# Repository-grounded Arkle research matrix

## M1 — Are programme totals, sample populations and intervals comparable?

Guidance: knowledge/drill-results/extraction.md; knowledge/general/coaching.md (active).

Evidence: rns-9769143, rns-9773898, rns-9635389.

Finding: 2, 712 vs~2, 900 m; 1, 558 vs 1, 559 m; 974 scope; 403 vs 420 surface population; 62 collected vs 45 submitted vs 62 interim.

Visual/data-room implication: Unique hole/sample manifest; preserve dates and result types. No invented 87 trajectories.

## M2 — Does a surface footprint or contract resource trigger establish a resource?

Guidance: knowledge/resource-update/extraction.md; knowledge/cross-context/coaching.md (active).

Evidence: rns-9392626, rns-9769143.

Finding: No uranium resource estimated. Contract 20 Mlb/120 ppm threshold is contingent consideration only.

Visual/data-room implication: No orebody/resource solid or grade-tonnage arithmetic; target outline only.

## M3 — What is the selection bias, support and method fitness?

Guidance: knowledge/drafts/exploration-results-factsheet.md (draft; pending analyst validation).

Evidence: rns-9392626, rns-9769143, cim-exploration-2018.

Finding: 178 selected pits, trench/channel and pavement samples, gamma-selected 204 samples have different support. Control counts are not control performance.

Visual/data-room implication: Keep assays, eU, visual mineralogy and symbol bins separate. Request failed batches, blanks/CRM/duplicates, preparation and dispatch records. No draft score bands.

## M4 — Can historical holes and geophysics be reconciled to modern records?

Guidance: knowledge/drafts/exploration-intelligence-research.md (draft).

Evidence: rns-9501735, rns-9581203, rns-9635389.

Finding: 95 planned →106 logged on 8995/8298. Historical operator identities/files unresolved; 50 mvs 200 mline spacing not equivalent to positional accuracy.

Visual/data-room implication: Campaign provenance and raw grids; holdout registration controls; do not assert validated model performance.

## M5 — Which permit instance is application, readiness notice, consent or issued clearance?

Guidance: knowledge/drafts/permitting-factsheet.md (draft; second analyst pass pending).

Evidence: eis-8995-20, eis-8995-22, eis-8995-23, eis-8995-26, eis-8995-27.

Finding: 8026→8995 documentary history; expired 8026 ECC; conditional notice; wrong-licence heritage attachment. Current grants unresolved.

Visual/data-room implication: Instrument register with holder, commodity, application/grant/expiry and conditions; no green approved badge.

## M6 — Does proximity establish continuity, access or processing rights?

Guidance: knowledge/drafts/jurisdiction-assessment-factsheet.md (draft).

Evidence: rns-9392626, eis-8290-esia, eis-8995-20.

Finding: Neighbour deposits are contextual only. Communal/state-land and rights access described in dated reports, not an access agreement.

Visual/data-room implication: Separate land access, heritage, environmental and mineral rights; no inferred plant access or geological continuity.

## M7 — What ownership, cash and contingent obligations were acquired?

Guidance: knowledge/drafts/property-transactions-factsheet.md (draft).

Evidence: rns-9392626, rns-9773898.

Finding: 85%subsidiary;£2.032 mconsideration; minority carry/royalty details absent.

Visual/data-room implication: Ownership/carry/royalty and transfer instruments; dated £375 k/£242 k/£195 k schedule; resource acceleration separate.

## M8 — What is usable cash and what dilution can be established?

Guidance: knowledge/drafts/capital-raises-factsheet.md (draft).

Evidence: rns-9773898, rns-9761835, rns-9504442.

Finding: June/Septembercashdates differ; investing dominates spending; optiontable inconsistent.

Visual/data-room implication: Costed programme and cash bridge; no simple operating-burn runway or unverifiedfully diluted total.

## M9 — Which statements are compliant result categories and which are forecasts?

Guidance: knowledge/drafts/reporting-standards-factsheet.md; knowledge/drill-results/extraction.md; knowledge/resource-update/extraction.md (promoted topics governed by active packs; remaining draft questions only).

Evidence: rns-9769143, rns-9392626, cim-exploration-2018.

Finding: Assayawaited; eU needs calibration/disequilibrium; noresource. QP disclosure does not validate a digitised figure.

Visual/data-room implication: Carry units, method, date, uncertainty and evidence type on each feature; no proposed draft thresholds as scoring rules.

## M10 — Who owns technical, corporate and database review?

Guidance: knowledge/drafts/corporate-management-changes-factsheet.md (draft).

Evidence: rns-9581203, rns-9769143, rns-9773898.

Finding: Named technical roles recovered; data custodians to be confirmed.

Visual/data-room implication: Nominate company approver, exploration database owner, geophysicist and finance reviewer; no inferred track-record score.

## M11 — Can named holes be reconstructed as actual trajectories?

Guidance: knowledge/drill-results/extraction.md rules7,17,25,26; knowledge/cross-context/coaching.md (active).

Evidence: rns-9635389, rns-9769143, edn-paper-v3.

Finding: Five named legacy intervals recovered; 9 modern ULG collar labels retained with uncertainty. Per-hole orientation/EOH/RL and deviation rows unavailable. GSN archive access failed; no absence claim.

Visual/data-room implication: Five nonspatial downhole interval columns; source-local collar symbols; null unknown metadata; programme ranges not assigned to individual holes.

## M12 — What additional evidence is recoverable from embedded source images?

Guidance: knowledge/drafts/exploration-results-factsheet.md; knowledge/drafts/exploration-intelligence-research.md (draft questions; no scoring thresholds used).

Evidence: presentation, rns-9635389, rns-9769143.

Finding: Native presentation images reveal sample classes, channel banks, regional target outlines, geological fragments and Rössing context. Surface sample 2924 in image vs2923 in acquisition prose unresolved. Low-resolution RNS images are originals, not merely low-quality previews.

Visual/data-room implication: Additional source-local selectable geometry; retain originals, occlusions, fragments, dates and sample-selection caveats; no interpolated solids.

## M13 — Can broad public terrain support all four licence contexts?

Guidance: knowledge/drill-results/extraction.md rule7; knowledge/drafts/exploration-intelligence-research.md (active coordinate extraction; draft feasibility questions).

Evidence: copernicus-context-v3, eis-8995-25.

Finding: Copernicus DSM and September Sentinel mosaic cover regional bbox14.3–15.4E,22.65–21.65S. Provider georeferencing reconciles raster frames; independent surveyed drawing controls remain absent.

Visual/data-room implication: True-scale geographic terrain presets separate from source-local exploration frames. No surveyed collar RL inferred from DSM.

## M14 — Can the published ULG grid support a disclosed approximate terrain overlay without pretending it is survey control?

Guidance: knowledge/drill-results/extraction.md; knowledge/general/coaching.md; company-spotlight visual reconstruction and viewer guidance (active extraction guidance; display decision explicitly approximate).

Evidence: rns-9769143, geographic-ulg-v4, eis-8995-25.

Finding: Regular printed grid supports a candidate WGS84/UTM33S transform; qualitative regional/imagery consistency, but issuer datum and independent surveyed controls remain absent. Internal residuals are not geographic accuracy.

Visual/data-room implication: Optional approximate overlay with permanent badge, provenance and unmodified source pixels. Independent controls remain necessary for validated geographic layers.