One and a Half Percent, a Story of a Flood Foretold
How developer bias captured Shorewood City Hall — and what it costs the people who live here.
Part of Anatomy of a Captured City. Start here for the short version. This one is about water, and it is the one where somebody gets hurt.
There is a house near the Watten Ponds development whose basement has already taken water. There is a yard not far away that floods in a hard rain. The neighbor closest to that wetland says the water already comes up higher than the level the developer's own flood model treats as normal.
Those are the people on the other side of this one. Not the lake. Not the trees. Houses, with people in them, downhill.
The rule, and the number
Shorewood's code says a new development cannot send more water off the property than the land sends today. Not a little more. Any more. Three storms get checked.
Read the bottom row left to right. On the small storm there is real cushion. On the hundred-year storm — the one that actually puts water in basements — the cushion is one and a half percent.
The entire difference between this project complying with the City's flood rules and not complying is about a quarter of a cubic foot per second.
And here is the thing to hold onto: if that number is wrong, the water doesn't vanish. It goes to the nearest low ground. Which is a yard, and then a basement.
In other words, this is about protecting residents from developments that ruin their houses. It’s not check the box, it’s people’s lives. But a flood model is not a measurement. It is an estimate built out of assumptions, subject to all models to the rules of GIGO (garbage in, garbage out), and the state's own stormwater manual says these models are not exact simulations of the processes occurring in nature, and the way to check one is against actual monitoring data or observations. Which means the model can be wrong, and if it’s wrong, houses may flood.
Nobody at the City checked it against anything.
The Code told them exactly what to ask. This is the part that turns it from sloppiness into something else.
Shorewood's Code does not stop at the runoff numbers. Section 1202.20 requires the City to decide whether a subdivision is premature, and one of the grounds is inadequate drainage — which exists if, among other things, the subdivision will cause pollution of water sources or damage from erosion and siltation on downhill or downstream land, or harmful and irreparable damage from erosion and siltation on downstream land.
Read who that section is written for. Downhill. Downstream. The neighbors.
And the Code does not leave it vague. It lists the factors to weigh. Here is each one, and what the file shows.
Average rainfall for the area. The model uses 7.32 inches for the hundred-year storm, from the national rainfall atlas published in 2013. The figure before it, used here for decades, was around six inches — when the data got refreshed, the hundred-year storm grew about twenty percent, and the federal government is revising it again now. Nobody used a wrong number. Nobody asked what it means that the last correction to this input was more than ten times the size of this project's entire margin. No record.
The relation of the land to floodplains. The report promises no increase — 0.00 ft — in the hundred-year flood level of the channel. No record of anyone testing that.
The nature of soils and subsoils and their ability to adequately support surface water runoff. Soil type was picked off a desktop map. The report says: Soil borings will be completed prior to applying for building permits. The digging happens after the approval that depended on it. No record.
The slope of the land and its effect on effluents. Downhill, toward wetlands, toward houses that have already flooded. No record.
The presence of streams. The channel running between the two wetlands. No record.
Five factors named in the Code. Nothing in the record indicating that the City addressed any of them.
And this is not some dusty provision nobody remembered. The Council adopted it on July 14, 2025 — six months before this report was filed. It has been in force every day since, first as Ordinance 617 and now as Section 1202.20 of the Code.
Somebody did look at this applicant's work
Just not the City.
The Minnehaha Creek Watershed District hired an outside engineering firm to review the same applicant's modeling on a related permit. On January 27 their reviewer wrote: The applicant's current analysis for high water level impacts and proposed floodplain fill appears to be based on modeling which does not consider the upstream and downstream hydrologic conditions of the site (i.e. wetlands, drainage area, etc.).
A professional reviewer, saying the modeling left out the wetlands and the land draining into the site. That is the same modeling behind the 0.00 ft promise above.
One agency paid somebody to look and sent it back. One didn't bother to look at all.
Nobody even read the report
Here is the detail that needs no engineering at all. Every page of the model output in that report is headed Prepared by {enter your company name here}. The template placeholder. Never filled in. One hundred and two pages, handed to the City, accepted into the record, and nobody — not the consultant, not the applicant, not the City — ever noticed.
What is the report actually for?
There are two ways to treat a stormwater report. One: it's a document the Code requires, and the job is to get a finished one into the file so the project can move. Paper with ink on it. Two: it's the tool the City uses to find out whether this project is going to put water in somebody's house, and to fix it before that happens.
Those two produce the identical stack of paper and completely different questions. The first asks: is there a report, does it say compliant, is it signed. The second asks the five questions the Code actually lists.
Read the file and there's no doubt which one happened. The report arrived. It said compliant. It went in the drawer.
Nobody decided to ignore the flood risk. That is not how this works. The question just never got asked, because the paperwork was complete — and when the paperwork is the point, one and a half percent is a pass, exactly like fifty percent is a pass.
Who pays if it's wrong
With the trees, being wrong costs the city trees. Here, being wrong costs a specific family a specific basement.
And there is no neutral mistake available. Every assumption that broke the developer's way breaks against the people below him. Optimistic about the soil, optimistic about the rain, optimistic about how much a cleared lot absorbs — the water those assumptions didn't account for still falls. It just lands somewhere, and the somewhere is downhill.
Helping the applicant here is not a favor that costs nobody. It is a transfer. The developer gets a buildable lot now; the neighbor gets the risk, forever, with no way to prove whose it was.
A developer who is wrong about this sells the houses and leaves. The family mopping the floor is still here in twenty years.
What we asked for, and what we got
We never asked the City to reject the project. We asked it to do the analysis its own Code requires before accepting the report: test the assumptions and see how far they can move before the project fails, do the soil borings before approval instead of after, and have one independent expert in water look at it.
All three can be required at the developer's expense. None of the three happened.
A one and a half percent margin, on the biggest storm, resting on numbers nobody checked, uphill of houses that already flood — and not a single question in the file.
The Code gave the City the questions. Asking them was discretionary. A captured department does not use discretionary tools. It completes the file.
The stormwater management report, the watershed district correspondence and the project file are public records. We live in Shorewood.
How this fits with the rest
This post connects the series in a sixth direction: the same discretionary gap appears in stormwater review. Read it alongside the tree-policy analysis in Eleven Tools, Not One Used - a Tree Preservation Policy that Doesn't Preserve Trees; the scope problem in The Amazing Shrinking Environmental Review; the appellate omissions in Letting the Clock Run Out; and the explanation and proposed fixes in What Capture Is, and Why You Pay and How It Happens, and What to Do About It. A Eulogy for Arthur shows the human cost of the tree decision.
Public watchdog or developer lapdog? Capture and Why You Pay — The name for what happens when a regulator starts seeing things the way the people it regulates do.
The Amazing Shrinking Environmental Review — An environmental review ordered in February covered one lot by September. Nobody voted. A chronology from the City's own records.
Eleven Tools, Not One Used - a Tree Preservation Policy that Doesn't Preserve Trees — The City's tree policy gave it eleven powers over how these woods come down. There is no record it used one of them.
One and a Half Percent, a Story of a Flood Foretold — The stormwater report left only a 1.5% margin on the largest storm, uphill of homes that have already flooded, without the City's file showing that the assumptions were checked.
Letting the Clock Run Out — The City is defending its own decision on appeal. Read what its brief leaves out.
How It Happens, and What to Do About It — Why this happens to normal people, and the two small things we're asking the City Council to do.
Series overview · What Capture Is, and Why You Pay · The Amazing Shrinking Environmental Review · Eleven Tools, Not One Used - a Tree Preservation Policy that Doesn't Preserve Trees · Letting the Clock Run Out · How It Happens, and What to Do About It · A Eulogy for Arthur · If Somebody Tells You Agency Capture Isn't a Real Thing
Comments