Site identification for carbon sequestration in Latin America: A grid-based economic approach

https://doi.org/10.1016/j.forpol.2004.12.003Get rights and content

Abstract

Latin America harbors a large potential for carbon sequestration and biomass production. This paper deals with the estimation of carbon supply curves for afforestation and reforestation and its implicit carbon sequestration in wood products. The methodology presented aims at determining sequestration costs for individual geographical entities, based on unit-specific land use and ecosystem information, and economic data. This approach allows us to supplement local statistics that are typically scarce and unreliable in developing countries, with independent remotely sensed data in order to have a consistent method that can be applied over a large region. The results are mapped, which allows in-depth appraisal of results in an interactive mode and quick identification of least-cost carbon sequestration sites. The model is dynamic to support decision making at various stages in the Kyoto process. After model calibration and sensitivity analysis, we conducted scenario analysis. For a carbon price scenario of $20/tC, we find that the cumulative carbon sequestration by 2012 and 2020 is about 125 MtC and 337 MtC, respectively. The net benefit by 2020 could amount up to US$ 2.3 billion using less than 4% of the area suitable for afforestation and reforestation in the next 20 years. Our long-term estimates of the cumulative sequestration potential for 100 years imply that tree planting could compensate for more than 7 years of current CO2 emissions of the region's energy sector at low costs.

Introduction

Global warming as a consequence of human-induced emissions of greenhouse gases (GHG) is perceived as a major environmental concern threatening future welfare. Scientists predict that by 2100, the globally averaged surface air temperature will increase by 1.4–5.8 °C leading to major disturbances for human settlements and natural ecosystems (IPCC, 2001). The Kyoto Protocol of Climate Change aims at capping GHG emissions from industrialized countries and allows emission trading between industrialized countries and developing countries through the Clean Development Mechanism, CDM (UNFCCC, 1998).

The CDM is applicable for energy-related projects as well as for afforestation and reforestation projects, where the latter are referred to as CDM-sinks or AR-projects. While for the first commitment period of Kyoto, 2008–2012, the market for CDM-sinks is limited (den Elzen and de Moor, 2002), the importance of CDM-sinks is in the large potential for afforestation and reforestation in developing countries that could be used beyond 2012. According to Nilsson and Schopfhauser (1995), the area available for plantations in the developing world is twenty-six times larger than in Europe, eleven times larger than in the US and three times larger than in the Former Soviet Union. Therefore, there is a need to develop methods for deriving cost curves of carbon sequestration in these regions and identify areas where carbon sequestration is cost-efficient. We take Latin America as a case study because of its land availability and ecological conditions favoring forestry projects, as well as its active participation for implementing carbon sequestration projects in the early stage of the Kyoto process (Brown et al., 2000). We estimate carbon supply curves for AR-projects and its potential benefits for carbon trading under the Kyoto agreement, and provide a geographic representation of the distribution of carbon costs.

Economic studies on carbon sequestration in Latin America have so far provided single point estimates of sequestration costs associated with particular sequestration levels [e.g., Fearnside (1995) for Brazil, Pereira et al. (1997) for Venezuela, Masera et al. (1997) for Mexico and, Olschewski and Benítez (in press) for Ecuador]. These studies provide information on average costs of carbon sequestration for particular regions, but do not assess how these costs increase when large-scale afforestation and reforestation programs are implemented. In contrast to these studies, we evaluate how the heterogeneity of prices (e.g., land and timber prices), and the heterogeneity in land attributes (e.g., net primary productivity and suitability for agriculture) influence sequestration costs and determine carbon supply patterns. In addition, we provide a framework for identifying least-cost sites for carbon sequestration by means of a grid-based analysis that scrutinizes all the available area for plantations in the region.

Section snippets

Methods

A myriad of economic land use change models have been developed to derive supply curves of carbon sequestration measures. Some are based on cost–benefit analysis (Sathaye et al., 2001), while others involve more comprehensive analyses like partial and general equilibrium approaches (Callaway and McCarl, 1996), econometric models (Plantinga et al., 1999, Stavins, 1999), timber supply models (Sohngen et al., 1999, Sohngen and Sedjo, 2000), and land use optimization models (Parks and Hardie, 1995

Data

The analysis considers 0.5 degree cells (about 50×50 km depending on latitude) and includes 8 countries that represent more than 90% of the Latin American region (FAO, 2001, FAO, 2002). The ecological and economic data used for the analysis are discussed below.

Cost curve for 2012 and 2020

Based on the model and data described in the previous sections, we estimated the Latin American carbon supply for the years 2012 and 2020 considering 2000 as starting year. We consider 2012, since it is the end of the first commitment period of Kyoto and 2020 for providing insights into the post-Kyoto era. As shown in Fig. 1, we find zero-cost options for carbon sequestration at the left-side of the curve (the carbon price appears to be negative), where timber benefits would provide sufficient

Conclusions

This paper describes a methodology for deriving supply curves of carbon sequestration through AR-projects. The method is based on determining sequestration costs for cells of a homogenous geographical grid. For each cell, spatial information obtained from GIS databases was used for estimating carbon uptake, timber production and land prices. Major advantages of the method are that (i) there is no need to entirely depend on comprehensive data that are often scarce in developing countries,

Acknowledgments

Special thanks to Ian McCallum, Harrij van Velthuizen and Sylvia Prieler for the hours they spent extracting data in the appropriate forms and producing high-quality GIS maps, and to Henk Folmer, Jos Sijm and Ger Klaasen for their valuable comments. We acknowledge the financial support of the Energy Research Centre of The Netherlands (ECN).

References (37)

  • EROS Data Center, 2002. Global land cover characterization database, Version 2. Available at:...
  • FAO

    Global forest resources assessment 2000—main report

    (2001)
  • FAO, 2002. FAOSTAT database. FAO, Rome. Available at...
  • A. Gritsevskyi et al.

    Costs of reducing carbon emissions: an integrated modeling framework approach

    Climatic Change

    (2003)
  • IPCC

    Climate change 2001: impacts, adaptation and vulnerability

    (2001)
  • B. Locatelli et al.

    Accounting methods for carbon credits: impacts on the minimum area of forestry projects under the clean development mechanism

    Climate Policy

    (2004)
  • G. Marland et al.

    National CO2 emissions of fossil-fuel burning, cement manufacturing and gas flaring

    (2000)
  • A. Martinez-Yrizar et al.

    NPP tropical forest: Chamela, Mexico, 1982–1995

    (2001)
  • Cited by (36)

    • Towards carbon neutrality: How much do forest carbon sinks cost in China?

      2023, Environmental Impact Assessment Review
      Citation Excerpt :

      4) The parameter of the θ is drawn on Benítez and Obersteiner (2006) and considers that 50% of forest products are durable products with a half-life of 20 years, and the rest are short-term perishable products with a half-life of 1 year. ( 5) The discount rate r is set at 8% referred to Benítez and Obersteiner (2006) and Zhong and Xing (2012). Besides the parameters above, data of remaining parameters select provincial data instead of corresponding city-data.

    • A global climate-economy model including the REDD option

      2019, Journal of Environmental Management
      Citation Excerpt :

      Lubowski and Rose (2013) identify four classes of REDD-relevant economic models: fixed market, partial equilibrium, general equilibrium and integrated assessment models. Fixed market models have been applied by Benitez and Obersteiner (2006) focusing on forests in Latin America, by Busch et al. (2012) in Indonesia and by Nepstad et al. (2007) in the Brazilian Amazon. Partial and general equilibrium models are capable of capturing the effects of long-run investment decisions and implications of large-scale changes with regional or global adoption of REDD (Table 1 in Lubowski and Rose, 2013).

    • Land system science in Latin America: challenges and perspectives

      2017, Current Opinion in Environmental Sustainability
      Citation Excerpt :

      These four avenues are related to the unique characteristics of the region in terms of the diversity of its social-ecological systems, regional history, demographic patterns and connections to other global regions. Latin America harbors one fifth of the world’s forests [4••], the greatest terrestrial biological diversity of ecosystems on Earth [5] and areas of crucial importance for carbon sequestration and climate regulation [6]. Latin America is also the largest world region with an overall positive bio-capacity reserve [7], namely the natural ability to provide the resources and ecosystem services that are consumed by people each year [8].

    View all citing articles on Scopus
    View full text