The corrosion of those aluminium hangers would very likely have been accelerated if water from spray or a film trickling down the wall, perhaps only in wet conditions, met it. Water exiting as vapour in the air is effectively distilled but cave water holds dissolved calcium-carbonate or is slightly acid from atmospheric CO2 and/or organic compounds in the soil above the cave.
After all, it is this natural acid that dissolves the limestone to make the cave.
These hangers replaced M8-threaded expanding steel inserts in holes drilled in the wall. The rope was attached to these via aluminium hangers taken along with the ropes, and removed on exit. All very well but the steel of these building fixings obligingly corroded even without dissimilar-metal help, and constant use wore the threads.
.
My work involved dangling test-pieces in a tank of tap-water – hard in Dorset – kept clean with swimming-pool filters and additives: calcium hypochlorite disinfectant, copper sulphate algicide and acid or alkali pH-corrector. The water was perfectly safe for us, but I was forever trying to tell people that the lovely black anodising of their prototypes of equipment intended for sea-water immersion, cannot protect the aluminium from the stainless-steel screws* even in slightly electrolytic, fresh tap-water! (Anodising is slightly porous.)
.
What of aluminium railway track?
The whole thread does make me think of miniature railways assembled from aluminium-alloy rail using mild-steel or stainless-steel fishplates and fastenings.
A raised track might not suffer too badly, and is readily-accessible for repairs, but even the best-maintained ground-level track will inevitably collect some leaf-litter and even soil around it, especially in wet windy Winters; helping keep the joints moist with more electrolyte-rich water.
A steel fishplate can be insulated from the rail by an interposed plastic or rubber shim, but there are still the screws to consider.
Incidentally, even stainless-steel can corrode in certain contact conditions, depending partly on the alloy; but in contact with aluminium it is still the latter metal that is eaten.
.
*Stainless-steel screws and nuts…
As we know these can gall, a topic sometimes aired on this site.
My work also involved assembling some test-pieces, and one had a ring of M5 st-stl screws and ‘Nyloc’ nuts on flanges inside a narrow recess. One galled… The only answer was to keep tightening it until the screw sheared.