GC13H-1252
Reconciling Agricultural Needs with Biodiversity and Carbon Conservation in a Savanna Transformation Frontier
Monday, 14 December 2015
Poster Hall (Moscone South)
Marcus P Spiegel, Lyndon D Estes, Kelly K Caylor and Timothy Searchinger, Princeton University, Princeton, NJ, United States
Abstract:
Zambia is a major hotspot for agricultural development in the African savannas, which will be targeted for agricultural expansion to relieve food shortages and economic insecurity in the next few decades. Recent scholarship rejects the assumption that the large reserves of arable land in the African savannas could be converted to cropland with low ecological costs. In light of these findings, the selection of land for agricultural expansion must consider not only its potential productivity, but also the increase in greenhouse gas emissions and biodiversity loss that would result from the land conversion. To examine these tradeoffs, we have developed a multi-objective optimization technique to seek scenarios for agricultural development in Zambia that simultaneously achieve production targets and minimize carbon, biodiversity, and economic cost constraints, while factoring in the inter-annual variability in crop production in this highly uncertain climate. Potential production is determined from well-characterized yield potential estimates while robust metrics of biodiversity and high resolution mapping of carbon storage provide fine scale estimates of ecological impact. We draw production targets for individual crops from potential development pathways, primarily export, commodity-crop driven expansion and identify ecologically responsible agricultural development scenarios that are resilient to climate change and meet these demands. In order to achieve a doubling of production of nine key crops, assuming a modest 20% overall increase in yield potential, we find a range of scenarios that use less than 1600 km2 of new land without infringing on any protected areas or exceeding 6.7 million tons of carbon emissions.