CND01 / Annual
National review
Nine chapters would cover the year, major events, the work of SEDEC units, the federation, public policy and links to the Sendai Framework.
CND / Reading edition02 NOV 2026 / BR · CN
Project under review · methodological pilot completed
National DisasterSituationReport
An annual account of what happened across Brazil, who was affected and how public authorities acted. The proposal would bring municipal, state and federal civil defence experience into one national reference.
Fig. 01Cartographic plateBR / 001
4,262municipal records
2,551municipalities with a record
76,191municipal records
5,256municipalities with a record
Historical series1991—2025
Geographic coverage27 federative units
Proposed cycleAnnual review
FIG. 02Brazil · historical series1991 — 2025
35 years of municipal records
Select a year to see the national total and open the same selection in the atlas.
Year2025
- municipal records
- 4,262
- municipalities with records
- 2,551
municipal recordsBR / 35
Source: S2iD / Digital Disaster Atlas · IBGE 2022 boundaries
A single event can generate records in several municipalities. No record does not establish the absence of a disaster.
View data table
| Year | municipal records | municipalities with records |
|---|---|---|
| 1991 | 403 | 378 |
| 1992 | 590 | 544 |
| 1993 | 1,324 | 1,300 |
| 1994 | 231 | 222 |
| 1995 | 818 | 766 |
| 1996 | 490 | 469 |
| 1997 | 748 | 705 |
| 1998 | 1,178 | 1,095 |
| 1999 | 741 | 735 |
| 2000 | 595 | 549 |
| 2001 | 1,824 | 1,679 |
| 2002 | 1,816 | 1,587 |
| 2003 | 1,606 | 1,465 |
| 2004 | 2,424 | 1,935 |
| 2005 | 2,314 | 2,095 |
| 2006 | 1,580 | 1,366 |
| 2007 | 1,831 | 1,642 |
| 2008 | 1,921 | 1,616 |
| 2009 | 3,005 | 2,246 |
| 2010 | 2,086 | 1,714 |
| 2011 | 1,740 | 1,298 |
| 2012 | 2,325 | 2,122 |
| 2013 | 2,911 | 2,453 |
| 2014 | 2,505 | 2,179 |
| 2015 | 2,113 | 1,707 |
| 2016 | 2,352 | 1,963 |
| 2017 | 3,089 | 2,267 |
| 2018 | 2,560 | 1,807 |
| 2019 | 2,538 | 1,785 |
| 2020 | 3,880 | 2,482 |
| 2021 | 3,556 | 2,218 |
| 2022 | 5,021 | 2,776 |
| 2023 | 5,115 | 2,803 |
| 2024 | 4,699 | 2,676 |
| 2025 | 4,262 | 2,551 |
01The publication
How did Brazil experience disasters over the year?
The answer is usually scattered. A municipality records damage in S2iD. A monitoring centre keeps the time a warning was issued. A local team knows when a bridge reopened. The report would put those pieces on the same timeline.
02 / Territorial
State profiles
One profile per state or federal district would show municipalities reached, local capacity and progress on recovery. State civil defence teams would help explain changes.
03 / Four-year
Full report
A four-year review would examine changes in exposure, losses and response capacity. Open tables and methodological notes would accompany the series.
02Five phases
What would each team contribute?
Choose a phase to see the editorial question, the evidence needed and the teams that would contribute. Editorial assignments are proposed and subject to agreement.
Where did risk information change a land-use decision?
- Evidence to gather
- Susceptibility and risk maps, municipal plans, the register of susceptible areas and land-use decisions, with dates and geographic boundaries.
- Proposed contribution
- DPM, state teams and municipalities would check the connection between the risk assessment and the decision.
Which measures reduced an existing risk?
- Evidence to gather
- Slope stabilisation or drainage projects, works completed, the population potentially protected and residual risk. Completion and effectiveness are separate measures.
- Proposed contribution
- DPM, implementing agencies and local civil defence would document the objective, the work delivered and what still needs monitoring.
Did a warning arrive in time to inform local action?
- Evidence to gather
- Times of the technical bulletin, warning and local action, channel coverage, contingency plans and completed drills.
- Proposed contribution
- DPS and CENAD would assemble preparedness evidence with Cemaden, other monitoring centres and the civil defence teams disseminating warnings.
Who needed rescue, what assistance arrived and when?
- Evidence to gather
- Disaster forms, decrees, recognition decisions, operations, humanitarian assistance and transfers, each with its date and counting unit.
- Proposed contribution
- The preparedness and rescue and humanitarian assistance departments would assemble federal records. States and municipalities would explain shelter use, assistance and gaps.
What service was restored, and when could residents use it again?
- Evidence to gather
- The work plan, funds released, physical progress, delivery and local verification. Emergency restoration precedes long-term reconstruction.
- Proposed contribution
- The restoration and recovery department would trace projects with implementing authorities. The profile would report disbursement and physical delivery separately.
03From the federation to the council
A local record in its national context
S2iD organises administrative records and cases. The Digital Atlas consolidates the historical series. The annual report would add an account from those who monitored, responded and rebuilt.
- 01
Municipality
Records the disaster, damage and needs. Identifies what was not assessed and revises the record as evidence becomes available.
- 02
States and Federal District
Review regional coverage and overlaps, document warnings and support, and discuss the profile with participating municipalities.
- 03
SEDEC
Units take editorial responsibility for their chapters. DPM and the information team would coordinate drafting, version control and reconciliation. Planning and budget staff would qualify financial figures.
- 04
CONPDEC
Brazil’s national civil protection council could use the report in its advisory role to monitor the National Plan, discuss territorial gaps and propose priorities. A separate proposed editorial board would commission, review and approve the publication.
Trace a figure from S2iD to the publication
The consolidated Atlas extract contains 4,262 municipal records for 2025. Of these, 2,253 are classified as recognised. Recognition is a status in this selection. Adding the two figures would count some of the same rows twice.
Each edition would retain its reference year, cut-off, version, counting rule and corrections. CONPDEC could then ask a state why occurrences appear in the Atlas while warning or assistance records for that territory remain incomplete.
S2iD / Atlas · 9c3f4404acfc
Demonstration profile / SCSC27° SSanta Catarina in 2025
The Atlas contains 738 municipal records from 252 of the state’s 295 municipalities. Heavy rainfall was the most frequent type in this selection. The state team would add the sequence of warnings, operations and delivered works.
Inspect Santa Catarina in the atlas ↗04Editorial calendar
How a year becomes a comparable edition
Pilot issue No. 0, completed in August 2026, demonstrated the method using 1991–2025 data. An internal rehearsal covering 2025 and three state profiles is proposed before the public launch.
- 31 Mar
Proposed data freeze
Consolidate records for the reference year, freeze a version and document outstanding items and revisions.
- Apr—Jun
Joint drafting
Each unit would draft a chapter. State and municipal teams would help verify coverage and context.
- Jul—Sep
Review and reconciliation
Review discrepancies, explain limitations and submit the edition to the proposed editorial board.
- 13 Oct
Proposed launch
The first public edition would cover 2026 and launch on 13 October 2027. The first four-year report is proposed for 2030.
05Brazil–China dialogue
How do two countries count the same kind of disaster?
China already publishes an annual account of natural disasters. In its 2025 review, released in January 2026, the Ministry of Emergency Management and the office of the national commission drew on several agencies and described human, economic and territorial impacts.
Read China’s 2025 national review ↗BR CN
A first technical workshop
Each delegation could choose a flood episode and bring a public record. The workshop would compare reference periods, territorial units, people affected, warnings and recovery stages. A bilingual glossary and two annotated timelines would be the first outputs.
- When one event crosses several municipalities, what is the counting unit?
- When the number affected is revised, how is the previous version preserved?
- Which local observations help explain the interval between warning and recovery?
A proposal for discussion. Exchanges and institutional participation would require agreement.
06Next edition
What decisions would allow work to begin?
Review the proposal, establish an editorial board, assign responsibilities by function and agree the calendar with the units and state civil defence teams. The brief sets out the deliverables and decisions for each participant.
Sources and interpretation
Atlas extract for 1991–2025, version 9c3f4404acfc. Municipal records are subject to revision and under-reporting. A place with no record may still have experienced a disaster. Economic losses with different time bases require reconciliation.
Digital Disaster Atlas · IBGE 2022 boundaries. A single physical event spanning several municipalities can generate several records.