CHALLENGE
There is growing concern that Brazilian agriculture and forestry sectors are increasingly vulnerable to climate variability and change (eg. a decline in productivity of subsistence crops in northeastern Brazil; increased risk of fire and Amazonian forest dieback). To meet development, food security, climate adaptation and mitigation, and trade goals over the next several decades, Brazil will need to significantly increase per area productivity of food and pasture systems in central and southern Brazil while simultaneously reducing deforestation, rehabilitating millions of hectares of degraded land for cropping and forest plantations in the Amazon, the cerrado, the Atlantic forest, and the Pampas.

However, because of previous climate modeling and data limitations, there was still significant uncertainty associated with the projections for Amazonian rainfall (timing, seasonality) as well as the magnitude and locations of climate impacts in Brazil over the next 50 years. Improved climate change impact assessments were urgently needed to guide policy makers on priorities, geographical targeting (hot spots), and phasing of investments for adataptation and mitigation to climate change.

APPROACH
This activity implemented by the World Bank in partnership with leading Brazilian agencies and supported by PROFOR focused on establishing a robust and integrated decision and policy support framework that could empower Brazil’s policy makers and the agricultural sector to undertake the adaptation necessary to cope with projected climate change.

Its objectives were:

• 1.To refine the available climate change projections for Brazil via the coupling of global, regional, and local scale modeling currently being tested by the Brazilian National Institute for Space Research (INPE) in Brazil and the regional climate program for South America (CREAS).
• 2. To integrate the INPE and CREAS suite of tested global (300 km horizontal resolution) and regional models (50 km horizontal resolution) with the state of the art Brazilian developments in Regional Atmospheric Model (BRAMS) that incorporates aerosol and land cover/deforestation/burning feedbacks for much improved local weather and climate (especially rainfall) projections.
• 3. To make the Brazilian Agro-Eco Zoning Model that is currently used by the Central Bank of Brazil for rural credit programs "climate-smart" by integrating the high resolution climate projection outputs from 1 and 2.
• 4. To make the existing Brazilian Land Use Model (BLUM) climate-sensitive by coupling it with the outputs described above to assess: (a) Climate change induced changes in supply and demand of agricultural commodities at a national level, (b) Changes on the distribution of land use and production (agriculture, forestry, pasture) in Brazil for given supply and demand scenarios, and (c) Economic effects on agricultural and forestry production and profitability.

RESULTS
Projections for land use and forest land cover change, and an assessment of the distribution of land use and production were undertaken. Specifically, previous climate change projections were refined using a global, regional and local scale modeling; Regional Climate Change Scenarios for South America (CREAS) were used for regional and sectoral impact assessments; the Agro Climatic Risk and Vulnerability Zoning Model developed by EMBRAPA and UNICAMP was refined; soils were classified and cropping areas that are less vulnerable to climate change impacts were identified based on temperature effects through 2020 and 2030. A final report, in English and Portuguese, is forthcoming.

FINDINGS
The study showed that while for some crops (soybean and cotton) the projected negative climate impacts to 2020 are likely to be more moderate than previously projected, for other crops (beans and corn), however, the impacts could be significantly more severe than projected in previous studies. All the climate change scenarios simulated in this study resulted in a reduction of ‘low risk’ cropland area in 2020 and 2030.

More specifically, the findings suggested that the South Region of Brazil, currently an agricultural powerhouse, could potentially lose up to 5 million hectares (ha) of its highly suitable agricultural land due to climate change, while Brazil as a whole could have around 11 million ha less of highly suitable agricultural land by 2030. The findings of this study are being incorporated by EMBRAPA into the EMBRAPA/UNICAMP Agroecozone Model to improve the climate projections that underpin the national rural credit and insurance programs in Brazil.

Read the report in Portuguese.

Follow us on twitter and facebook or join our mailing list for regular updates.