Integration of Land Use, Fire, and Carbon Flux in Critical Amazon Landscapes: the Xingu River Headwaters and the BR163 Highway Corridor
Oriana Almeida IPAM - Instituto de Pesquisa Ambiental da Amazônia (SA-PI)
2001 as global economic forces and technological advances have increased the profitability of cattle and soybean production in the region. Cattle ranching interacts synergistically with logging to cause forest understory fires that can double land-use driven carbon emissions from the Amazon during years of severe drought. However, the growing role of international markets in the region’s land-use dynamics also means that producers are under pressure to improve their land management practices, potentially weakening the positive feedback between land-use, drought, and fire.
Accurate simulation of the potential impact of these new market forces on Amazon landscapes and carbon emissions will require integration of land-use models that are sensitive to changes in the profitability of production and land management strategies. These models must have sufficiently high resolution to capture property-level land-use decisions, for this is where land-use policies are played out. They must also, however, cover landscapes that are large enough to be relevant to the regional planning processes that are underway in the region’s most dynamic frontiers.
We propose to build upon our previous LBA-ECO research to develop two high-resolution (0.2-0.5 km pixel), sub-regional models that integrate economic models of four land uses, an ecosystem model, and fire regime algorithms to simulate future trajectories of land-use change, fire, and associated carbon fluxes. We will model two dynamic agricultural frontiers:
the headwaters of the Xingu River and the BR-163 highway corridor. We will test the hypotheses that: (a) forest fire ignition sources are highest in landscapes dominated by smallholders and lowest in soy landscapes, (b) cattle rancher dependence on fire as a management tool declines as ranch profitability increases, (c) many smallholder settlements are located where the potential profitability of farming is low, perpetuating dependence on semi-subsistence agriculture, and (d) “best” land management practices and compliance with environmental legislation are favored through higher prices for environmentally certified soy, beef, and timber, potentially lowering carbon emissions. The resulting modeling systems will incorporate several LBA datasets, including maps of selective logging scars, soy production, and secondary roads. These modeling systems will provide market- and policy-sensitive simulations of future trajectories in land use that are important input layers for hydrological, erosion, and climate models, for habitat fragmentation analyses, and in support of land-use zoning projects.
Hence, the proposed research will address one of the most elusive LBA goals:
support for sustainable development. It will support the PhD dissertations of five students (three Brazilian), and has been requested by both Amazon state agencies and the Brazilian federal government.