How to estimate dam storage volume from satellite imagery
Every entitlement audit eventually comes down to one question: how much water can that dam actually hold. The licence register tells you what was approved. It doesn't tell you what got built since, and if you're working off a satellite pass over a catchment rather than a fresh survey, you need some way to turn a waterline polygon into a storage figure.
Surface area is the easy part
Delineating the waterline from 0.5-2 m imagery is straightforward enough. You trace the edge of standing water, measure the polygon, and you've got a surface area in square metres. Do that for a ring tank or a gully dam and you've got your starting number. The problem is that area tells you almost nothing about volume on its own. A shallow, wide turkey nest and a deep embankment dam across a valley can show the same surface area from above and hold wildly different amounts of water.
Volume depends on the shape of the basin below the waterline, and that's the part a single overhead image can't see directly.
Area-to-volume curves and why they vary by dam type
This is where the hydrology gets interesting and the estimating gets rough. Catchment management authorities and state water agencies have published area-to-volume relationships for farm dams for decades, usually as a power curve: volume rises faster than area as a dam gets bigger, because deeper storages also tend to be built with steeper walls and more fill. The coefficients differ by construction type. A dam built across a natural gully behaves differently to an excavated tank or a ring tank built up from flat ground, because the cross-section of the basin is different in each case.
None of these curves are precise for an individual storage. They're regional averages built from surveyed dams of a known type, and they're meant to give you a rough capacity, not a number you'd put in a compliance notice without checking it. If you're triaging a few hundred storages across a catchment, a curve like this gets you from "big blue polygon" to "probably somewhere in this range of megalitres," which is enough to decide what needs a closer look.
Where the estimate breaks down
A few things make this harder than it sounds from a desk.
The waterline you're measuring reflects water level on the day the image was captured, not the dam's full capacity. A storage photographed at 60% full during a dry spell will under-report its true volume against any curve built for a full dam. Turbid or algae-affected water can also throw off the edge detection, especially on older or lower-resolution passes. And none of this accounts for sediment that's built up in the base of an older dam, which quietly eats into real capacity without changing the surface footprint at all.
Dam wall height is the other blind spot. Two storages with identical surface area can differ by a metre or two of wall height, and that difference compounds fast once you're estimating a volume rather than measuring one. Without a known dam type and a recent full-supply reading, any imagery-based volume stays an estimate with a wide error band.
Using it to prioritise, not to sign off
For an entitlement audit, that's usually fine, because the question isn't "what's the exact capacity of this dam," it's "which of these three hundred storages looks bigger than what's on file." An area-based estimate, even a rough one, is enough to rank storages by how far they've drifted from the licensed volume and send the biggest outliers to the top of the site-visit list. The on-ground inspection, with an actual depth reading or a surveyed bathymetry, is still what closes out the file.
That's the gap a yearly pass over the catchment is built to fill: a map layer of candidate structures checked against the current licence register, so the estimating work starts from a shortlist instead of a blank aerial photo of the whole district.
If your audit season starts with "where do we even look first," that's worth a closer look.