AI driven forest survey system reduces costs and speeds up mapping in Nagano

On November 13^th, the Nikkei Shimbun reported that Nagano Prefecture’s Kiso Town has launched a first-of-its-kind pilot program this autumn to carry out forest boundary clarification and cadastral surveys at the same time - a pairing long treated as separate processes in Japan.

By harnessing AI that analyzes aerial imagery to estimate land boundaries with high precision, the town aims to produce legally valid cadastral maps while cutting survey time in half and reducing costs by 30 - 50%.

Branded as the “Kiso Model,” the initiative seeks to accelerate Japan’s chronically delayed forestland registration and strengthen the foundation for revitalizing local forestry.

How the Kiso Model Works

Kiso Town is developing the model with Shinshu University and Precision Forestry Measurement, a university - affiliated startup based in Ina City. The partners have created technology that uses drone - captured images to automatically identify individual trees - species, position, and trunk diameter - with a level of precision that previously required extensive fieldwork.

Forest boundary estimation map created from drone imagery courtesy of Masato Kato at Shinshu University via the Nikkei Shimbun.

AI then estimates forest boundaries by combining these measurements with topographic features such as ridgelines, streambeds, and service roads, as well as existing public land records.

Masato Kato, a specially appointed professor at Shinshu University, explains that shifts in vegetation patterns - for example, between planted hinoki or larch and naturally occurring broadleaf trees - often signal where a boundary lies. These datasets are compiled into a geographic information system (GIS) map called the “Forest Boundary Estimation Map.”

Reaching Agreement Without On-Site Inspections

A defining feature of the Kiso Model is its desk-based approach. Stakeholders gather at community facilities to review boundaries displayed on monitors, referencing the estimation map and on-site photos.

Elderly residents with decades of forestry experience take part as advisers, contributing knowledge of past planting histories.

“Because the boundary estimation map is already generated using precise survey data, once the boundary is settled, we can immediately reflect the information and produce the cadastral map,” says Masaru Yamada, head of Kiso Town’s Construction and Forestry Division.

This eliminates the need for labor-intensive on-site attendance. For the 80 hectares scheduled for mapping this fiscal year, Kiso Town expects the project period to shrink to 18 months - half the original plan - with costs lowered by 30 - 50%.

Traditionally, surveyors had to navigate steep terrain to identify boundaries, and landowners were required to be physically present, creating significant delays.

Why the Dual Process Is Significant

Boundary clarification and cadastral surveys have historically proceeded separately because their roles differ. Boundary clarification is a practical, simplified process for forest management and land use.

A cadastral survey, by contrast, is a legally mandated, high-precision measurement used to create cadastral maps - official records that define property boundaries for land registration, taxation, and real estate transactions.

Merging these two processes has long been difficult due to differing technical and legal standards. But with advances in remote-sensing technologies such as aerial laser scanning and photogrammetry, the national government in 2022 issued guidelines allowing boundary clarification results to be incorporated into formal cadastral surveys.

With support from Kiso Town and Shinshu University, the Forestry Agency selected their initiative for its “Forest Consolidation Model Area Pilot Program,” enabling both processes to advance simultaneously.

A National Challenge: Slow Progress and Outdated Maps

Japan’s forest cadastral surveys remain severely delayed, with national progress stuck at around 47% as of the end of FY2023. Many public land maps still originate from the Meiji-era land tax reform and contain inaccurate boundaries and area measurements.

These uncertainties limit forestry development, forcing loggers to avoid unclear borders and leaving unharvestable strips of woodland that hinder consolidation and large-scale operations.

Kiso Town Aims to Build a Replicable Model

Kiso Town, where roughly 90% of land is forested, is pursuing its “Wood Industry Development Project” to promote the area as a major forestry hub. With postwar Japanese larch plantations reaching harvest age, the Kiso Model is expected to provide a critical foundation for strengthening local forestry activity.

Amid nationwide issues such as aging populations, depopulation, and the rise of absentee forestland owners due to inheritance-driven subdivisions, Yamada emphasizes the model’s wider relevance. “We want to establish this as a model that even small municipalities like ours - with a population of just under 10,000 - can put to practical use.”

Source:

Nikkei Shinbun (Japanese only; paywalled)