Field notes: the seep that wasn't on any permit map

Bright orange, iron-stained water flowing over rocks in a shaded rocky streambed

October 14. Permit file for the site says three discharge points. We counted four from the ridge road, and the fourth one was the color that gets your attention: that flat orange-brown you only see where iron is coming out of rock that's been exposed to air and water for longer than anyone meant it to be.

It had no outfall number and no monitoring station, just a stain running down a talus slope below an adit that the file says closed in 1987.

This is the part of the job nobody puts in the training manual. Permit boundaries describe where a mine operator is supposed to be discharging. The water comes out wherever the rock lets it. Old workings don't respect survey lines.

A shaft collapses, the water table shifts a few feet, and twenty years later it finds a new way downhill through a fracture nobody logged. The paperwork stays still. The hydrology doesn't.

We get maybe two or three of these a season on a watershed this size, usually on the legacy side rather than active operations, since active sites at least have someone checking the pond liners. Legacy sites have nobody checking anything until a hunter reports orange rocks or a stream survey crew trips over it, which is roughly how this one got found too, except we'd flagged the pixel three weeks before the field visit confirmed it. The stain had been visible in imagery since early September. It just took that long to get boots there.

That gap is the whole problem with ground-truth-only screening. A sampling budget can cover plenty of sites in a season, but it can only cover the sites that made the list, and the list is only as good as whoever built it. A list built from permits, old inspection records, and institutional memory is a nice way of saying it's built from whatever the last regulator happened to write down before they retired. Nobody drafts a line item for a seep that opened up on its own three summers ago.

What the imagery gives you instead is a scan of the whole drainage, not just the parts with a file number. Iron staining and acid rock drainage both have a spectral signature that reads differently from the surrounding soil and vegetation, month over month, whether the Water Quality Division knows the discharge exists or not. The question shifts from whether a known point is still within limits to where in this basin the chemistry looks different from last month, which is a much shorter route to a site like this one.

We've started running a watershed like this one through a monthly screen built for exactly this signature, flagging ponds and plume extent from wide-swath imagery before a crew ever drives out, and it's changed the order operations run in. Instead of inspecting the permitted points on schedule and hoping nothing else surfaces in between, you get a flagged layer first and send the truck to the thing that actually changed. You can read more about how that screen works on the Mine Water Quality site.

The seep is logged now. It'll get a discharge monitoring requirement next quarter, assuming the enforcement side moves at its usual pace, and someone will eventually trace it back to a stope that was supposed to be sealed in the original reclamation plan. None of that changes the fact that it sat there unregulated and unmonitored for who knows how long, visible from above the whole time to anyone looking at the right band.

If you're screening a watershed and want to see what's staining it this month, that's the whole reason this product exists.

Start a pilot

Back to the blog