Published online 2 December 2010 | Nature | doi:10.1038/news.2010.640

News

Researchers row over whether regional projections are based on sound science.

Mexico's climate-change programme has been criticized as being based on poor projections. Reuters

MEXICO CITY

Some of the key climate-change projections used by the Mexican government to support its environmental policies could be flawed, local scientists have warned as the country hosts this year's UN climate summit in Cancún, which runs from 29 November to 10 December.

The government uses national and regional projections on changes in temperature and precipitation to assess the vulnerability of the country's population to extreme climate events, water scarcity, floods and heat waves, and to calculate the economic cost of the impacts, as well as of emissions reductions and adaptation measures. Mexico hopes to cut its greenhouse gas emissions to 50% of the 2000 levels by 2050.

But according to a group of researchers at the National Autonomous University of Mexico's Atmospheric Sciences Centre (CCA-UNAM) in Mexico City, the country's climate change programme may be based on flawed science.

The group, led by climate-change economist Francisco Estrada, is questioning the set of regional climate-change scenarios produced by Victor Magaña, a well-known climatologist also based at the CCA-UNAM who is one of the key academics advising the government on climate impacts.

Downscaling doubts

Magaña's national-scale estimates of future temperatures and rainfall were first published in 2007 in a technical report commissioned by the government's National Institute of Ecology (INE) for policymakers. He recently published another report setting out guidance on how to produce regional climate-change scenarios and use them for decision-making.

Estrada says that these scenarios, which form part of the basis of climate-impact assessments in government reports on climate change, used obsolete and inadequate methods and shouldn't be used to guide decision-making. A paper including the claims has been submitted to the journal Climatic Change. Prominent climatologists based at the CCA-UNAM — Cecilia Conde, Benjamín Martínez and Carlos Gay García — are co-authors on the paper, which has yet to be accepted.

The group is arguing that the scenarios are an example of the misuse of statistical downscaling — a tool that allows scientists to calculate local effects of larger-scale patterns predicted by global circulation models (GCMs), mathematical models of planetary atmospheric or oceanic circulation. GCMs work at resolutions of 150–300 square kilometres, and downscaling may allow estimating information at scales of 50 kilometres or less.

Using this technique, Magaña obtained results that were markedly different from projections by the Intergovernmental Panel on Climate Change, the main body assessing climate-change science.

When such differences arise, they need to be explained, says climatologist Eric Salathé, who works with regional climate-change models at the Joint Institute for the Study of the Atmosphere and Ocean at the University of Washington in Seattle.

"There are regional atmospheric processes that could yield divergence from a global model," says Salathé. "A high-resolution climate model might do a better job simulating these effects and produce a different regional pattern of warming. However, statistical downscaling methods such as those used [by Magaña] are not designed to improve on the modelling of physical processes and feedbacks."

Magaña denies that there is a problem with his projections. "There is not a definitive conclusion on what is the best statistical downscaling option," he says. In any case, he adds, "Adaptation policies don't need to have the very highest high-resolution scenarios, but general trends on how climate change will affect the population at risk."

He agrees with Estrada, however, that dynamical downscaling methods, which use both statistics and the physics of climate to produce regional climate-change projections, are a better alternative to purely statistical downscaling. "But that requires a lot of computational capacity and a lot of time and money," says Magaña.

A further problem, both scientists say, is the lack of a strong national observational network, which diminishes the quality of field data.

Climate economics

Estrada has also criticized a July 2009 climate report commissioned by the government, which like the United Kingdom's Stern Review on the Economics of Climate Change attempts to estimate the impact of climate change on the economy.

The Economics of Climate Change in Mexico, coordinated by UNAM-based economist Luis Galindo, suggests that by 2100, the costs of climate-change impacts will be at least three times greater than the costs of cutting Mexico's greenhouse-gas emissions by 50%.

But Estrada says that the report has significantly underestimated the overall costs of climate change for the Mexican economy. The economic models used in this report, he says, are normally used in short-term projections, and are not suitable for long-term ones. In addition, the report only included projections based on one scenario — which assumed that emissions would continue to rise unchecked.

"To determine if a given emission-reduction policy is appropriate for Mexico the report should include cost estimates for different stabilization scenarios," he says.

ADVERTISEMENT

Horacio Catalán, an economist at UNAM who helped Galindo to produce the July 2009 report, told Nature that all criticism is "very welcome". The report is preliminary, he says, and had to deal with "a lot of uncertainties". "We expect to improve our estimates," he adds.

Like the UK government, Mexico does not ask independent bodies to peer-review its technical reports. But Estrada argues that the review process for all these documents should be at least as rigorous as it is for other scientific publications.

Magaña disagrees. "You don't need peer review in official reports to be sure that you've got science good enough to make the right decisions," he says.

Salathé, however, says it is crucial that these reports have the confidence of the community. "Given these issues, this confidence can probably only be restored through an evaluation by an independent panel," he adds.