Back Creek

Geological setting and mineral system

Back Creek is a five-occurrence prospect cluster (WN0006–WN0010) in hornfelsed Carboniferous Texas Beds on the western flank of the New England Orogen, ~30 km west of Tenterfield, NSW. GSNSW classifies all five occurrences as cassiterite (Sn) stockworks hosted in the Texas Beds (Brown, Henley & Stroud 2001, GS2001/087); the As–Ag–Pb–Zn–Bi associations carried into targeting are the WMC 1992–95 rock-chip overlay (R00003116). CAADIA's working hypothesis re-frames the cluster as an intrusion-related gold system (IRGS) with distal Ag–Pb–Zn–Bi replacement, zoned around a buried body defined by post-2020 airborne gravity, magnetics and radiometrics: a coherent ~2 km × 1 km gravity-high, magnetic- and radiometric-quiet anomaly east-northeast of the historic cluster. The body is a geophysical inference: its 200–400 m depth-to-top is a desktop estimate, no historical hole reached granite (deepest hole 204 m), the body has not been intersected and its mineralogy is unknown. The interpretation is preferred-but-not-exclusive; alternatives and their discriminants are set out in synthesis §5.4.

The grades — what the system has already demonstrated

Demonstrated result (GSNSW open-file)	Confirmation status
14.2–14.4 % As, 36–38 g/t Ag — Anomaly 155 North arsenopyrite gossan, rock chips DA653947 / DA709548 (R00003116; R00001192)	Independently confirmed by WMC re-sampling in a subsequent programme year and separate assay batch (14.2 → 14.4 % As; 38 → 36 g/t Ag). Trace Au 0.030 g/t reported on DA709548 only; DA653947 was not assayed for Au. The gossan itself has never been drilled.
2.5 % Pb, 65 g/t Ag, 240 ppm Bi, 1,570 ppm Zn — Genesis-area gossan, single rock chip DA653986 (R00003116)	Single sample; continuity along strike not yet tested. Spatial attribution is flagged for reconciliation against the GSNSW Genesis-zone centroid (synthesis App. C item C-13) and is resolved by Phase 1 mapping and ground-truthing ahead of BCRC003–004.
1,700 ppm As, 1,700 ppm Zn, 880 ppm Sn, 360 ppm Pb at 90–93 m — PDH GC2 (204 m, deepest hole; sample 900366, R00009457 App. I)	Cross-checked at v4.1 against the primary Comlabs appendix (App. C item C-10, closed); earlier misreported down-hole values were withdrawn at v3.0 and do not appear here. GSNSW independently summarises the Shell drilling (GS2001/087): best intersection 940 ppm Sn over 3 m. Not assayed for Au; system plausibly open at depth — suggestive, not conclusive.
Trace Au ~0.02 g/t on three BHP rock chips (SYD 69–71, SGS Job LA8255, R00011348); 0.030 g/t Au on DA709548 (R00001192)	The only gold determinations at Back Creek — two independent operators and laboratories, both on surface rock chips at Anomaly 155. No drill sample was ever assayed for Au (see §4).

Exploration history — three operators, documented exits

Every campaign found real mineralisation; every documented exit was model-driven under a tin-only target — no exit decision ever saw a gold assay.
Period	Operator (licences)	Model targeted	Documented departure
1979–82	Dampier Mining / BHP (EL 1235–1237)	Sn-greisen on the Ruby Creek Granite roof zone	Relinquished 1982 after 374 m of shallow percussion (PDH1–PDH8) returned only weakly anomalous Sn under the tin model; no Au assays run (R00011348, R00011350, R00015204).
1982–85	Kennecott → Shell → Renison → Billiton (EL 1474)	Deeper Sn granite-cupola test at Anomaly 153	Relinquished 1985: the two deep holes (GC1, GC2) intersected no granite or signs of granite, and GC1's calc-silicate mineralisation was recorded as fracture-related, not skarn-related — read as closing the case under the Sn model (R00009457, paraphrased).
1991–95	Western Mining Corporation (EL 1947 / 3976 / 3977)	Sn-stockwork in the apical zone of a buried Ruby Creek-equivalent granite	All titles relinquished May 1995 after the most intensive surface programme at the cluster; on the data-room evidence WMC did not drill within the title area. GSNSW subsequently closed the area against the Sn-greisen target model (R00003116, R00001192, R00001193; GS2001/087).

Why the grade is still available — the operators' documented record

Wrong commodity, wrong model. Shell's 1984 progress report records — paraphrased per the synthesis's verified reading of R00009457 — that no granite or signs of granite were intersected in either deep hole, and that the GC1 calc-silicate mineralisation is fracture-related rather than skarn-related. Under the Sn-cupola model that closed the case; under an IRGS model the same down-hole result would warrant a deeper follow-up hole (synthesis §4.4). GSNSW closed the district file against the Sn-greisen target model after the WMC relinquishment (GS2001/087); the IRGS / distal-replacement model has never been tested.

The analytical gap. The government-database audit finding is unambiguous: across 707 m in ten historical percussion holes, no sample was ever assayed for Au (MinView drillhole register; R00011350, R00009457). Te and W have never been assayed anywhere on the project and are [not yet measured]; Bi was measured on only two samples — 86 ppm (DA709548, R00001192) and 240 ppm (DA653986, R00003116). The only gold determinations — ~0.02 g/t on three BHP chips (SGS, R00011348) and 0.030 g/t on a WMC chip (R00001192) — come from two operators and two laboratories, both at surface. The pathfinder suite the IRGS model predicts was analytically invisible to every historical programme.

The untested target. No deeper follow-up was ever recommended — R00009457 contains no explicit forward programme — and none was drilled. The synthesis's precise formulations stand: no hole has tested the post-2020 airborne gravity–magnetic anomaly; the body has not been intersected and its mineralogy is unknown. The two highest-grade surface samples have likewise never been drilled: the Anomaly 155 chips lie outside the footprint of the shallow 1981 percussion holes, and the Genesis-area gossan sits ~1.9 km south of the part of the anomaly tested by PDH5–PDH6 (drill design §3.2–3.3). 

The miss, precisely. The historical programmes did not fail to find gold — they never looked for it. Each campaign asked one question ("is there an economic tin deposit?"), answered it in the negative, and left; because no sample across 707 m of drilling was assayed for Au, Te or W (R00011350, R00009457), those exits carry no evidentiary weight against the Au–Ag–Bi system the modern model targets. The one feature that could have redirected the search — the ~2 km × 1 km buried body — only became visible in airborne datasets released to MinView after 2020, more than 25 years after the last operator closed the file. The gap is therefore analytical and temporal, not geological: the data that would kill or confirm this play has simply never been collected.

System scale and the 2026 price lens

Footprint (dimensions, never tonnes). Airborne body ~2 km × 1 km, with a 200–400 m depth-to-top that is a desktop estimate (App. C item C-01). The main occurrence cluster spans ~4 km east–west on the GSNSW centroids (GS2001/087), with WN0006 a discrete occurrence ~7 km WSW. WMC dipole-dipole IP defines 50–60 mrad decoupled-phase anomalies against a ~20 mrad background on two 1 km lines (R00003116 §8.2.3). Depth-tested: the deepest historical hole is 204 m — historical drilling stops at or above the modelled top of the body.

Model class (attributed to the class, not the project). As the synthesis words it: Pogo averages 17 g/t Au; Fort Knox averages 0.85 g/t Au; the geological range across IRGS deposits is approximately one order of magnitude (synthesis §7.1.3). The conservative case carries equal weight: an intersection at the lower end of the class range (~0.5 g/t Au) would require multiple Phase 2 holes to assess, and Phase 1 alone will not deliver economic certainty; the independent reviewer's judgment at v3.0 is that the probability the body is IRGS-fertile is materially lower than 50 % on present data (synthesis §5.4). The rock-class similarity to Pogo / Fort Knox applies to the inferred host class only.

The 2026 lens. The synthesis makes no commodity-price recharacterisation and none is offered here. The commercial observation is narrower and documentary: all three historical programmes were funded, judged and closed on tin economics; the commodities the modern model targets — Au, Ag, Bi — were never assayed and therefore never entered any historical exit decision. CAADIA commercial framing; context only, no recovery, mining or processing costs applied, not attributed to any historical operator.

The targets

Primary — Anomaly 153 IRGS test (BCDD001–002). The known technical limitation is stated plainly: the airborne signature is non-unique — a sulphide-bearing skarnised metasediment package, or a buried sediment package of distinct mineralogy, can produce the same gravity-high / magnetic-quiet / radiometric-quiet response with no intrusion at all (synthesis §5.4.1, Alternatives C–E), and the desktop depth-to-top carries ±100 m-class uncertainty (§7.1.1). The designed test resolves both: ground magnetics and gradient-array IP constrain depth and geometry before collars are fixed; BCDD001 (300 m diamond) undercuts the GC2 90–93 m arsenic spike; BCDD002 (260 m) tests the core of the airborne body; down-hole magnetic susceptibility and petrography discriminate ilmenite- from magnetite-series rock directly on core.

Secondary — Anomaly 155 (BCRC001–002, 150 m each): first bedrock test beneath the re-sample-confirmed 14.2–14.4 % As / 36–38 g/t Ag gossan, with a 60 m step-out along the interpreted structure.

Secondary — Genesis (BCRC003–004, 120 m each): first test beneath the DA653986 Pb–Ag–Bi gossan, drilled after the C-13 spatial reconciliation; 50 m step-out.

Testing the thesis — gates, kill criteria, cost to answer

Phase & cost	Scope	Gate & kill criterion
Pre-drill gates — ~A$8,000 re-assay + ground geophysics	Au + 48-element ICP-MS re-assay of ~200 archived Londonderry splits (GC1/GC2 and BHP percussion); ground magnetics and gradient-array IP over each target before collars are fixed; DGPS validation of historical collar positions.	The kill answer for the analytical-gap thesis costs ~A$8,000 and is binary. Per the drill design (§13): “If the parallel gold and multi-element re-assay of the archived GC1 / GC2 splits returns no gold and no Bi–Te–W pathfinder enrichment, the central ‘never assayed for gold’ upside thesis weakens materially and the case for Phase 2 is reconsidered.”
Phase 1 — A$475,800 central case (ex-GST)	Six holes, ~1,100 m: BCDD001 (300 m) and BCDD002 (260 m) diamond at Anomaly 153; BCRC001–002 (150 m each) at Anomaly 155; BCRC003–004 (120 m each) at Genesis. QAQC: CRMs, coarse blanks and field/pulp duplicates at a combined ≥1-in-20 (≥5 %); DGPS collars; gyro down-hole surveys; residues and half-core lodged at Londonderry.	Anomaly 153 gates are split so a depth-model miss is not a play kill: failure to intersect the body “refutes the depth model and may justify a relocation rather than a relinquishment” (synthesis §5.3); failure of the class gate (not reduced / ilmenite-series) or the grade gate (no ≥0.5 g/t Au in any 1 m sample in or within 50 m of the body) is the deposit-model kill. Anomaly 155 / Genesis: a rootless gossan skin with no bedrock continuity and no gold downgrades the target to a surface anomaly — the conservative re-framing, not a zero outcome (drill design §13).
Phase 2 — success-contingent	Decision tree per synthesis §6.4: the positive case drills T2 plus infill / step-out and triggers independent JORC-compliant Competent Person review; the mixed case (T1 positive, T3 negative) re-frames the model from IRGS to distal-replacement only.	The negative case is bounded: Phase 2 is not pursued; the maximum write-off is the Phase 1 spend. Every failure mode above has a pre-specified, costed, binary test.

Technical Manager: Dr Tim McConachy FAusIMM (1974), Senior Fellow SEG, former Rio Tinto Chief Geologist.

Data room (under NDA): technical synthesis v4.1, Phase 1 drill programme design, source register and figure set.

Contact: Doug Alcock, Managing Director, CAADIA Metals Pty Ltd — dalcock@caadiametals.com.au

No JORC (2012) Mineral Resource, Ore Reserve or Exploration Target is declared or implied. Historical results are compiled from GSNSW open-file (DIGS) records, have not been verified by a Competent Person for public reporting, and are provided for context only. The IRGS / distal-replacement interpretation is a working hypothesis under test; alternative interpretations of the geophysical data exist. Earn-in terms are indicative and non-binding, subject to contract, due diligence and grant of the Exploration Licence.