Queimadas/INPE - BDQueimadas¶

Overview¶

Field	Value
Institution	INPE — Instituto Nacional de Pesquisas Espaciais
Website	queimadas.dgi.inpe.br
agrobr access	Direct (public CSVs)

Data Origin¶

Source¶

URL: https://dataserver-coids.inpe.br/queimadas/queimadas/focos/csv/
Format: CSV (latin-1 or UTF-8), ZIP for historical data
Access: Public, no authentication
Granularity: Daily (focos_diario_br_YYYYMMDD.csv) and monthly (cascading fallback)

Available Data¶

Fire Hotspots¶

Satellite detection of heat spots (hot spots) over the Brazilian territory:

Geographic coordinates (lat/lon)
GMT date and time of detection
Detecting satellite (13 satellites)
Municipality and state
Biome (6 Brazilian biomes)
Indicators: days without rain, precipitation, fire risk, FRP

Coverage¶

Temporal: Since 2003 (annual data); monthly since 2023; direct CSV since 2024
Spatial: Entire Brazilian territory
Frequency: Daily (updated several times a day)

Cascading fallback (monthly)¶

The INPE server changed the organization of historical data. The client tries in order:

Period	Format	URL
2024+	monthly `.csv`	`mensal/Brasil/focos_mensal_br_YYYYMM.csv`
2023	monthly `.zip`	`mensal/Brasil/focos_mensal_br_YYYYMM.zip`
2003-2022	annual `.zip`	`anual/Brasil_todos_sats/focos_br_todos-sats_YYYY.zip`

For annual data, the full CSV of the year is downloaded and filtered by the requested month.

Usage¶

Monthly Hotspots¶

import asyncio
from agrobr import queimadas

async def main():
    # All hotspots of September/2024
    df = await queimadas.focos(ano=2024, mes=9)
    print(f"{len(df)} focos detectados")

    # Filter by state
    df = await queimadas.focos(ano=2024, mes=9, uf="MT")

    # Filter by biome
    df = await queimadas.focos(ano=2024, mes=9, bioma="Cerrado")

    # With metadata
    df, meta = await queimadas.focos(ano=2024, mes=9, return_meta=True)
    print(meta.source, meta.records_count)

asyncio.run(main())

Daily Hotspots¶

# Hotspots of a specific day
df = await queimadas.focos(ano=2024, mes=9, dia=15)

Combined Filters¶

# Hotspots in Mato Grosso in the Amazon by reference satellite
df = await queimadas.focos(
    ano=2024, mes=9,
    uf="MT",
    bioma="Amazonia",
    satelite="AQUA_M-T",
)

Schema¶

Column	Type	Description
`data`	date	Detection date
`hora_gmt`	str	GMT time (HH:MM)
`lat`	float	Latitude (-35 to 6)
`lon`	float	Longitude (-74 to -30)
`satelite`	str	Satellite name
`municipio`	str	Municipality name
`municipio_id`	Int64	IBGE code
`estado`	str	State name
`uf`	str	State abbreviation (2 characters)
`bioma`	str	Brazilian biome
`numero_dias_sem_chuva`	float	Days without precipitation
`precipitacao`	float	Precipitation (mm)
`risco_fogo`	float	Risk index (0-1)
`frp`	float	Fire Radiative Power (MW)

Satellites¶

INPE monitors fire hotspots with 13 satellites. The reference satellite is AQUA_M-T (MODIS), used in the official statistics for having the longest and most consistent time series.

Cache¶

Aspect	Value
TTL	12 hours
Policy	Fixed TTL

Updating¶

Aspect	Value
Frequency	Daily
Reference satellite	AQUA_M-T passes ~13h and ~01h30 UTC