mineral systems
Alteration, structure, a favourable host, the right tectonic setting: each of these is commonly found around mineralisation, and each is also found, far more often, around nothing at all. This is why Lithoen records associations and refuses to score them.
About: Mississippi Valley-type system · Porphyry copper system
The clearest statement of the problem is in the USGS Mississippi Valley-type model, which is describing districts it knows well. Each ore-controlling feature typically has an unmineralised counterpart in the same district, which suggests that the coincidence of several controls must have been important in localising ore. The same faults, the same collapse breccias, the same lithological transitions occur barren, in the same rocks, a short distance away.[1]
So a fault is not a pathway because it is a fault. The model treats ore controls as factors in fluid transmissivity that focus flow and create opportunities for deposition, which means a structure matters when it did that — and whether it did is a separate question from whether it exists.[1,2]
Alteration mineralogy is the most detectable expression of a hydrothermal system. Sericitic, argillic, advanced argillic and propylitic assemblages, along with iron oxides and hydroxides and silica, can be mapped from orbit through their absorption features in the visible near-infrared, short-wave infrared and thermal infrared bands.[1,2]
What the instrument detects is the mineral. It does not detect the cause. Hydrothermal systems that never concentrated a metal produce the same minerals, and propylitic alteration in particular is widespread and weakly diagnostic. A map of alteration is a map of where water was hot, which is a much larger and much more common thing than a map of where metal accumulated.[1]
The ratio is starker in magmatic nickel systems. Occurrences are present on most continents; deposits of sufficient size and grade to be mined are rare, with the major examples confined to five countries. The setting — flood basalt near a craton margin — covers a substantial fraction of several continents.[1]
Every indicator in the corpus is stored as an edge that must carry a statement of what else produces the same observation, and the schema refuses the edge without one. An indicator edge may never be recorded as settled fact: association is inferred at best. This is enforced by a validator, not by convention.[1,2]
There are no scores. Lithoen assigns no probability, ranks no ground, colours no map and recommends no drilling. Phase 3 exists to make the question well-formed — what evidence would we expect if this system operated here? — and answering it about a place needs spatial data, validation against known outcomes, and an accounting of uncertainty that this corpus does not have and does not pretend to.[1,2]
Often said
“Strong alteration means a deposit is nearby.”
What the evidence supports
Alteration minerals are produced by any hydrothermal system, mineralised or not, and infrared mapping detects the mineralogy without distinguishing its cause. Propylitic assemblages in particular are widespread and weakly diagnostic.
Supporting records: pc-remote-expression
Often said
“If enough indicators line up, the conclusion follows.”
What the evidence supports
The coincidence of several controls appears to matter, which is a statement about what localised known mineralisation — not a rule that produces a deposit when satisfied. The USGS porphyry model notes that many features of even that best-studied system remain poorly understood as controls of ore formation.
Supporting records: one-control-not-enough · pc-controls-poorly-understood
Bracketed references link to the structured entity records whose claims support each paragraph. Last updated 2026-09-07.