---
title: the cost-benefit case for the ground upstream
canonical_url: https://ensurance.app/guide/the-cost-benefit-case-for-the-ground-upstream
markdown_url: https://ensurance.app/guide/the-cost-benefit-case-for-the-ground-upstream.md
subtitle: "a worked example on a hypothetical water utility: the loss, the land, the counterfactual, and a payment the budget can book"
category: how-to
---

# the cost-benefit case for the ground upstream

*a worked example on a hypothetical water utility: the loss, the land, the counterfactual, and a payment the budget can book*

**What is climate adaptation cost benefit analysis?** It compares the present value of the losses an adaptation measure avoids, plus its other benefits, with the full cost of building and keeping that measure, both measured against a no-action baseline on the same events. If the benefits are larger, and stay larger on cautious assumptions, the measure has earned a budget line.

This analysis is for ground, not a levee or a pump station: the forest that holds the slopes above a reservoir, the riparian corridor that catches sediment before the intake. They were setting the hazard long before anyone scored it, whether or not anyone pays for them. [**Ensurance**](/manual/ensurance?from=guide) is how the people who depend on that ground can pay for its keeping. It is not the forest, and it does not stop a storm.

Below is the method in seven steps, yours to use whether or not you talk to us, then a worked case on a **hypothetical mid-size water utility**. Every number in it is illustrative; no real utility, town or river is meant.

## step 1: name the loss and who pays it

Start with a bill, not a hazard map. Below a burned watershed the utility's bill is usually treatment: more coagulant when turbidity spikes, sludge to haul, a settling basin to dredge, an intake shut while the river runs dark. Count bills you already pay ([what ensurance saves you](/guide/what-ensurance-saves-you?from=guide)). A public figure, such as the Federal Emergency Management Agency's (FEMA) [tract-level expected annual loss for your place](/place?from=guide), is a lead; your invoices are the case.

## step 2: find the ground that moves it, and who owns it

Trace the loss upstream: burned slopes that shed debris in rain, cut banks feeding the channel, overstocked stands ready for the next fire. Name the owner of each parcel: a national forest, a ranch, a land trust, the utility itself. Work on ground you do not own needs that owner as a partner.

Name the mechanism too: debris flows, erosion and slope failures need different models, and US Geological Survey post-fire debris-flow assessments cover one link, not what treatment changes.

## step 3: run the same events twice

Set the **no-action baseline**: what the utility loses if nothing is done upstream, including the forest getting worse. Then the **with-project case**: the same storms, fires and plant, with the work in place. The difference is the avoided loss.

The US Army Corps of Engineers' HEC-FDA flood damage software is the public example.

Change only the hazard or vulnerability term the work moves; do not assume a cut because the forest looks healthier. We do not run these models. A modeler you hire does, and the result is an estimate under stated assumptions, not proof. For ground a town already has, see [price what the creek already does](/guide/price-what-the-creek-already-does?from=guide).

## step 4: count the right losses

**Expected annual loss** (EAL) averages good and bad years into one figure. Split it into bands first. Work on the ground mostly moves frequent and mid-size losses: the turbidity storm most years, the post-fire storm that fills the basin once a decade. It does little against the extreme year, when a record fire meets a record storm; keep insurance and reserves for that year and leave it out of this case ([the average year is not the year that breaks you](/guide/the-average-year-is-not-the-year-that-breaks-you?from=guide)).

That post's table holds the modeled studies. This case counts only averages by band, on purpose: a saving on a deductible, a retention or the price of cover would add to it once a carrier quotes it, so the case is a floor.

## step 5: cost the work for its life

A pump is costed for its life. Do the same for the ground: upfront thinning and riparian repair, then re-treatment as the understory grows back, monitoring, crews, repair after a storm. A case priced on year one fails in year eight.

## step 6: discount, compare, and take the lower estimate

Put both in **present value** (PV), discounted at the utility's own rate. The **benefit-cost ratio** (BCR) divides the PV of avoided loss by the PV of cost; **net present value** (NPV) subtracts it. [The quote beside the cost of doing nothing](/guide/the-quote-beside-the-cost-of-doing-nothing?from=guide) walks through the arithmetic.

The 2026 *Guide for Adaptation and Resilience Finance* from Standard Chartered and KPMG asks for benefits against a no-action case and lists conservatism among its attribution safeguards. Use the low end of the modeled range, show half the effect and double the cost, and say where the case breaks.

## step 7: turn it into a payment a budget can book

A BCR wins an argument; it does not move money. Spread the life-cycle cost into a yearly figure, then into a unit the budget uses: **per acre of named ground, per year**, on a line beside treatment. The most you should pay is in [what a source is worth paying for](/guide/what-a-source-is-worth-paying-for?from=guide).

## the worked case: a hypothetical utility

A hypothetical mid-size water utility draws from a river below a fire-scarred, eroding watershed: 4,000 acres of named source ground, burned, overstocked and losing its banks. Every figure is illustrative.

| band | what happens | EAL without the work | EAL with the work | avoided a year |
|---|---|---:|---:|---:|
| Frequent | Storm turbidity most years: extra coagulant, sludge hauling | $350,000 | $230,000 | $120,000 |
| Mid-size | A post-fire storm every decade or so: basin dredged, intake shut for days | $450,000 | $210,000 | $240,000 |
| Extreme | A record fire, then a record storm: weeks offline | $200,000 | $170,000 | $30,000, not counted |
| **Counted** | | | | **$360,000** |

The work costs $1,600,000 upfront (about $400 an acre), then $60,000 a year for re-treatment, monitoring and crews, over 20 years at a 4% discount rate. For simplicity, avoided loss and the baseline are held flat from year 1, in today's dollars; a real case would ramp the benefit in and let the baseline worsen, as step 3 asks.

| measure | base case | half the effect | double the cost |
|---|---:|---:|---:|
| PV of avoided loss | $4.89M | $2.45M | $4.89M |
| PV of cost | $2.42M | $2.42M | $4.83M |
| BCR | 2.0 | 1.0 | 1.0 |
| NPV | +$2.48M | +$0.03M | +$0.06M |

The base case cuts the counted bands about 45%, mid-range for those modeled studies, not the low end; under step 6's rule the half-effect column is the conservative case. Break-even is a cut of about 22%.

At half the effect or double the cost, the case only breaks even; both at once, it fails at a BCR of 0.5. That is why others below the same slopes, a county road department or a downstream town, can be asked to pay against their own avoided bills ([split the bill for what lowers it](/guide/split-the-bill-for-what-lowers-it?from=guide)). Insurers so far rarely recognize the work.

Spread over 20 years at 4%, the $2.42 million is about $178,000 a year, **about $44 per acre**.

| budget line | per acre per year | 4,000 acres |
|---|---:|---:|
| Source protection, named upstream ground | $44 | $178,000 |
| Avoided treatment and outage loss (modeled, counted bands) | $90 | $360,000 |

At half the effect, avoided loss falls to $45 an acre, barely above the payment (about 57 cents an acre).

## real benchmarks beside the hypothetical

| case | what it shows | read it as |
|---|---|---|
| Yuba Forest Resilience Bond, California | Pilot (2018), about 15,000 acres; Yuba II (2021), $25M for about 48,000 acres; Yuba Water Agency committed $1.5M, then $6M | Dependents, a water agency among them, help fund the work. No avoided-loss figure published |
| National Institute of Building Sciences, 2019 | Wildland-urban interface fire grants $3 per $1 | Societal benefit, not one payor's saving; rests on 25 grants |
| Gulf Coast nature-based adaptation, 2018 | Average BCR above 3.5; some marsh and oyster reef cases above 7 | Modeled; 7 is the upper end, not the average |

:::johnson
**price the ground the way you price the pump.**

A pump gets a capital line, a maintenance line and a design life. The slopes above the intake can get the same three, beside the treatment budget.

[see what a utility can fund upstream →](/solutions/utilities?from=guide&topic=after-the-score)
:::

## how a utility pays for source water protection, per acre

- **Exposure:** treatment and outage loss below named source ground, in your own invoices.
- **What we deliver:** a source-protection account for the named ground, certificates for its parcels, and proceeds routed to the stewards and crews. Behind it, on the [utilities page](/solutions/utilities?from=guide&topic=after-the-score): watershed and hydrology services; wildfire, flood and erosion resilience; measurement, reporting and verification. Prices are on [services](/services?from=guide&topic=after-the-score).
- **The work it funds:** forest thinning and riparian repair on the named upstream parcels, then re-treatment and monitoring.
- **The unit:** per acre of named source ground, per year, beside the treatment budget line.

Each certificate starts **ensured**, funded year by year. The path runs toward **entrust**, permanent protection; ensured alone does not finish it. Someone fronts the upfront work and is repaid from the yearly line.

The line is named and not yet minted: **source-protection** under [Risk Resilience](/risk-resilience.ensurance?from=guide). You can [co-create it](/specific/create?mode=suggest&agent=risk-resilience.ensurance&name=source-protection&from=guide), and a person reviews it. Or [bring us your bill and the ground you think sets it](/contact?from=guide&topic=after-the-score).

## what this case cannot claim

**Modeled, not observed.** Every avoided-loss figure here is model output, and specific to its place and mechanism; a figure from another forest does not transfer to your slopes.

**Premiums rarely move.** A 2025 wildfire policy at Tahoe Donner was priced on forest treatment: 39% cheaper, with an 89% lower deductible, by the sponsors' figures, on homeowners-association land. It is the exception. Resources for the Future found retrofit costs "orders of magnitude greater" than the insurance savings. Leave a premium cut out unless a carrier has quoted it.

**No case of ours yet.** We have not run a completed case end to end. Ensurance does not run catastrophe models, compute a tail or hold tail risk.

## frequently asked questions

### what is climate adaptation cost benefit analysis?

Comparing the present value of avoided losses and other benefits with an adaptation measure's life-cycle cost, both against a no-action baseline on the same events.

### how do you value nature-based adaptation?

Like gray adaptation: model the loss without and with the work on the same events, cost the work for its life, discount both, and bring the ground's owner in as a partner.

### how can a water utility pay for source water protection?

As a yearly payment per acre of named source ground, booked beside treatment, with other dependents paying their own shares.

## sources

[US Army Corps of Engineers, HEC-FDA](https://www.hec.usace.army.mil/software/hec-fda/) — Flood damage software that computes expected annual damages and compares without-project conditions with alternative plans

[US Geological Survey, post-fire debris-flow hazard assessments](https://www.usgs.gov/programs/landslide-hazards/science/postfire-debris-flow-hazard-assessment-data) — Likelihood and volume of debris flows under rainfall scenarios

[Standard Chartered and KPMG, *Guide for Adaptation and Resilience Finance*, 2026](https://www.sc.com/en/uploads/sites/66/content/docs/Adaptation-Finance-Reports.pdf) — Benefits against a no-action baseline; conservatism among the attribution safeguards

[World Resources Institute, Yuba II Forest Resilience Bond](https://www.wri.org/news/release-new-forest-resilience-bond-will-finance-25-million-restoration-reduce-wildfire-risk) — Yuba II, $25 million for about 48,000 acres after a 15,000-acre pilot; Yuba Water Agency's $1.5 million and $6 million commitments

[National Institute of Building Sciences, *Natural Hazard Mitigation Saves*, wildfire grants fact sheet, 2019](https://nibs.org/wp-content/uploads/2025/04/ms_v3_grants_fire.pdf) — Wildfire grants save $3 per $1, from 25 grants with data

[Reguero et al., *PLOS ONE* 13(4): e0192132, 2018](https://doi.org/10.1371/journal.pone.0192132) — Gulf Coast nature-based adaptation, average benefit-cost ratio above 3.5

[The Nature Conservancy, Tahoe Donner wildfire resilience insurance, 2025](https://www.nature.org/en-us/newsroom/ca-wildfire-resilience-insurance0/) — Premium 39% lower and deductible 89% lower than without the forest work, by the sponsors' figures

[Resources for the Future, Working Paper 25-30, *From Risk to Reward*, 2025](https://www.rff.org/publications/working-papers/from-risk-to-reward-insurance-discounts-for-wildfire-mitigation/) — Retrofit costs "orders of magnitude greater than the insurance savings"

## the series

1. [the score is the easy half](/guide/the-score-is-the-easy-half?from=guide) — **Read first:** What a climate risk score leaves undecided, and the five questions after it
2. [the average year is not the year that breaks you](/guide/the-average-year-is-not-the-year-that-breaks-you?from=guide) — Expected annual loss is an average; the bad year sets the capital, and living systems move some of it, not all
3. [the bet that isn't in your scenarios](/guide/the-bet-that-isnt-in-your-scenarios?from=guide) — A stress test reports the risk you modeled; the source your site depends on is usually the scenario nobody ran
4. [the watershed isn't in the asset register](/guide/the-watershed-isnt-in-the-asset-register?from=guide) — A climate risk model needs hazards, assets and damage curves; the ground that sets the hazard has no row
5. [waiting is a position](/guide/waiting-is-a-position?from=guide) — When the bad years get worse, the protection is worth more and each year of waiting costs more
6. [the cost-benefit case for the ground upstream](/guide/the-cost-benefit-case-for-the-ground-upstream?from=guide) — A worked example on a hypothetical water utility, from the loss to a payment a budget can book