Central America Project, Environment: Conservation and Competitiveness. HIID 2001. Chapter 15.
In this chapter we consider potential gains derived from preventing deforestation, drawing heavily from information from Chapter 14. It uses the same economic model and econometric technique and the same land use/land cover data. It also uses the carbon stock estimates presented there. The key difference is that, instead of using proxies for land-use returns such as ecological characteristics related to higher productivity, we attempt to directly estimate dollar-valued returns. We use these as an independent variable to explain and predict deforestation patterns. This allows us to simulate the potential supply of carbon sequestration in response to dollar-valued returns for certified emissions reductions. Payments for CERs will reduce deforestation by lowering the net return from forest clearing. The loss of the reward for carbon sequestration will partially offset the positive return from agricultural uses. To estimate the effect of such payments on deforestation, and hence CER supply, we need to estimate the response of deforestation to changes in returns to land use. An increase in agricultural returns is empirically equivalent to a reduction in carbon CER payments. Thus, we construct a variable that estimates the potential return of a plot of land if it is cleared. We construct a variable that varies across space (different crop suitability and yields) and time (changes in export prices, technology, and labor costs). We then use this variable in our econometric estimation. The results are used to calculate a supply curve of CERs. These results are illustrative only. They are produced as part of an ongoing effort at estimation (Kerr, Pfaff, Hughes et al. 2000) and are used to show some underlying features of a dynamic supply curve.
Central America Project, Environment: Conservation and Competitiveness. HIID 2001. Chapter 14.
The chapter is structured as follows. First, below, we begin this analysis of the process influencing land changes with a dynamic model of land-use choices. Such models have often been suggested, but crucial features have often been neglected in application. This model generates testable hypotheses regarding factors underlying patterns of land-use changes in tropical areas. The next section describes the data collected for this project and discusses the quality of land-use data. It also outlines the variables used to test the implications of the model. Following that, we present our results and then discuss the linkage from land-use changes to implied carbon sequestration, and the quality of information currently available on carbon sequestration. Finally, we present some conclusions and lessons learned.
EOS (American Geophysical Union), May 15 2001
Earth-science predictions of natural phenomena are increasingly seen as valuable aids to improved societal decision making. Pielke et al. recently (EOS 7/13/99) argued persuasively that good predictions alone won’t achieve better societal decisions. These authors’ call to change the decision environments in which scientific predictions are used, though, may be more relevant to the daily activities of policy makers than to those of scientists. We see a role also for changing the information that scientists feed into those decision environments. In particular, scientists could better serve societal needs by generating not only possible scenarios, but also improved probabilities that decision makers need, including for decisions to be taken in the near future.