A flood claim is a precise document. It names a date, an address, a water depth, and a dollar figure. It never names the wet ground upstream that used to hold that water before it reached the house.
For an underwriter, a flood is an event. For a basin, a flood is a condition — water arriving faster than the land can take it. Between the two sits an acre that used to be wet and isn't anymore, and nobody has priced its absence.
the claim is a receipt for the missing acre
Riverine flooding is a timing problem before it is a volume problem. Rain falls across a watershed, and the question is how much of it arrives at the same place at the same hour. Wetlands change that arithmetic three ways: saturated soils and shallow depressions store water on the land, dense vegetation slows overland flow, and a floodplain still connected to its river lets a rising stage spread sideways instead of stacking up and running downstream.
A wetland is the floodplain that still works — marsh, swamp, wet meadow, prairie pothole, playa. Potholes and playas hold water where it falls; floodplain wetlands hold it where it gathers — same job, different position in the basin. Drain it, fill it, or cut it off behind a berm, and the water does not disappear. It arrives somewhere else, sooner, with a claim attached.
Nobody built that drainage to cause floods. Most of it was built to make land farmable, buildable, or passable, one reasonable decision at a time, often a century before anyone was writing flood policies on the parcel downstream. The ditch is a system, not a character flaw. But the water keeps its own books, and those books are now being read.
what the claims modeling shows
In June 2026, Nature Water published work by Jesse Gourevitch and Adam Gold of the Environmental Defense Fund with Helena Garcia of the University of North Carolina at Chapel Hill. The authors matched National Flood Insurance Program residential claim payments from 1985 to 2023 against annually observed wetland-area change in upstream subwatersheds, controlling for heavy rainfall and for growth in impervious surface.
What the model found:
| finding | figure |
|---|---|
| Effect per hectare (about 2.5 acres) of upstream wetland lost | Individual claim amounts rise 0.01–0.03% |
| Added residential NFIP claim payments attributed to wetland loss since 1985 | $10.1 billion, about 9% of riverine claim payments |
| Where the added cost concentrates | Houston metro, southeastern Louisiana, coastal Florida |
| Share of average annual U.S. flood losses the NFIP actually insures | About 30% |
| Total cost of historical wetland loss once uninsured and privately insured damage is counted | Could exceed $33 billion |
Two things have to be said plainly about those numbers.
First, they are modeled estimates from a quasi-experimental panel of repeat-claim properties, with rainfall and impervious-surface controls — strong for what it is, still a model, not a gauge. The per-hectare effect is small, and honest hydrologists will argue about magnitude, lag, and which flood types transfer.
Second, $10.1 billion is not what American wetlands are worth. It is one avoided-damage line, for one flood type (riverine), inside one program (the NFIP), over one window (since 1985), for one property class (residential). Anyone who tells you wetlands are worth $10 billion has read the headline, not the paper.
The paper is more careful than its headline. It also reports — still as a modeled association — that wetland loss raised flood costs more in lower-income and non-white communities. It also reports that in 16% of U.S. subwatersheds, the marginal flood-mitigation benefit of keeping wetlands exceeds the marginal cost of conserving that land. That is a smaller number than an advocate would want, and a far more useful one: it tells a state floodplain program, a levee district, or a flood underwriter where the trade already pencils on flood benefit alone — before anyone counts water quality, habitat, or recharge.
flood wall, buyout, or floodplain
Three responses to the same water. They are not interchangeable, and the difference that matters is what you hold when the next storm comes.
| response | what you build or buy | what happens to the water | what you hold afterward |
|---|---|---|---|
| flood wall or levee | An engineered structure sized to a design storm | Confined and moved downstream faster | A depreciating asset with a maintenance obligation, a design limit, and a residual-risk problem behind it |
| buyout | Public purchase and removal of a repeatedly flooded structure | Still arrives; there is no longer a building in the way | Vacant land, an ended repetitive-loss cycle, and a smaller tax base |
| wet floodplain | Ground kept wet, or reconnected to its river, and held that way — after land acquisition or an easement, plus O&M, and the same tax-base question as a buyout | Spread, stored, slowed, released late | A living stock that keeps producing storage, filtration, and habitat while it does the job |
All three are legitimate. Levees protect cities that cannot be moved. Buyouts end repetitive-loss cycles that were never going to resolve any other way, and they are often the most humane money in the whole program. The floodplain option is the only one of the three that appreciates as it matures, and the only one that was already there before anyone paid for it.
the value of wetlands, stated honestly
Ask what the value of wetlands is and you usually get a per-acre dollar figure. That answer is doing something subtly wrong even when the arithmetic is right.
A wet acre is not valuable because a claims model can attribute damage to its absence. It is valuable because it is continuously doing work: holding water, releasing it late, filtering nitrogen and sediment, recharging shallow groundwater, growing the nursery habitat that fisheries downstream depend on. Avoided flood damage is one flow among many, and it is the one that happens to leave a paper trail in an insurance program.
The value of wetlands is not a price per acre. It is a set of services that keep arriving as long as the ground stays wet — and a bill that arrives downstream when it doesn't.
We do put numbers on condition. That is how capital gets routed to a specific place instead of to a general feeling. But the price is a bridge, never the claim that a dollar figure is the worth of a living floodplain. If the number ever starts standing in for the wetland, the number is wrong.
For the restoration-economics version of this — what a wet acre out-earns and why — see the most valuable acre in america is a wet one.
a floodplain being put back: east fork lewis river
Three river miles of the East Fork Lewis River in Clark County, Washington, upstream of La Center, spent decades as a single confined channel running past nine abandoned gravel pits. In 1995 and 1996, floods breached the levee between the river and the pits; the river shifted south through the mine ponds and abandoned a salmon spawning reach it had used for centuries.
The Lower Columbia Estuary Partnership's reconnection project, in construction from April 2025 through October 2026, is removing berms and levees, has filled four of the nine abandoned pits, is regrading roughly 300 acres of floodplain, and is rebuilding a multi-channel system so the river can spread at high flow again. Funding came from the Washington Department of Ecology's Floodplains by Design program, the state Salmon Recovery Funding Board, and NOAA, with revegetation continuing for years after the equipment leaves.
This is what buying back storage looks like: two construction seasons, heavy equipment, and a public funding stack assembled across agencies (Floodplains by Design, the Salmon Recovery Funding Board, NOAA) on a roughly $23.5 million project — to restore a function the reach performed for free until someone needed the gravel.
Keeping a floodplain wet is cheaper than reconnecting one. That asymmetry, not the poetry, is the argument.
fund the acre that still holds the water
Here is the part worth sitting with. The beneficiaries of a working floodplain are already paying for its absence, just late and through other line items:
- NFIP claim payments, and the program debt and reauthorization fights that follow them
- Levee and flood-wall bonds, plus the maintenance that never ends
- Dredging and channel work where the sediment now lands
- Drinking-water treatment upgrades for the nitrogen and sediment nobody filtered upstream
- Drainage-district assessments, road washouts, and post-event disaster appropriations
Every one of those is a payment for a missing wetland, made after the water arrives. None of them buys a wet acre.
ensurance is the other end of that sequence: funding the wet condition of a named place now, while it is still producing, rather than compensating for the loss later. Two instruments carry it — a coin funds protection protocol-wide and indirectly, a certificate funds one named natural asset directly. Neither instrument is the wetland. The wetland is the wetland. The instrument is how the people downstream pay for keeping it that way.
Our own doors here are named and small: inland-wetlands.ensurance is the live agent for this stock, with modest volumes and real onchain history. The wet acre is usually someone's private ground. Today an easement is what pays a landowner to keep it wet; who pays to keep the floodplain wet is the convert on that bill.
If you want the same physics framed as an investment arbitrage rather than a claims problem, that is a different post with a different keyword and the same water: flood mitigation: the hidden arbitrage in natural infrastructure.
frequently asked questions
how do wetlands reduce flooding?
Wetlands store water in saturated soils and surface depressions, slow overland flow with dense vegetation, and give a rising river somewhere to spread. The net effect is to lower and delay the downstream flood peak, which matters most in riverine flooding across small and mid-sized watersheds. They are not a universal substitute for engineered protection: storm surge, extreme urban downpours, and very large basin-wide events can overwhelm wetland storage, and a wetland already saturated before the storm has less capacity left.
why does wetland loss raise flood insurance claims?
Because the water still shows up. When upstream wetlands are drained, filled, or disconnected from the river, the same rainfall reaches downstream properties faster and higher, so more structures get wet and each claim runs larger. Gourevitch, Gold, and Garcia estimate that each hectare of upstream wetland lost raises individual residential claim amounts by 0.01% to 0.03% — small per hectare, and material once summed across decades of national wetland loss: $10.1 billion in additional NFIP residential payments since 1985, roughly 9% of riverine claims.
what is the value of wetlands for flood control?
It depends entirely on where the wetland sits and who is downstream of it, which is why a single per-acre figure is the wrong unit. The strongest published U.S. estimate is the 2026 Nature Water analysis: $10.1 billion in added NFIP residential claims attributable to wetland loss since 1985, and potentially more than $33 billion once uninsured and privately insured losses are included, since the NFIP covers only about 30% of average annual flood losses. The same study finds the flood-mitigation benefit alone already outweighs conservation cost in 16% of U.S. subwatersheds — and flood control is only one of the services a wet acre provides.
what to do with this
If you underwrite flood risk: treat upstream wetland extent as an exposure variable, not scenery. The study's subwatershed-level marginal values are published for exactly that use.
If you manage a floodplain or a public works budget: price the wet acre against the wall and the buyout before you price the wall against the buyout.
If you are new to the stock itself: start with the definition, then come back — what wetlands actually are.