Blindly trusting data will leave you all at sea

Zillow recently published a report on how many houses are at risk from sea level rise around the US. It’s a good idea, but looking closely at where I live reveals some… issues in their analysis. Here’s a screenshot from the Seattle Times’ article with the local angle on the report:

Homes that Zillow thinks are at risk of flooding from sea level rise in Seattle

If you know Seattle, you can probably spot the first problem. For those who don’t: the Eastern shore, and the cluster N of the word “Seattle” on that map are all lakefronts, separated from the sea by the rather large Ballard Locks. The local article’s been updated to mention that, but Zillow’s downloadable data hasn’t. And even the local article misses a detail: if sea levels were to rise by a foot more than the analysis assumed, the highest tides would still only be 4¾ inches over the lock gates—a problem for some houses for sure, but nothing like the 7 feet of flooding we’d have on our sea side, and much less than the amount the lake is already allowed to rise and fall by.

But there’s another problem, unrelated to quirks of Seattle’s Herculean engineering, and much more worrying for the reliability of this analysis. Let’s zoom in on an area that is on the sea and surely will suffer under 6 levels of sea level rise:

Zillow's sea level rise map, zoomed in to part of West Seattle

What’s going on here? Are they trying to tell us that we have a row of houses already underwater? And what’s that partial second line? A look at the King County Parcel Viewer gives a hint:

King County Parcel Viewer zoomed in to a section of West Seattle's shoreline

The red outlines are individual property parcels, and the green dots are the much more plausible locations where the County’s GIS department thinks the main building on each property is. Take a look at the first few properties going South from the road junction at the top of both screenshots. We see two wide parcels (4701 & 4705 Beach Drive SW), then two narrower ones (4707 & 4711) with publicly-owned tidelands West of them. Looking back at the screenshot from Zillow’s data, they seem to have assumed that every house is right in the middle of its parcel. Oops.

It is definitely possible that all the properties on the seaward side of Beach Drive SW are in danger. That is a low-lying area, and even if the houses aren’t immediately flooded at high tide they’ll suffer from beach erosion. But it’s also hard to take an analysis seriously that starts with treating the house as though it’s already in the sea. And there’s more! The Seattle Times report said houseboats were excluded, but here’s a zoom in on one of the bigger houseboat areas on Lake Union:

Houseboats that Zillow thinks will flood

Most of the houseboats are excluded, but there are still a few dots there. Houseboats on Lake Union mostly don’t have individual property parcels—the parcels reflect ownership of the docks—and looking back at the parcel viewer gives me the suspicion that most, albeit not all, of those parcels are in fact included.

It’s hard to tell how much these kinds of glitches really affect the overall picture that Zillow paints with their report, because they don’t say much about methods or assumptions. It may well be that they make a much smaller difference in the areas that will be harder hit by this problem than Seattle, because where the ground is flatter the problem extends far inland of the current shoreline. But looking at where I know, a large proportion of the properties they say are at risk seem to be affected by misunderstandings of the data. Sadder still, Zillow is based in Seattle, so someone working on this should have noticed the same things that I immediately went to check. There’s a broader lesson here, about the dangers of trusting data files without carefully checking that they really say what we think they do.