Field notes from our first appeal season on satellite imagery

A small single-engine propeller airplane flying against a clear blue sky.

We'd run the same aerial contract for three cycles straight. Every four years, a plane came through in March, the vendor delivered ortho tiles that fall, and our GIS tech spent the winter re-digitizing rooftops and driveways parcel by parcel. It worked. It was also the reason our fee roll was always a little behind the actual ground.

This year we tried something different for the impervious layer behind the stormwater fee, and appeal season was the real test of whether it held up.

Week one. The usual letters came in: homeowners who'd torn out a patio, a church that paved half its gravel lot for parking, a strip mall owner convinced his square footage was wrong. Normally this is where I'd pull the four-year-old ortho, squint at a driveway that may or may not still exist, and tell the appellant we'd need to schedule a site visit. This year I pulled the current layer instead. For about a third of the appeals, the imagery already matched what the appellant was describing, because it had been refreshed more recently than the paving job they were complaining about.

Week three. The hard case showed up: a warehouse owner who'd added a loading dock extension sometime in the last year. Old workflow, that's an automatic deferral to next cycle, because nobody flies a plane for one parcel. This time the next scheduled refresh caught it within a few months, not years, and we adjusted the billing unit without him having to file a second appeal the following season. That's the part that changed how I think about the roll: the gap between the ground changing and the fee roll catching up got a lot shorter.

Week five. Our appeal board asked the question I expected: is this as sharp as the old flyover? I told them the truth, which is that we're working at sub-half-meter resolution and we're not claiming pixel-for-pixel parity with a low-altitude aerial pass. What we gained instead is cadence.

The old contract gave us one good look every few years and a slow fade after that. A rooftop addition from year two of the cycle sat undetected until year four, quietly undercounting someone's impervious area the whole time, and nobody noticed until a neighbor pointed it out at a rate hearing. That's the failure mode nobody budgets for: not bad data, just old data wearing a recent date stamp.

Week seven. One thing I didn't expect: fewer "I want a second opinion" escalations. When the imagery behind a denial is eighteen months old instead of three years old, appellants have less room to argue the lot "probably changed since then." They still argue. They just argue about less.

By the end of the season our denial rate didn't move much, and I'm not going to pretend otherwise. What moved was the number of appeals we could resolve on the first call instead of scheduling a field crew, and the number of parcels sitting in the roll with impervious data that predated the current mayor's first term. For a program where the whole fee structure rests on "square footage of sealed surface per parcel," that's not a cosmetic improvement.

We're not done with field visits. Boundary disputes and shading questions still need a person standing on the lot. But the baseline layer doing the heavy lifting now gets refreshed on a cycle built for a billing roll, not an engineering survey, and that's the piece that was broken before. If your own roll is still running on whatever the last flyover caught, Stormwater Fee is worth a look before your next appeal season starts. Pull a current layer for one problem parcel and see for yourself what a shorter refresh cycle actually buys you at the appeal table.

← Back to the blog