CAMPA monitoring
a satellite layer for compensatory-afforestation transparency

1. The CAMPA stake, in numbers

CAMPA was created to ensure that when forest land is diverted for non-forest use under the Forest (Conservation) Act, the diverting user-agency pays Net Present Value (NPV) and Compensatory Afforestation (CA) into a dedicated fund. After the 2018 Supreme Court ruling and the 2018 CAF Act + 2018 rules, the corpus was placed under the CAMPA framework with National and State authorities. The intent is straightforward: when a hectare is diverted, an equivalent (or better) hectare gets restored — measurably, accountably, and on time.

The execution has been uneven. Annual CAG audits and the Supreme Court's Central Empowered Committee have repeatedly flagged plantation non-maintenance, delays in plan approval and fund release, and instances of fund use outside the programme's intended scope. State-level audits in Uttarakhand and Odisha, in particular, have been picked up in national press. None of this is a secret — and none of it is something morimieru is uniquely positioned to fix in isolation. But the data layer that closes most of these gaps is, today, freely available from space.

2. The structural monitoring problem

The CAMPA reporting chain currently runs through state forest department officers, divisional inspections, and periodic compendium reports to National CAMPA. Several features of this chain limit what the CAG and CEC can verify externally:

• Plot boundaries are often not in a queryable spatial format. Compensatory plots are recorded by survey numbers and verbal descriptions. Without vector polygons, third parties cannot test claims against ground truth.
• Baseline imagery for "before" condition is rarely archived alongside the approval. When a clearance is granted under the FCA, the precise pre-diversion forest cover at that polygon is not snapshotted in a way that can be later proven.
• Plantation success is verified by site visits, not continuously. A site-visit-based system measures the day of the visit; canopy collapse in month 11 after a month-1 visit goes undetected.
• Cross-state comparability is weak. Each state's CAMPA reports on its own template, often without coordinate-system consistency.

None of these constraints come from negligence — they are artefacts of an inspection-based system designed before pixel-level satellite data was free, weekly, and global. The data environment changed; the monitoring layer hasn't fully caught up.

3. What a satellite monitoring layer adds

A satellite monitoring layer — built on Sentinel-2, Sentinel-1, and HydroBASINS, with cryptographic timestamps on every observation — adds five capabilities that the inspection chain cannot deliver on its own:

Capability	Concrete output
Pre-clearance baseline lock-in	NDVI snapshot of the exact diverted polygon at the date of clearance, cryptographically time-stamped so it cannot be back-dated
Continuous compensatory-plot tracking	Weekly NDVI for every CA plot since planting; alerts on canopy collapse, fire, or stalled growth
Like-for-like comparison	Side-by-side time series of the diverted plot vs the compensatory plot, on the same scale
Watershed-context overlay	Each plot's catchment and downstream basin shown alongside its growth — making the "did we restore in a useful place" question answerable
Tamper-proof audit chain	TPM 2.0 attestation + Merkle hash + RFC 3161 timestamp on every report. CAG / CEC can independently verify dates and content years later.

4. Our stance: complement, not critique

A state DFO's job already involves more inspections than headcount allows. Satellite monitoring multiplies their visibility without adding fieldwork. A CAG auditor's mandate already requires verification — satellite evidence makes a previously expensive check trivial. The technology benefits the people inside the system as much as the people auditing it.

5. A worked example: monitoring a hypothetical CA polygon

Suppose a 50 ha forest patch in district A is approved for diversion in 2023, with a 75 ha compensatory plot allocated in district B. The diverting user-agency deposits NPV + CA into State CAMPA. The state forest department plans the CA plot, plants it in monsoon 2024, and reports survival rates back to National CAMPA. Where does satellite evidence enter?

1. 2023, at clearance: Sentinel-2 NDVI snapshot of the 50 ha diverted polygon at the exact date of FCA approval, frozen in the cryptographic chain. Provides the contested "what was there before" baseline for every future audit.
2. 2024, at planting: Snapshot of the 75 ha CA polygon at the planting season, marking the year-zero condition of the new plantation.
3. 2024–2030, monthly: NDVI tracking of the CA polygon. Year-on-year growth visible; canopy collapse, fire, or stagnation flagged automatically.
4. 2024–onward, watershed view: Both polygons placed on the HydroBASINS Lvl 10 layer. Auditors can see whether the CA plot sits in a watershed that hydrologically substitutes for the diverted patch, or in a distant basin that does not.
5. Anytime, externally: CAG, CEC, civil society researchers, or climate finance entities can pull the report bundle and verify that the values match the time-stamped raw data — no permissioned access required.

This is exactly the pattern morimieru runs for forests in Japan. The same engine, applied at the State CAMPA level, opens a national-scale evidence layer.

6. Stack: what stays, what swaps

Layer	Source (already global)	India substitution
Optical satellite	Sentinel-2 L2A (Copernicus, 10 m)	—
Cloud-penetrating	Sentinel-1 SAR (Copernicus)	—
Climate	NASA POWER	IMD gridded products for finer monsoon detail
DEM	SRTM / Copernicus DEM	ISRO CartoSat where state-released
Watershed boundaries	HydroBASINS Lvl 10 (CC-BY)	India-WRIS finer sub-basins
Forest inventory	—	Forest Survey of India regional volume tables
Land-use baseline	—	ISRO Bhuvan LULC + FSI ISFR
Plot register	—	State CAMPA plot list (where shared) + e-Green Watch where available
Cryptographic chain	TPM 2.0 + Merkle + RFC 3161	—

The architecture is identical to our Japan pilot. The data layer differences are mostly small and well-documented. See our India pilot article and Tier 2 walkthrough.

7. A 90-day state-level pilot scope

Phase	Days	Output
1. Plot ingest	1–10	State CAMPA CA plot list converted to polygon geometry (where survey-number-only, buffer the centroid until ground-truth polygons available)
2. Baseline snapshots	11–25	Sentinel-2 NDVI baseline for each polygon at planting date (or most recent year if older)
3. Time-series build	26–45	Year-by-year NDVI tracks for every CA plot in the pilot batch
4. Watershed overlay	46–60	HydroBASINS catchment per plot; downstream-village layer where required
5. Cryptographic chain	61–75	Per-plot signed report bundles; aggregate Merkle root anchored for the state
6. Public dashboard	76–90	State-CAMPA dashboard (read-only, sortable by district / year / size) — analogue of our existing JP-side reports

We have not yet run a CAMPA pilot. We have run every constituent step for our Japan forest pilots and have documented the India-layer substitutions. The pilot above is what we would propose to a state forest department, a research institution, or a CAG-affiliated reviewer who can bring (a) a plot list, (b) the political space to run a parallel transparency layer, and (c) the willingness to publish the result.

8. Honest constraints

• Plot geometry is the blocker. The pipeline only works once polygons exist. Where CA plot records are still survey-number-based or descriptive, there is a one-time digitisation step.
• NDVI is not species-aware. A mono-plantation can score the same NDVI as a biodiverse forest. The satellite layer can flag presence of canopy; composition still needs ground botanical inventory.
• Smallholder co-implemented plots. Where CA plots are mosaic-style on community lands, single-polygon analysis is too coarse. Pixel-level aggregation helps but does not fully solve this.
• Imagery archive only goes back so far. Sentinel-2 baseline is reliable from late 2015; Landsat extends earlier with coarser resolution. Plots approved before 2015 need a different baseline strategy.

9. Who this is for

• State forest departments wanting an audit-defensible monitoring layer for their own use — no licence cost, no vendor lock-in.
• National CAMPA and the MoEFCC, for cross-state comparability.
• CAG, the Central Empowered Committee, and the Parliamentary Standing Committee on S&T, Environment, Forests & Climate Change for independent verification.
• Research institutes (FRI Dehradun, IIFM Bhopal, IISc CES, IIT Bombay CESE) studying compensatory afforestation effectiveness.
• Civil society and climate journalism tracking implementation.
• Climate finance entities assessing whether Indian afforestation programmes are delivering measurable, additional climate outcomes.

10. Frequently asked questions

Q1. Is morimieru proposing to replace State CAMPA's own monitoring?

No. The proposal is a parallel public-good layer that state forest departments can adopt voluntarily, and that external reviewers (CAG, CEC, civil society) can use independently. The official monitoring chain continues to be the authoritative one.

Q2. Are state forest departments asking for this?

Mixed. Some senior officers have publicly welcomed satellite-based transparency (it reduces the inspection burden on under-staffed divisions). Others reasonably worry about misinterpretation of pixel-level data by people unfamiliar with field reality. The cryptographic chain matters precisely because it lets both sides reason from the same data.

Q3. Is the data accuracy good enough to call out a state?

Used responsibly, yes for the gross failure modes (no plantation present, complete canopy collapse, fire), no for fine-grained claims (exact species mix, exact stem count). The intent is to surface where the system is working and where it isn't — not to litigate marginal cases via remote sensing.

Q4. What does it cost a state to participate?

The morimieru web layer is free. Adapting it to a specific state's plot list is a one-time digitisation effort that the state holds the data for. Compute and storage at scale are in the low single-digit lakh rupees per state per year — trivial against the CAMPA corpus.

Q5. Where does this connect to CCTS afforestation credits?

Directly. The same satellite layer that monitors a CAMPA CA plot can serve as baseline and MRV for a CCTS afforestation project, if the state or user-agency registers it. CAMPA monitoring and CCTS MRV converge on the same data infrastructure.

11. References & sources

• Compensatory Afforestation Fund Act, 2016 and CAF Rules, 2018 — Ministry of Environment, Forest and Climate Change (MoEFCC)
• T.N. Godavarman Thirumulpad v. Union of India — Supreme Court of India, ongoing forest case
• National CAMPA — Annual reports and CAG audit findings
• Central Empowered Committee (CEC) — Periodic reviews on CAMPA implementation
• Parliamentary Standing Committee on Science & Technology, Environment, Forests and Climate Change — review reports on CAMPA
• Forest Survey of India — India State of Forest Report
• e-Green Watch (NIC / MoEFCC) — central plantation-tracking system
• Copernicus Sentinel-2 / Sentinel-1 — open satellite data
• WWF HydroSHEDS / HydroBASINS — global watershed boundaries (CC-BY)

Last updated 2026-05-26. morimieru is operated as a public-good infrastructure. This article is descriptive, not a regulatory commentary. State-specific implementation notes will be added as pilot conversations progress.