofs.
4. Obtain explicit usable imagery rights before adding a real aerial texture. The verified City 2021 imagery and SGIC routes are better leads than assuming public viewing permits copying.
5. Represent grassland, shrubs, riparian vegetation and bison in a way consistent with official conservation sources, while labelling placement as illustrative. A measured vegetation inventory, current paddock plan, and building model remain unconfirmed.

No claim of a complete digital twin, surveyed trail alignment, current canopy inventory, raw LiDAR scan, or publication-cleared park media is supported by this research.