Best Available Science?

Yesterday, the Scientific Committee on Problems of the Environment (SCOPE) released a report, “Biofuels: Environmental Consequences and Interactions with Changing Land Use,” based on the proceedings of its International Biofuels Project Rapid Assessment.

The Environmental Working Group immediately praised the report, claiming that it “confirms that corn-based ethanol is a dead end.” (I recommend reading this response to EWG by Biofuels Digest’s Jim Lane.)

Far from that confirmation, the report’s recommendations recognize that whether the maximum environmental benefit of biofuels is achieved depends on how they’re produced, and that “many of the adverse effects of biofuels on the environment could be reduced by using best agricultural practices.” In fact there are 17 papers included in the report, drawing a range of conclusions about the benefits or dangers of biofuels.

Still, the executive summary cites a number of premises regarding biofuels that environmental groups now consider articles of faith:

The rapidly growing production of biofuels requires additional cropland. In some cases, this additional land comes from agricultural land previously used to grow food or feed crops. In a hungry world, these diverted crops must be made up elsewhere, thus driving land conversion– perhaps in different countries and on different continents – to compensate for the loss of food-crop production.”

I’ve heard these premises asserted as fact too many times now, yet no one has yet publicly demonstrated direct evidence of them. Within academic papers, it has become customary to footnote the paper by Searchinger et al in Science last February, but even that paper cited the same assumptions.

It is certainly a reasonable hypothesis to test whether crop production throughout the world has expanded into ecologically sensitive carbon sinks as a result of the U.S.’ inability to meet its share of increased worldwide demand for food and feed. It is also reasonable to test the alternative hypothesis that increases in crop productivity are able to meet the incremental increases in biofuel production under the RFS. Either way, it would likely be most effective to work for an international accord to directly protect ecologically sensitive carbon sinks, but that would take more work.

It should also be noted that production of agricultural commodities was shifting to other countries prior to the 2005 and 2007 laws. In point of fact, Argentina, Australia, Brazil, Canada, Costa Rica, the European Communities, Guatemala, India, Nicaragua, Mexico, Thailand and Uruguay have all joined a WTO dispute against the United States aimed at expanding in most cases their own agricultural production.

Iowa State University’s Center for Agricultural and Rural Development noted in response to that WTO case:

Farmers base their decisions about what and how much to plant on numerous factors, including rotation considerations, production costs, expected market prices, availability of crop insurance, and expected benefits from farm programs. The complicated nature of these decisions makes it quite difficult to determine if U.S. farm programs for crops other than cotton are vulnerable to a WTO case against them on the basis of price suppression. The role that these programs play in farmers’ planting decisions varies across crops, regions, and crop years. Simple “rules of thumb” that use total payment levels as a guide or the belief that the programs work as a cheap food policy are inadequate measures of the impacts of farm payments on U.S. supply and international commodity prices.”

Those conclusions are applicable to the current debate on indirect land use change.

But now the assumptions cited by environmentalists are set to be enshrined within California’s Low Carbon Fuel Standard, despite the fact that California admits the available evidence contradicts the premises.

Nathanael Greene of the Natural Resources Defense Council considers the California law and the EPA’s expected rule on the RFS, which utilizes similar modeling, to be the best available approach:

In order to develop a full lifecycle accounting protocol that includes emissions from indirect land-use change, both regulators are relying on economic models. They use these models to look at the world first without the biofuels and then with them; the change in pollution is assigned to the biofuels. While the models are complex, both agencies have relied on the best peer-reviewed science and economics and will update their rules regularly over time.”

Jim Lane of Biofuels Digest has provided a very cogent and much needed analysis of the methodology being used to “measure” the indirect land use change carbon emissions attributed to biofuels:

In short, we’re arguing about whether the models make for a good forecast. A problem is that we haven’t backcast — that is, checked the predictions of the model against known outcomes in the past to see if the predictions were accurate.”

Another problematic aspect of the use of general equilibrium models in lifecycle analysis is that they require the assumptions that the environmentalists cite as articles of faith. The models are therefore incapable of testing the assumptions as hypotheses. General equilibrium models, by definition, must assume a point of equilibrium and then assume a shock to the system. They are properly used to assess the risks of a new policy or program that is expected to have an economic impact on markets. But the calculated outcome of the modeling, the new predicted equilibrium, is not an actual measurement.

So these models do not look at the world without biofuels and the world with biofuels to compare them. In fact, they make calculations that come directly from their starting assumptions. To mistake the calculations for conclusions is circular logic. It is not the best available science – it isn’t science at all.


Ethanol or Tar Sands?

Those who are pushing the inclusion of indirect land use change (ILUC) in government regulations have thus far proposed ILUC as the only indirect effect and that it only apply to biofuels. A letter from more than 100 scientists pointed out this would create unequal boundaries for transportation fuels and unfairly disadvantage biofuels.

But are there indirect effects from producing gasoline? In a March 16 hearing, staff for the California Air Resources Board said they looked but couldn’t find any. Maybe they should subscribe to National Geographic Magazine. In the cover story of their March issue, they take a look at the oil produced from the Canadian Oil Sands. This paragraph gives you a good synopsis:

Nowhere on Earth is more earth being moved these days than in the Athabasca Valley. To extract each barrel of oil from a surface mine, the industry must first cut down the forest, then remove an average of two tons of peat and dirt that lie above the oil sands layer, then two tons of the sand itself. It must heat several barrels of water to strip the bitumen from the sand and upgrade it, and afterward it discharges contaminated water into tailings ponds like the one near Mildred Lake. They now cover around 50 square miles.

Only 150 square miles of the oil sands have been mined thus far but the government has leased 1,356 square miles that are minable. According to the article, the U.S. imports more oil from Canadian Oil Sands than any other country. Imports from Canada are about 19 percent of our total oil imports. That number is expected to grow and according to a new paper in the journal Environmental Research Letters, it will contribute to significantly higher greenhouse gas emissions for gasoline.

Why is this important? Well, if you read the latest EIA Outlook, you will see that the U.S. will continue to require increasing amounts of liquid transportation fuels. The report also said that growth in U.S. crude production is “limited after 2010, however, because newer discoveries are smaller, and capital expenditures rise as development moves into deeper waters.” So the two most likely source to meet the increased need are biofuels, which are constantly improving their GHG profile, and oil sands, which are moving in the other direction. Which one do we want?

Controversy on Discounting Greenhouse Gas Emissions

On March 13, Sens. Tom Harkin (D-Iowa) and Charles Grassley (R-Iowa) along with 10 others sent a new letter to EPA Administrator Lisa Jackson urging “EPA to refrain from including any calculations of the ILUC components in determining life-cycle GHG emissions for biofuels at this time.” The new letter referenced a letter sent in November to then-Administrator Stephen Johnson, which was signed by four of the current 12 signers as well as then Sen. (now Interior Secretary) Ken Salazar.

The latest letter argues that quantification of indirect land use change is difficult because there are many factors that affect it. Current models do not take into account the many causes of land use change or their overlapping effects. California’s recently proposed rule on the life cycle accounting shows how difficult it can be. While California’s Air Resources Board staff use modeling that relies on an assumption that use of land to grow crops for biofuels will push agricultural production to other countries and that this effect can be estimated through use of a general equilibrium market model, they acknowledge that the prediction is inconsistent with actual trade data.

The Senators wisely say,

It is possible that future domestic and international climate change policies will include major provisions restricting land use changes. Indeed, that may be the most appropriate and effective way to reduce GHG emissions associated with land use changes.”

The group also raises the issue of the use of a discount rate in the analysis of the reductions of greenhouse gases attributed to both biofuels and gasoline. (See earlier post on discounting.)

Michael J. Roberts, and Agricultural Resource Economist at North Carolina State University, makes a case for setting the discount rate at zero:

I have no idea how much we should spend reducing GHG emissions because I have no idea how costly reductions would be, the costs should warming occur, or how GHG affects all the possibilities. But I think a low discount rate, in the ballpark of zero, should be used when weighing current expenditures to future expected benefits.”

So how costly might reducing GHG emissions through use of advanced biofuels be? Sandia National Laboratories’ 90 Billion Gallon Biofuel Deployment Study suggests that the infrastructure costs of developing a large-scale advanced biofuels industry are essentially equivalent to those of drilling for more oil in U.S. onshore locations such as ANWR and the Rockies.

Use of a discount rate of any sort in the EPA rule seems highly inappropriate. It is well-noted that the choice of a discount rate can greatly affect the outcome of a cost-benefit comparison. But consider that the Renewable Fuel Standard requires biofuels to achieve a percentage reduction in greenhouse gas emissions compared to a gasoline baseline. The use of a discount rate on both biofuels and the gasoline baseline arbitrarily closes the gap between the two, understating both the costs of gasoline and the benefits of biofuels.

California Proposes Numbers for Indirect Land Use Change Emissions

California issued a staff report last week for its “Proposed Regulation to Implement the Low Carbon Fuel Standard.” As expected, the rule proposes a measure of indirect land use change emissions for select biofuels – corn and sugarcane ethanol and soy biodiesel. The report defines the assumptions behind the analysis – in a word, that use of existing crops for biofuels reduces supplies, increases prices, and thereby induces agricultural expansion:

Land use change effects occur when the acreage of agricultural production is expanded to support increased biofuel production. Lands in both agricultural and non-agricultural uses may be converted to the cultivation of biofuel crops. Some land use change impacts are indirect or secondary. When biofuel crops are grown on acreage formerly devoted to food and livestock feed production, supplies of the affected food and feed commodities are reduced. These reduced supplies lead to increased prices, which, in turn, stimulate the conversion of non-agricultural lands to agricultural uses. The land conversions may occur both domestically and internationally as trading partners attempt to make up for reduced imports from the United States. The land use change will result in increased GHG emissions from the release of carbon sequestered in soils and land cover vegetation.”

In addition to releases of 90 percent of above-ground carbon in the first year, and 25 percent of below-ground carbon over a period of 30 years, the ILUC penalty calculated by California says, “The carbon that would have been sequestered in the lost cover vegetation is also included in the total emissions value.”

A letter signed by more than 100 researchers and biofuel company executives and sent to California Gov. Schwarzenegger on March 2 calls the analysis being conducted by California “unusually sensitive to the assumptions made by the researchers conducting the model runs.” It also points out that the concept of an indirect effect is only being proposed for biofuels, even though, “Petroleum, for example, has a price-induced effect on commodities, the agricultural sector and other markets.”

California’s analysis does account for the direct land use change effects of oil drilling – within California. However, despite an acknowledgment that, “The cost of energy appears to have been the largest contributor” to the recent sharp increase in corn prices, the report does not analyze the possible indirect land use effect of that oil price shock.

Further, the report indicates that the evidence one would expect to see of indirect land use change caused by using corn for biofuels is lacking:

The model predicts, for example, that the expanded use of domestic corn for the production of ethanol will reduce U.S. corn exports. That prediction appears to be inconsistent with the actual trade data appearing in Appendix C. Those data show that the production of corn, soybeans and wheat in the United States has generally been on the increase over the last decade. Exports meanwhile have remained relatively steady. In the case of corn, production increases have been sufficient to supply the ethanol industry while maintaining export levels.”

So, despite any inconvenient contrary evidence, California’s Air Resources Board concludes precisely what it assumes from the beginning – that Midwest corn ethanol emits more CO2 than California gasoline.

Dr. Robert Brown of the Iowa State University Bioeconomy Institute gave an interview to Brownfield Network to discuss the letter sent by the academics. Dr. Brown notes that California is formulating policy on conjecture — not even a testable hypothesis. Listen to the entire interview here.

Europe to Study Indirect Land Use

The European Parliament on Dec. 17 adopted amendments to the Renewable Energy Sources Directive, raising targets for production of biofuels but at the same time setting strict sustainability standards to monitor and reduce greenhouse gas emissions from the use of road transport fuels. The Parliament’s adopted text makes clear that it intends to calculate climate change emissions from international land use, but that the science is not currently available to do so:

(11) In calculating the greenhouse gas impact of land conversion, economic operators should be able to use actual values for the carbon stocks associated with the reference land use and the land use after conversion. They should also be able to use standard values. The work of the Intergovernmental Panel on Climate Change is the appropriate basis for this. That work is not currently expressed in a form that is immediately usable by economic operators.”

The text also includes this assessment of the risk of indirect land use change and the need for an accurate measurement:

(18) Even if biofuels themselves are made using raw materials from land already in arable use, the net increase in demand for crops caused by the promotion of biofuels could lead to a net increase in the cropped area. This could be into high carbon stock land, in which case there would be damaging carbon stock losses. To alleviate this risk, it is appropriate to introduce accompanying measures to encourage an increased rate of productivity increases on land already used for crops; the use of degraded land; and the adoption of sustainability requirements, comparable to those laid down in this Directive for EU biofuel consumption, in other biofuel-consuming jurisdictions. The Commission shall develop a concrete methodology to minimise greenhouse gas emissions caused by indirect land use changes. In doing this the Commission shall analyse, on the basis of best available scientific evidence, in particular, inter alia, the inclusion of a factor for indirect land use changes in the calculation of greenhouse gas emissions and the need to incentivise sustainable biofuels which minimise the impacts of land use change and improve biofuel sustainability with respect to indirect land use change. In developing this methodology, the Commission should inter alia address the potential indirect land use change effects of biofuels produced from non-food cellulosic material and from ligno-cellulosic material.”

The agreed upon amendments to Directive 98/70/EC include a two-year study of indirect land use change that is to include methods to ensure that sustainable biofuels avoid causing land use change:

7d. (6). The Commission shall, by 31 December 2010, submit a report to the European Parliament and to the Council reviewing the impact of indirect land use change on greenhouse gas emissions and addressing ways to minimise this impact. This report shall where appropriate be accompanied, in particular by a proposal, based on the best available scientific evidence, containing a concrete methodology for emissions from carbon stock changes caused by indirect land use changes, ensuring compliance with this Directive, in particular Article 7b(2).”

Annex IV. Rules for Calculating Life Cycle Greenhouse Emissions from Biofuels, includes the calculation of GHG reductions for different types of biofuels without land use change.

Note that the U.S. Energy Security and Independence Act also called for a National Academies study of indirect land use impact, to be completed within 18 months of the law’s enactment. That study has not been funded.

Yes, Virginia

One of the most strikingly circular arguments put forward to support inclusion of current estimates of indirect land use change emissions in both California’s Low Carbon Fuel Standard and the EPA’s Renewable Fuel Standard Life Cycle Assessment is that these estimates are so large. The University of California Berkeley Letter to EPA from Michael O’Hare et al. and the previous letter to California’s Air Resources Board by the same group (Mark Delucchi et al.) are examples of the argument:

The salience of this requirement lies in the size of current estimates of these indirect emissions: added to typical direct emissions values, they indicate that substituting certain biofuels, especially corn ethanol, for gasoline will actually increase the global warming (GW) intensity of motor fuel, or decrease it so little (depending on how it is calculated) that these biofuels would fail to meet EISA required GHG reductions.”

And again:

The best methods currently available for estimating market-mediated effects are economic models such as partial and general equilibrium models. Several groups are currently employing these models to estimate indirect LUC, and despite considerable uncertainty, none has concluded that zero grams of CO2 per megajoule is the best estimate of the effect. Ignoring an effect that may be large simply because it is uncertain is unjustifiable.”

And once again:

So far no models, in particular no peer-reviewed models, have been advanced that come up with values for iLUC that are significantly lower than those in the Searchinger et al paper.”

So in the spirit of the holiday season, I’d like to offer a similar argument. This year, 2008, will be the 50th anniversary of the North American Aerospace Defense Command’s (NORAD) tradition of tracking Santa’s flight from the North Pole around the world. The tradition began in 1955, but NORAD inherited it in 1958. Fifty-plus years of scientific modeling and measurement of the phenomenon ought to be considered proof positive that Santa Claus exists.

Happy Holidays everyone.

EPA Continues to Move Toward Calculation of Indirect GHGs

Dr. Bruce Dale of Michigan State University published a Commentary in the Washington Times on Nov. 26, calling for sound science in development of the EPA’s rule on the Renewable Fuel Standard.

Dr. Dale points out, “No actual data exist that connect U.S. domestic ethanol production with, to cite a widely quoted example, the clearing of the Amazon rain forest. Instead, the paper’s conclusions depend entirely on economic modeling and assumptions.” He then discusses his own recently completed research that shows that changes to the assumptions and different calculations for some of the variables can dramatically change the outcome of the lifecycle greenhouse gas equation. His research is to be published in a forthcoming edition of Environmental Science and Technology.

See Dale: Unexplored Variables and Dale: Unexplored Variables Supporting Information

The New Fuels Alliance on Oct. 23 sent a letter to the California Air Resources Board raising similar cautions about using estimates of indirect impacts in regulating biofuels. The letter notes:

CGE [computable general equilibrium] models like GTAP [Global Trade Analysis Project] provide estimates of land use change in distant locations, but at the price of severe limits in accuracy and at the expense of a realistic inclusion of complex causes of land use change.”

The NFA notes that CARB has not investigated the indirect effects of other fuels, particularly petroleum, an idea explored in previous posts to this blog. (See here and here.) The indirect effects of petroleum prices are vast, according to the letter. “If the rising price of agricultural commodities is a concern – as the catalyst for additional planting – it is now clear that oil prices have a profound effect on agricultural commodity markets.” In fact, researchers at FAPRI include oil prices as a significant cause of uncertainty in calculating the indirect effects of biofuels on land use.

In general, Dr. Dale’s conclusion is consistent with NFA’s statement, “The fundamental assumption of the current ILUC argument – that using an acre of land in the U.S. for fuel will require almost an acre of crop development somewhere else – produces questionable results when applied to ‘good’ public policy initiatives.”

The EPA’s Regulatory Impact Analysis from the 2005 version of the Renewable Fuel Standard makes clear that they do use this assumption in several places. The reason for accepting these assumptions, according to the EPA, is that existing life cycle models fail to take into account general equilibrium principles on market impacts.

The agricultural sector modeling results indicate that, compared to the 2012 Reference Case, approximately two and a half million acres will come out of CRP land as a result of increased renewable fuel production. Not all of these two million acres will go directly into corn production used to produce ethanol. However, the entire amount of CO2 emissions from the CRP land use change is attributable to the increased amount of ethanol produced, as without the increased demand for corn there would be no change in CRP land.”

And they indicated with the publication of the RIA that they will use the assumption in estimating indirect land use in future legislation:

Due to decreasing corn exports some changes to international land use may occur, for example, as more crops are planted in other regions to compensate for the decrease in crop exports from the U.S. While the emissions associated with domestic land use change are well understood and are included in our lifecycle analysis, we did not include the potential impact on international land use and any emissions that might directly result.”

EPA at the time also applied this assumption to oil production:

Under a full lifecycle assessment approach, the savings associated with reducing overseas crude oil extraction and refining are included here, as are the international emissions associated with producing imported ethanol. This assumes that for every gallon of gasoline that’s not imported into the US, the corresponding quantity of crude oil is not extracted or processed to make this gasoline regardless where the extraction or production takes place.”

A fuller exploration of the market impacts of biofuels and oil are warranted, particularly after this year’s food and gas price swings.