When I worked with Welsh Water's Scientific Services (predecessor of National Rivers Authority, now the Environment Agency) in 1985-86 on monitoring Cwmrheidol and on preliminary proposals for ameliorating its polluting discharges, the total zinc emmanating from Level 6, Level 9 and leacing from the dumps averaged around 50kgs per day or around 20 tones a year (daily figures fluctuated considerably depending on rainfall - anything from 20kgs to 200kgs per day). If I recall correctly, lead levels were around one-fifth the levels of zinc (i.e. around 10 kgs a day / 4 tonnes a year), and cadmium were less than one-tenth the levels of zinc (i.e. around 5 kgs per day / 1.5 tonnes a year). In total this amounts to around 25 tonnes a year. I can't recall how much iron was contained in the effluent though its levels also reflected rainfall and rates of through-flow.
The quoted figure of 98 tonnes a year may reflect either a radical change in the hydrology of the mine or a worst-case calaculation based on the worst individual days magnificed to annual levels, or iron as well as zinc/lead/cadmium?
The spent filter medium is likely to be sufficiently different to the usual types of zinc/lead/cadmium bearing wastes that reprocessors might well offer very little for it or charge for reprocessing it. At the very least, its large volume compared to its reclaimable metallic content may tip it from being an asset to be sold, to a liability that has to be subsidised to be disposed of. Nevertheless, its metallic content will very likely defray at least part of its processing costs.
A 1975 plan of Level 9 prepared when Rheidol Vale Properties proposed to drill from within the level states that the level was blocked by a major fall 500 ft from the portal. It also states that the tramway was 1' 11.5" gauage and that there were eight steel rings at three foot centres between the portal and rock head - do these details tally with what you and Simon found on site John?
More general plans of the mine prepared at the same time (undoubtedly based on older plans as most of the mine was inaccessible) show a pair of closely spaced winzes approximately 200ft west of Level 6 crosscut, descending from Level 6 on what was termed the South Lode to Level 9 on the South Lode. Presumably the water referred to in John's fascinating account was descending one or both of these winzes. (The two apparently seperate lodes named on old plans appear to actually be the south and north sides of one very wide lode which is up to 150 feet wide).
The Bwlchgwyn stream has repatedly sunk into the stopes on the top of the mine. Either Welsh Water or the county council installed concrete channels in about 1984 to carry the stream across areas where it was sinking but finances limited the extent of the work and it did not completely solve the problem even then. As John's photos eloquently showed, the stream found new places to sink into the stoped-out ground. Presumably the stream had been diverted and/or channelled across these areas in the nineteenth century to stop it sinking into the workings.