Methodology, revised on
We publish how we assess: structure, scales, sources, validation, governance and limitations of the ESGreen Score and the IRC-ESGreen (Climate Risk Index).
Evidence, not declarations. Every number carries a source, a date and a version.
Every result can be traced back to its source, with collection date and methodology version.
The ESGreen Score and the IRC-ESGreen have scales running in opposite directions. Every reading shows the scale name and direction.
The methodology is versioned and every IRC-ESGreen run is dated.
The IRC-ESGreen was validated on a real portfolio and compared with events that actually happened.
ESGreen delivers risk, value and a suggested action. The decision always belongs to the institution.
These principles underpin the Verification stage of the Monitor, Report, Verify and Decide cycle.
Always read the scale name before the number.
799 / 1.000
the higher, the lower the riskThe ESGreen Score 2.0 is an ESG and climate risk indicator per company. It consolidates public and private information into a comparable score, read across 5 dimensions and published with the Confidence Index.
Each layer brings together signals from specific sources and feeds a dimension. Each dimension appears separately in the score, to show where the risk comes from.
The Score is presented as a numeric value, with the scale direction, the risk band and the Confidence Index. A higher value indicates lower risk. The breakdown by dimension and by layer shows where the risk comes from, and the monthly history shows when it changed.
The bands are the same ones the platform shows. The band is an input for the institution's policy, which makes the decision.
A climate risk index per land parcel, operation and portfolio, from 0 to 100. The higher, the greater the risk.
The weighting between the three pillars is described in the methodology note.
Each horizon is a window projected from the run date. “hist” is the historical baseline, used as the reference for the 7, 15, 30, 60 and 90-day horizons.
RUN 09/30/2026, METHODOLOGY irc-v1.0 Illustrative data
The climate component uses forecast models from the NVIDIA Earth-2 ecosystem. ESGreen is a member of NVIDIA Inception.
The analyst does not need to interpret a climate model. They get risk, value and visibility.
Portfolio climate exposure over the horizon analyzed (example). Not a forecast of accounting loss.
The suggested action is an operational recommendation. The credit, underwriting or provisioning decision belongs to the institution.
414 contracts, 1,023 land parcels and 160 cities in PR, SC and RS, in the portfolio of a credit cooperative in southern Brazil.
A −1.6 °C frost forecast for July 14 in Caxias do Sul, identified 12 days in advance.
The April 20, 2026 run pointed to two waves of instability over Rio Grande do Sul. The May 1, 2026 event recorded 324 mm in seven hours in Rosário do Sul. In the study, which included ten companies in the South and Southeast, the signals arrived up to 18 days in advance.
94.3% of the portfolio analyzed showed some level of exposure to extreme climate events in the previous two years, in a study covering more than 1,000 areas and properties in southern Brazil.
| Official zoning (ZARC) | IRC-ESGreen |
|---|---|
| Static, by municipality and by crop season. | Dynamic, per land parcel, recalculated with every run. |
Cross-referencing with a dynamic analysis can reveal exposures that aggregate zoning does not show at the operation level. The comparison requires a methodological basis, a sample and statistical evidence for the subset analyzed.
ESGreen Evidence Assessment measures a company's ESG maturity with a questionnaire adapted to its size and CNAE (Brazilian industry classification). Each answer is backed by documents, reviewed with AI support for consistency and fit with the criterion. The result is a maturity score, with an annual history.
The maturity score uses evidence provided by the assessed company itself. It is not a credit rating, certification or legal opinion.
Every data point with source, collection date and version.
SOURCE: IBAMA, UPDATED 09/30/2026 08:53, METHODOLOGY score-v3.2 Illustrative data
| Group | Examples |
|---|---|
| Registration and corporate structure | Federal Revenue Service (Receita Federal): CNPJ, CNAE, registration status, legal nature and shareholder structure; politically exposed persons |
| Tax | Tax clearance certificates; federal tax debt roll (Dívida Ativa) |
| Legal | CNJ and courts: critical lawsuits |
| Environmental and territorial | IBAMA and ICMBio (violations and embargoes), INPE, MapBiomas, SEEG, IBGE |
| Land tenure | SICAR/CAR, permanent preservation areas (MMA), Indigenous lands (FUNAI), quilombola territories (INCRA) |
| Labor | MTE (employer registry), workplace accident reports (CAT) |
| Domestic restrictive lists | CEIS, CNEP, Ibama, TCU, CVM |
| International sanctions | OFAC, United Kingdom, European Union, Interpol, UN |
| News and media | Continuous scanning of press sources |
| Climate | Climate forecast models with data updated daily |
| Client-provided data | Operation data submitted by the institution (IRC Operational pillar) and documents submitted by the assessed company (Evidence Assessment) |
Update frequency varies with each originating agency's publication schedule. The collection date is recorded on every data point.
References used in developing the methodology. Mentioning them does not imply partnership, endorsement, validation or certification by these institutions.
How we keep results explainable, reproducible and auditable.
Every methodology change creates a new version, and every result states the version used.
Every IRC-ESGreen run is dated, and each horizon is a window projected from that date.
Every result can be traced back to its source, with collection date.
Results compared with events that occurred, such as the Caxias do Sul and Rio Grande do Sul cases described above.
ESGreen delivers risk, value and a suggested action. Credit policy, underwriting and the final decision belong to the institution.
Information security, privacy and LGPD: See the Trust page
Incomplete, outdated or erroneous records at the originating agency are reflected in the result. The date of each data point makes it possible to assess the lag.
Not every source covers the entire country or every type of company with the same depth.
IRC-ESGreen uncertainty grows with the horizon. A high index indicates a greater risk of impact, not certainty of an event.
The published IRC-ESGreen validation refers to rural credit portfolios in PR, SC and RS. Other uses, regions and crops require their own scoping and testing.
Local phenomena smaller than the ~5 km resolution may not be captured.
The Operational pillar depends on the quality of the data submitted by the institution.
The amounts in reais describe the portfolio's exposure over the horizon analyzed. They are not a forecast of accounting loss, provision or insurance claim.
The maturity score starts from documents submitted by the assessed company, checked for consistency and fit with the criterion.
ESGreen's results are inputs for the institution's decisions. They do not replace the institution's own analysis, independent audit or legal advice.
The summary on this page is open. The methodology note details how the ESGreen Score and the IRC-ESGreen are built, for risk, model validation and audit teams.
Or write to contato@esgreen.com.br
An ESGreen specialist replies by email within one business day.
Short answers to the most common questions.
Still have questions? Talk to an expert
Because they measure different things. The ESGreen Score, from 0 to 1,000, measures a company's (CNPJ) ESG and climate risk: the higher, the lower the risk. The IRC-ESGreen, from 0 to 100, measures the projected climate risk of a land parcel, contract or portfolio: the higher, the greater the risk. Every reading shows the scale name and direction.
No. The ESGreen Score, the IRC-ESGreen and ESGreen's indices are assessments based on public and regulatory data and on proprietary models. They are not a credit rating, investment recommendation or legal opinion and should be used as input for the institution's own decisions.
The ESGreen Score is recalculated every month for the entire base and with every relevant new fact identified by continuous monitoring. The IRC-ESGreen uses climate data updated daily and is recalculated with every run, and each run has a recorded date.
Every result can be traced back to its source, with collection date and methodology version. For the IRC-ESGreen, the run date and the horizon are recorded. This lets the auditor retrace the path from the number back to its origin.
No. ZARC is the official zoning: static, by municipality and by crop season. The IRC-ESGreen is dynamic, calculated per land parcel and recalculated with every run. Cross-referencing the two readings can reveal exposures that aggregate zoning does not show at the operation level.
The index combines three pillars: Climate, ESG and Operational. The Climate pillar uses a coordinate-level forecast weighted by crop stage; the ESG pillar uses the operator's ESGreen Score; the Operational pillar uses the operation's data. The weighting between the pillars is described in the methodology note.
In a 30-minute conversation, an ESGreen specialist shows how a result is built, from source to number.
Or write to contato@esgreen.com.br
An ESGreen specialist replies within one business day to schedule a time.
