Agricultural fire use in the Brazilian Amazon: some evidences for the state of Pará regarding the economics of accidental fires and fallow management

• Autor(a) principal: Thiago Fonseca Morello Ramalho da Silva
• Data de Publicação: 2013
• Tipo de documento: Tese
• Idioma: eng
• Título da fonte: Biblioteca Digital de Teses e Dissertações da USP
• Texto Completo: https://doi.org/10.11606/T.12.2013.tde-13082013-201624

The use of fire as an agricultural tool perpetuates in Brazilian Amazon, despite its negative socioeconomic, environmental and public health impacts. Two topics of the problem are investigated, by looking to the current period (2009-2010) and for three municipalities of the State of Pará, namely Santarém, Belterra and Paragominas. The analysis is restricted to motivations and consequences with strictly economic nature and fires linked with deforestation are kept out of the scope of analysis. Slash and Burn Agriculture (S&BA) is practiced by smallholders mainly for growing annual crops. The first essay demonstrates that the profitability of S&BA is governed by the trade-off between cost-free fertilization through the burning of secondary vegetation and idleness of the land. Additionally, it is established that, a reduction of the fallow duration, depending on the initial duration, can generate a cash surplus that can be used to finance (at least partially) the transition to a fire-free agriculture. The second topic addressed is the one of accidental fires, conceived as a phenomemon that emerges from collective behavior. The second essay tests the hypothesis that eventual damage to assets belonging to other farmers is not internalized by farmers when they decide to start a fire. Such hypothesis is not refuted by georeferenced data for the municipality of Paragominas and for the year of 2010. For this, spatial econometric and instrumental variables models are estimated. The third essay tests the hypothesis that the risk of losses potentially imposed by fires started in neighboring farms is not accounted by farmers when deciding how to allocate their land among alternative uses. This hypothesis is not refuted by microdata at the farm level, collected through a field survey conducted in the municipalities of Santarém, Belterra and Paragominas. The analysis is restricted to 2009. The technique of Iterated Seemingly-Unrelated Regressions is employed to estimate a system of equations determining how much land is allocated to each class of land of use.