CENSOAGRO

Data Coverage

Data is collected at multiple geographic levels: - Country level: aggregate national statistics - State level: disaggregated by Brazilian states - Municipality level: available for select datasets (currently "livestock_production")

Historical data spans from 1920 onwards, with different time series available for different datasets based on IBGE’s survey methodology evolution.

1. agricultural_land_area

Provides comprehensive data on total agricultural land area and the number of agricultural properties.

• Key metrics: Total land area (hectares), number of properties
• Time period: 1920, 1940, 1950, 1960, 1970, 1975, 1980, 1985, 1995, 2006, 2017
• Geographic levels: Country, State
• Use case: Track long-term trends in farm consolidation and total agricultural land expansion

2. agricultural_area_use

Details how agricultural properties use their land (crop farming, pasture, forests, etc.).

• Key metrics: Area by use category (temporary crops, permanent crops, natural pastures, planted pastures, forest for forest production, protected natural vegetation, other areas)
• Time period: 1970 onwards (1970, 1975, 1980, 1985, 1995, 2006, 2017)
• Geographic levels: Country, State
• Use case: Analyze land use transitions, deforestation patterns, and agricultural intensification

3. agricultural_employees_tractors

Captures information about the agricultural workforce and mechanization levels.

• Key metrics: Number of employees, number of tractors, employed persons
• Time period: 1970 onwards (1970, 1975, 1980, 1985, 1995, 2006, 2017)
• Geographic levels: Country, State
• Use case: Study agricultural mechanization trends and rural employment dynamics

4. agricultural_producer_condition

Describes the tenure status of agricultural land (ownership, rental, partnership, etc.).

• Key metrics: Number of properties by producer condition (owner, tenant, partner, occupant)
• Time period: 1920, 1940, 1950, 1960, 1970, 1975, 1980, 1985, 1995, 2006, 2017
• Geographic levels: Country, State
• Use case: Understand land tenure structures and changes in property ownership patterns

5. animal_production

Details the number of livestock animals farmed by species and type.

• Key metrics: Number of animals by species (cattle, pigs, poultry, sheep, horses, goats, water buffalo, etc.), number of establishments
• Time period: 1970 onwards (1970, 1975, 1980, 1985, 1995, 2006, 2017)
• Geographic levels: Country, State
• Use case: Monitor livestock herd sizes and sectoral changes in animal agriculture

6. animal_products

Quantifies production volumes of animal-based products.

• Key metrics: Production quantities (eggs, milk, honey, wool, hide, etc.)
• Time period: 1920, 1940, 1950, 1960, 1970, 1975, 1980, 1985, 1995, 2006, 2017
• Geographic levels: Country, State
• Use case: Track historical trends in dairy, poultry, and other animal product sectors

7. vegetable_production_area

Provides detailed crop production data including area planted and volume produced.

• Key metrics: Area planted (hectares), quantity produced (kilograms), number of establishments by crop type
• Time period: 1920, 1940, 1950, 1960, 1970, 1975, 1980, 1985, 1995, 2006, 2017
• Geographic levels: Country, State
• Use case: Comprehensive analysis of crop production patterns and agricultural productivity

8. vegetable_production_temporary

Focuses specifically on temporary crops (annual crops that must be replanted each season).

• Key metrics: Area planted, quantity produced for crops like soybeans, corn, beans, cassava
• Time period: 1970 onwards (1970, 1975, 1980, 1985, 1995, 2006, 2017)
• Geographic levels: Country, State
• Use case: Study annual crop production cycles and seasonal variations

9. vegetable_production_permanent

Focuses on permanent crops (perennial crops that produce for multiple years).

• Key metrics: Area planted, quantity produced for crops like coffee, sugarcane, cocoa, oranges
• Time period: 1940 onwards (1940, 1950, 1960, 1970, 1975, 1980, 1985, 1995, 2006, 2017)
• Geographic levels: Country, State
• Use case: Analyze long-cycle crop production and regional specialization

10. livestock_production

Specialized dataset on bovine cattle production and related establishments.

• Key metrics: Number of cattle establishments, herd size, number of properties
• Time period: 2017 (most recent census year)
• Geographic levels: Country, State, Municipality (unique to this dataset)
• Use case: Detailed regional analysis of cattle ranching, including municipality-level data

1. dataset

Selects which dataset to download. See dataset descriptions above.

dataset = "agricultural_land_area"  # character string

2. raw_data

Controls whether to download the original data or the processed/cleaned version.

• TRUE: Returns raw data exactly as published by IBGE
• FALSE: Returns treated data with standardized formatting, variable names in English, and consistent units

Default behavior: Raw data typically requires more cleaning and interpretation, while treated data is ready for immediate analysis.

raw_data = FALSE  # logical

3. geo_level

Specifies the geographic aggregation level.

• "country": National aggregate
• "state": Disaggregated by Brazilian state
• "municipality": Available only for "livestock_production" dataset

geo_level = "state"  # character string

4. time_period

Defines which year(s) to download. Availability varies by dataset:

You can request a single year or a range of years:

time_period = 2006           # single year
time_period = c(1995, 2006)  # multiple specific years
time_period = 1995:2006      # will select years within this range that are available

5. language

Output language for variable names and labels.

• "pt": Portuguese
• "eng": English

language = "eng"  # character string

Raw vs. Treated Data

• Raw data (raw_data = TRUE): Exactly as published by IBGE, with original formatting and Portuguese variable names
• Treated data (raw_data = FALSE): Cleaned and standardized with English variable names, consistent units (hectares for area, kilograms for production quantities), and NA values properly handled

Data Organization

When using treated data, the output is typically in long format with one row per observation unit, containing: - Geographic identifiers (state, municipality if applicable) - Year of the census - Product/category names (crop type, animal species, etc.) - Quantitative measurements (area, quantity, count) - Number of establishments/properties

Important Considerations

1. Time gaps: Census data is not collected every year. Years with no data simply won’t be available.
2. Geographic changes: Brazil’s state boundaries have changed historically; use caution when comparing very old data
3. Definition changes: IBGE’s classification of crops and agricultural activities has evolved. Variables may not be directly comparable across all decades.
4. Municipality data: Currently only available for livestock_production in 2017
5. Download size: Historical data requests with multiple years may be large; plan accordingly

Citing the Data

When using this data in research or publications, cite:

IBGE - Instituto Brasileiro de Geografia e Estatística. Censo Agropecuário. Available at: https://sidra.ibge.gov.br/pesquisa/censo-agropecuario