Partner review build Prepared for review by tribal natural resources staff. Not for public distribution. Every layer shows habitat capability, never occurrence.

Traditional foods · habitat capability

Manoomin (Wild Rice) habitat screen

Where north-central Minnesota waters are still biophysically capable of supporting Manoomin (Wild Rice), what limits them, and which basins carry mapped presettlement evidence. Lidar and imagery refine capability next; occurrence beyond DNR’s public rice waters is never inferred. Region: Beltrami, Cass, Clearwater, Hubbard, and Itasca counties.

525

basins identified by DNR as wild rice waters — protect first

39,737 acres of rice delineated by DNR across 281 of them; 646 basins on the MPCA wild rice inventory

992

basins are capable and carry mapped presettlement evidence, with no barrier detected

32

restorable basins where a barrier is identified — drained, impounded, ditched, or sulfate-impaired

0.80

AUC on the held-out counties — how well capability separates DNR rice waters from other basins (0.5 is chance)

347 of 391 held-out rice waters rated capable, against 0.58 of all held-out basins

A diagnosis, not a score

Every basin gets a category and a named limiting factor rather than a bare number. Capability uses a weighted geometric mean, so a single fatal factor sinks the result instead of being averaged away — and the limiting factor reported is, by construction, the one doing the most damage.

A detected barrier never lowers capability. It changes the category and names the obstacle, because a capable site with a ditched outlet is a restoration project with a known intervention, not a bad site.

What this map does not show

This shows capability — where habitat could support the guild — not occurrence. Precise locations of active stands and gathering sites are community knowledge, not ours to publish or sharpen. The only current-presence layers here are MN DNR's and MPCA's already-public wild rice waters lists, shown at their own resolution and never refined.

Water depth, substrate, and current are not directly measured. They are represented by basin morphometry, DNR classification, and — where lidar has been run — the low shore margin, which is why this is a screening tool and not a substitute for a site visit or a bathymetric survey.

Method & limitations

  • Reference condition. Presettlement vegetation comes from Marschner's reconstruction of Public Land Survey notes (1847–1907). It is mapped coarsely: basins under roughly 40 acres sit inside upland polygons and the reference map cannot speak to them. Those are reported as historical evidence unresolved — not as evidence of absence.
  • Calibration and validation. Factor bands were set from DNR-identified rice waters in Beltrami and Hubbard only. Rice presence is never a scored input, and the Cass, Clearwater, Itasca basins were not used to tune anything, so the AUC of 0.80 reported there is an independent check. The plain "share rated capable" is shown beside the base rate for all basins because on its own it flatters any model that rates nearly everything capable.
  • Barriers. DNR waterbody classification (drained, partially drained, impounded), ditch-named watercourses at the basin margin, and MPCA's 2024 wild-rice sulfate impairment listings (5 basins in region). Lidar outlet and ditch detection is the next build.
  • Through-flow. Each basin is classed river (a DNR river centerline runs through it), stream (a named watercourse connects) or isolated. Shown in the popup, not scored: on the calibration counties rice waters are 35 of 69 river basins, 5 of 90 stream-connected, 94 of 1729 isolated — connection is positive evidence at this resolution, so it is not a penalty.
  • Depth, wetland community, and shore soils. Where DNR bathymetry exists, the share of the basin no deeper than its shallowest mapped contour (usually 5 ft) is scored directly — rice grows in roughly 0.5–3 ft — with contour rings polygonised so shoals inside deep water count as shallow. The Minnesota updated National Wetlands Inventory gives the share of the basin and its 150 m shore mapped as deep marsh, shallow marsh or shallow open water, and USDA SSURGO gives the share of the shore in wet, organic or hydric soils — the soft substrate rice beds sit against. All three are scored only where they exist.
  • Lidar shore margin. Where etl_lidar.py has covered a basin, the share of its 150 m shore ring lying within 1 m of the water surface is scored as a low-margin proxy. Basins without it are scored without that factor.
  • Near-infrared imagery. Where etl_nir.py has covered a basin, the popup shows the evergreen and deciduous canopy share of its 150 m shore ring from Copernicus Sentinel-2 (context, not scored). The share of the water surface that is vegetated in late summer is computed for validation only and is not published: on a lake DNR has not listed it would say “something emergent grows here”, which is occurrence, not capability.
  • Sulfate. Only MPCA impairment listings are used; measured sulfate concentrations are not yet joined.
  • Screening only. No entry on this map should be treated as a restoration recommendation without a site visit, and any work on or near tribal land or treaty-reserved harvesting waters belongs to the nations whose waters they are.