Fixing a Critical Climate Accounting Error

K. L. Kline et al. misunderstand the important distinctions between global accounting of emissions for scientific and reporting purposes and accounting of emissions toward carbon limits under treaties and national laws. Our Policy Forum ("Fixing a critical climate accounting error," T. D. Searchinger et al., 23 October 2009, p. 527) focuses on the latter, and the perverse incentives created by the current legal exemption of emissions from bioenergy combustion.

Although Kline et al.'s E-Letter mainly supports the current accounting system, they ultimately call for counting bioenergy's greenhouse gas emissions based on a life-cycle analysis. Whatever its details, a life-cycle analysis for bioenergy is only necessary if bioenergy is not inherently carbon-neutral in the sense of having zero net effect on greenhouse gas emissions. Life-cycle analysis therefore differs fundamentally from the accounting of concern under many U.S. and European laws as well as the Kyoto Protocol, which treat all biomass as 100% carbon-neutral. By supporting a life-cycle–based approach and simultaneously defending the status quo, Kline et al. are supporting contradictory accounting systems. Ultimately, as our supporting materials point out, our solution parallels a life-cycle analysis while avoiding double-counting.

Kline et al. argue that exempting emissions from bioenergy combustion creates no accounting error when emissions from all sources and all countries are otherwise counted. In that event, the failure to count combustion emissions in one country is balanced by reported changes in carbon stocks in another. We made the same point in our Policy Forum. That was the rationale behind the Intergovernmental Panel on Climate Change's accounting recommendations for the U.N. Framework Convention on Climate Change, which requires emission reporting from all sources and countries. But the fundamental point of our article is that this accounting treatment of bioenergy is not valid for the Kyoto Protocol, nor for any existing or proposed national laws that enforce limits on emissions from energy but not from land use. That omission eliminates the carbon balance created by full accounting and incorrectly treats all bioenergy, from whatever feedstock, as having no net emissions.

Kline et al. suggest that more accurate, coordinated analysis of land-use emissions is the solution to this problem. However, under the treaties and laws, land-use emissions do not legally count toward greenhouse gas limits. Therefore, just analyzing them more accurately does not solve the problem. Kline and colleagues claim that our approach would violate the accounting principle that holds countries responsible only for emissions generated internally. According to them, the error we identify is not an error but only a variation of this general rule, which means importing countries are not responsible for the emissions generated in making the goods they purchase abroad. Kline et al. have the problem backwards. When developed countries use bioenergy, the combustion emissions occur within their borders regardless of where the feedstock was grown. Just as countries are responsible for emissions from burning imported, and domestic oil, this principle means they should also be responsible for emissions from bioenergy. Today, countries can ignore these emissions by implicitly claiming a credit for land-use activities that offset these emissions even if those offsetting activities were to occur abroad, or were never to occur at all. We do not object to a policy that provides such credits in principle, but the validity of the credits depends on the amount of "additional" carbon stored, or emissions reduced through the activities that produced the feedstock. The offset credit should therefore not be automatic, but should depend on that feedstock.

Kline et al. suggest that the problem we identify is unimportant because deforestation is declining, but they are factually incorrect. The recently estimated decline in the percentage of greenhouse gas emissions from deforestation does not result from a decline in the area deforested annually but rather from increases in total energy emissions, inclusion of non-CO₂ greenhouse gases in the calculation, and recalculations of earlier deforestation estimates (1). Regardless of the magnitude of overall deforestation today, which has many causes, the relevant focus for bioenergy policy is the additional deforestation potentially triggered by bioenergy production. Our Policy Forum cited credible estimates saying that ignoring all bioenergy emissions could lead to vast deforestation as carbon caps tighten.

Finally, we agree that counting the true carbon reductions from biomass for energy is more complicated than incorrectly assuming that it is always carbon-neutral. The analysis required is comparable to that needed to assess the net benefits of any non-capped activity, such as land-based offsets like afforestation. As one recent study for the British government fortunately indicates, forms of biomass that would achieve the greatest greenhouse gas reductions, like little-used or unused residues, are also those in whose reductions we can place the greatest confidence precisely because they are most obviously "additional" (2).

Timothy D. Searchinger

Princeton University, Princeton, NJ 08544, USA.

Steven Hamburg

Environmental Defense Fund, Boston, MA 02108, USA.

Dan Kammen

Energy and Resources Group, University of California at Berkeley, Berkeley, CA 94720, USA.

Gene Likens

Cary Institute of Ecosystem Studies, Millbrook, NY 12545, USA.

Ruben Lubowski

Environmental Defense Fund, Washington, D.C. 20009, USA.

References

1. G. R. van der Werf et al., Nature Geosci. 2, 737 (2009).

M. Brander et al., "Methodology and evidence base on the indirect greenhouse gas effects of using wastes, residues, and by-products for biofuels and bioenergy: Report to the Renewable Fuels Agency and Department for Energy and Climate Change" (Econometrica, Imperial College, London, 2010).

T. Searchinger et al. propose "Fixing a critical climate accounting error" (Policy Forum, 23 October 2009, p. 527). We agree that greenhouse gas (GHG) emission accounting needs to be more comprehensive, but believe that Searchinger's proposal would make matters worse by increasing the complexity and uncertainty of calculations. Solutions must be practical and verifiable to be effective.

Country borders have been chosen as system boundaries to inventory GHG emissions under the Kyoto Protocol. The use of country boundaries is clear and allows summing over all countries. The country inventories purposefully account for where and when both fossil-fuel combustion emissions occur, and changes in the biological stocks of carbon occur. The approach can be widely adopted, but this accounting is hampered by uncertain data (1, 2) and two basic shortcomings: Not all countries are required to report, and not all biological carbon stocks are inventoried. A first step to improve inventories would be to address these issues through concerted cooperation to improve the reliability of land cover and carbon stock data and establish comprehensive accounting of current stocks.

Under the Kyoto approach, sequestration is assumed to offset bioenergy emissions over the long term. On a short-term basis, emission reductions in one country (e.g., where bioenergy is used to replace fossil fuel) can be associated with emission increases elsewhere (where biomass is harvested). Such transfers are not unique to bioenergy. Developed countries import manufactured goods that are often produced with GHG emissions in developing nations (3). Fossil fuel exploration can lead to significant land-use emissions in supplying nations that go unaccounted in consuming nations (4–6).

To "fix bioenergy accounting," Searchinger et al. propose that all net changes in greenhouse gases (not just carbon) be traced globally (not by country) along with indirect effects, or "leakage emissions resulting from changes in land use." Unfortunately, national statistics and the organizations that compile them cannot come close to fulfilling this proposal. Searchinger's fix would require complex allocation of net emissions among diverse non-energy co-products as well as attribution among fluid and site-specific forces affecting land-use change, including governance, economic growth, international trade, poverty, energy and food production, policies, and demographics (7, 8). Such boundless indirect effects are impossible to measure or validate in practice (9) and although debate continues, the contribution of land-use change to global emissions is acknowledged to be small (10 to 15%) and shrinking compared with that of fossil fuels (10).

Calculating the effects of bioenergy on GHG emissions should be based on life-cycle analyses with the use of consistent systems boundaries and specifications supported by empirical evidence. Because accounting for emissions has implications for the treatment of biofuels in climate legislation, an appropriate and scientific approach that can be implemented by all nations is essential. We can greatly improve GHG accounting if all countries participate, all biological carbon sinks and sources are included, and all parties use consistent, verifiable methods. Better data on carbon stocks are essential for all the above.

Keith L. Kline, Virginia H. Dale

Center for Bioenergy Sustainability, Oak Ridge National Laboratory, Oak Ridge, TN 37831–6038, USA.

Alan Grainger

School of Geography, University of Leeds, Leeds LS2 9JT, UK.

References

1. A. Grainger, Proc. Nat. Acad. Sci. U.S.A. 105, 818 (2008).

2. P. Waggoner, Resources for the Future Discussion Paper 09-29 (2009).

3. G. P. Peters, E. G. Hertwich, Environ. Sci. Technol. 42, 5 (2008).

4. A. J. Liska, R. K. Perrin, Biofuel., Bioprod. Bior. 3, 318 (2009).

5. M. Finer et al. PLoS ONE 3(8), e2932 (2008).

6. S. Wunder,. CDR Working Paper 97.6 (Danish Institute for International Studies, 2007).

7. H. J. Geist, E. Lambin, Bioscience 52, 143 (2002).

8. K. L. Kline, V. H. Dale, R. Lee, P. Leiby, Issues Sci. Technol. 25(3) (2009); www.issues.org/25.3/kline.html.

9. B. A. Babcock, Iowa Ag. Rev. 15, 9 (2009); www.card.iastate.edu/iowa_ag_review/summer_09/article2.aspx.

10. G. R. van der Werf et al., Nat. Geosci. 2, 737 (2009).