Model predicts future deforestation in Central Africa

Nature, published online 20 November 2009; doi: 10.1038/news.2009.1100

Model predicts future deforestation

Projections could help Central African nations in Copenhagen climate talks

by Anjali Naya, Nature News, November 20, 2009

A computer model that predicts future changes in the world's forests could strengthen the case of Central African nations that are calling for compensation in exchange for protecting their natural resources.

Forest management is expected to be a key point of discussion at the United Nations climate summit in Copenhagen in December. Countries will negotiate on how to reward rainforest nations for protecting their forests, a mechanism dubbed REDD for 'reducing emissions from deforestation and forest degradation'.

Deforestation rates in the Congo Basin rainforest — the second-largest rainforest on Earth — have hovered at around 0.15% per year for the past 15 years. But preliminary results from the model, unveiled this week, predict that forest cutting in the region will increase to 0.3–0.5% per year by 2020–2030.

Major rainforest countries that have historically had high deforestation rates — such as Indonesia (2.0%) and Brazil (0.6%) — are pushing for compensation that is based on historical trends. With a relatively high business-as-usual scenario, they are expected to reap above-average rewards for any decreases in deforestation.

But using historical trends will short-change the countries of the Congo Basin, some argue. Although in the past this region has had low deforestation rates, recent improvements in the road network as well as planned mining and timber projects are likely to increase deforestation rates considerably in coming years. "There are strong indications that Central African forests are at a critical turning point for the future," says Carlos de Wasseige, the coordinator of an EU-funded project called Forests of Central Africa, whuch hopes to set up a regional forest monitoring centre.

"Most proposals for [REDD] suggest that history is the best predictor of tomorrow," says Michael Obersteiner, who is leading the development of the forestry model at the International Institute for Applied Systems Analysis, based in Laxenburg, Austria. "But for [Central African] countries, the forward-looking projections will be more reliable."

Model forest

The model, which has a resolution of 10–50 km², is a combination of three global land-use models called GLOBIOM, G4M and EPIC. Its predictions are based on key global drivers of deforestation, including population growth and gross-domestic-product growth, as well as global demand and production of biofuel, timber and agricultural crops. The model works by calculating the profitability of forest clearance in certain areas on the basis of topography, soil composition and climate.

Obersteiner recognizes, however, that the model is only as good as the data going into it — and those data can often be difficult to obtain, either through lack of resources to collect them or because the information is held by private companies.

"The idea of publicly available reliable statistics escapes our country," says André Kondjo-Shoko, head of forest inventory at the Democratic Republic of the Congo's environment ministry. "The statistics don't represent reality."

Another problem is that the model does not account for a major driver of deforestation: illegal tree cutting for charcoal and timber.

The conservation group WWF now hopes to fill this data gap using a "geo-wiki," also unveiled this week, which provides a repository for forestry information. Modelled on the open-source principles of the online encyclopaedia Wikipedia, "it combines the principals of social networking with the craft of spatial mapping," says Leo Bottrill, who heads the geo-wiki project at the WWF's office in Washington, DC.

Bottrill started the geo-wiki after struggling to collate information about the Congo Basin region. For the past two years, he has been collecting baseline information on mineral deposits, forest concessions and planned infrastructure such as roads, railways and transmission lines. He hopes that the site's users will be able to contribute data and maps, either through the Internet or by text message, as well as commenting on and editing the content.

The WWF will launch a pilot of its geo-wiki for the Democratic Republic of the Congo in March 2010, and hopes the system will be fully functional by mid-2010. "We hope people will run with the idea," says Bottrill.

Link:  http://www.nature.com/news/2009/091120/full/news.2009.1100.html