Forest Water Yield · Guide

Forest water yield
A complete guide to the simplified evaluation method

The Forestry Agency of Japan released a new standard simplified method in March 2026 that lets companies, governments, and citizen groups quantify forest water yield with public data alone. We walk through the formulas, the required inputs, and a Himi City worked example — and outline the same approach for India.

Published
2026-05-26
Aligned with
Forestry Agency of Japan Simplified Evaluation Method Ver.1.0 (March 2026)
Target reader
Sustainability teams · local governments · forest cooperatives · researchers

1. What forest water yield means

Forest water yield is the portion of rainfall that a forest captures in soil, releases slowly as baseflow into rivers, and ultimately delivers downstream as a reliable water supply. It is the quantitative form of the "forest as natural reservoir" idea.

Rain that falls on a forest takes one of three routes: (a) it evaporates from the canopy, (b) it runs off the surface directly, or (c) it infiltrates into the soil and emerges over weeks to months as groundwater or baseflow. The third path is water yield. It mitigates floods and sustains rivers through dry seasons — the part of a forest that literally waters downstream cities.

Forest water balance: precipitation, evapotranspiration, direct runoff, water yield
Figure 1 · Forest water balance. Precipitation minus evapotranspiration and direct runoff is water yield.

2. Why measuring water yield matters now

Water yield was historically treated as a qualitative property of forests — important but hard to count. That changed in the 2020s for several reasons:

Against that backdrop, the Forestry Agency of Japan released the Simplified Evaluation Method for Forest Water Yield, Ver.1.0 in March 2026, with bilingual documentation and a downloadable spreadsheet. The method was designed so anyone could compute water yield with publicly available data.

3. How the simplified evaluation method works

The method is a standardized packaging of the textbook water-balance approach. The headline formula is simple:

3.1 The headline formula

Water yield = Precipitation − Evapotranspiration − Direct runoff

The simplification is not in this equation — that part is decades old. It is in the standardized way to fill each term using free public data.

3.2 Required inputs

InputOpen data sourceResolution
PrecipitationJMA AMeDAS / NASA POWER / JMA mesh climatology10-min to annual
Temperature (for ET)JMA AMeDAS / NASA POWER10-min to annual
GeologyAIST Seamless Geological Map / GSI Japan20 m – 1 km mesh
Forest maskKSJ A13 / Forestry Agency ledger / Sentinel-2 NDVI10–25 m mesh
Stand info (species, age)Forest ledger / airborne LiDARPer stand

All sources are free and open. For India deployments, the equivalents are: IMD (precipitation, temperature), NASA POWER (climate backup), Geological Survey of India / Bhukosh (geology), FSI / Sentinel-2 (forest), and FSI ISFR (stand info).

Processing flow: climate, geology, forest data → water-balance → water yield
Figure 2 · The simplified method as implemented by morimieru.

3.3 Calculation steps

  1. Precipitation. Pull 5–10 years of annual precipitation for the area and average
  2. Evapotranspiration. Estimate potential ET from temperature (Thornthwaite or Penman) and multiply by a land-cover coefficient
  3. Direct runoff rate. Apply geology-driven coefficients (Tertiary sedimentary rocks yield higher direct runoff; volcanic rocks lower)
  4. Forest mask. Classify pixels with NDVI ≥ 0.5 as forest (or use the national ledger / KSJ A13)
  5. Spatial aggregation. Sum the per-pixel balance across the forest mask

4. Worked example: Himi City, Toyama, Japan

morimieru applied the method to an AOI of 16 km × 14 km around Himi City (Toyama Prefecture), covering 22,400 ha of mixed land cover. The result was ~119 million tonnes/year of water yield, or 961 mm/year over the forest area.

TermValueSource
Annual precipitation2,331 mm/yrNASA POWER 2020-2024 average
Evapotranspiration837 mm/yrThornthwaite method
Direct runoff533 mm/yrTertiary geology baseline
Water yield961 mm/yr41.2% of precipitation
Forest area12,398 haSentinel-2 NDVI mask
Cross-check vs municipal record92%vs Himi's reported 13,486 ha

119 million tonnes is roughly the annual domestic water demand of 1.6 million households. The full report is at Himi water yield report (JP).

Spatial map of water yield over Himi: blue intensity = yield density
Figure 3 · Spatial distribution of water yield across the Himi AOI (darker blue = higher yield).

5. Three ways to run the method on your own area

A. The Forestry Agency Excel tool (simplest)

The official spreadsheet released in March 2026 lets you enter precipitation, temperature, geology composition, and forest area for a target area and outputs the standard estimate. Best for 1 ha to one stand.

B. morimieru's web interface (free, end-to-end)

Enter an address or stand ID. We auto-fetch the satellite and standard datasets and produce a report. Best for 100 ha to many tens of thousands of ha, suitable for local governments, cooperatives, and corporate ESG teams.

C. Self-implementation (research / large-scale)

Python with rasterio + pystac-client + the public datasets gives full reproducibility. We plan to open-source the morimieru pipeline. Best for nation-scale batch processing.

6. How to read the numbers

7. Application: TNFD, CSR, and carbon credits

Forest water yield used to fall outside formal carbon credit scope. With TNFD now requiring disclosure of corporate dependency on water provisioning, water yield has become a headline indicator for companies reporting their upstream-forest exposure.

The morimieru watershed-matching feature takes an HQ or factory address, identifies the upstream forests, and computes their water yield in one workflow — directly executing the Locate and Evaluate steps of the TNFD LEAP approach. See the TNFD forest disclosure playbook for the disclosure-side framing.

8. India-specific notes

The method transfers directly. The substitutions:

The Western Ghats portion of Karnataka and Kerala typically yields 900–1,800 mm/yr of water — comparable to Japan's highest-yield basins, with much greater seasonal concentration due to the southwest monsoon. See the India pilot article for the full mapping.

Related

9. Frequently asked questions

Q1. Is water yield the same as storage / reservoir capacity?
Related but distinct. Storage capacity is a stock (mm of soil water at a moment); water yield is a flow (tonnes per year). The simplified method addresses the flow.
Q2. How does the simplified method compare with SWAT / MIKE-SHE etc.?
SWAT and MIKE-SHE are precision hydrological models requiring expert calibration and weeks-to-months of setup. The simplified method packages a coarser version that anyone can run with public data — lower precision, dramatically lower operational cost.
Q3. Can it quantify management effect (thinning, harvest)?
Yes. Sentinel-2 NDVI differencing before and after management gives the change in effective forest area and stand vigour. Re-running the simplified method yields a quantified effect. See Himi monitoring report (JP).
Q4. Is it usable for local-government accountability?
Yes. Since the method is a publicly standardised approach from a national agency, outputs can be cited as "computed per the official Ver.1.0 method", which carries weight in council presentations and constituent briefings.
Q5. Is morimieru free?
Yes — fully free as public-good infrastructure. PDF reports can be downloaded for any commercial or non-commercial purpose.

10. References & sources

Last updated 2026-05-26. Based on public materials from the Forestry Agency of Japan, organized by morimieru.