Wood you believe it? Claims that the world's tropical forests are declining can not be backed up by hard evidence, according to new research. This major challenge to conventional thinking is the surprising finding of a study by Dr Alan Grainger, senior lecturer in geography at Leeds University, and one of the world's leading experts on tropical deforestation.

"Every few years we get a new estimate of the annual rate of tropical deforestation," says Dr Grainger. "They always seem to show that these marvellous forests have only a short time left. Unfortunately, everybody assumes that deforestation is happening and fails to look at the bigger picture - what is happening to forest area as a whole."

In the first attempt for many years to chart the long-term trend, he spent more than three years going through all available UN data - and found some serious problems.

Rather than universal decline, the data appeared to indicate a slight increase in some types of tropical forest. In a few countries, such as Gambia and Vietnam, forest area has actually expanded since 1990, as the reforestation rate has exceeded the deforestation rate.

"The picture is far more complicated than previously thought," he says. "If there is no long-term net decline it suggests that deforestation is being accompanied by a lot of natural reforestation that we have not spotted."

But Dr Grainger does not claim that tropical deforestation is not occurring, as there is plenty of local evidence for that. His point is that, clearly, we cannot rely on our available data.

"What is happening to the tropical forests is so important, both to the peoples of tropical countries and to future trends in biodiversity and global climate, that we can no longer put off investing in an independent scientific monitoring programme that can combine satellite and ground data to give a reliable picture," he says.

Dr Grainger says we need a World Forest Observatory to monitor changes in forests in the tropics and elsewhere. "Only then will we really know what has happened to tropical forests over the last 40 years."

Political initiatives to tackle climate change have renewed the interest of western governments in tropical forestry. Much of the debate centres on biofuels, such as bio-ethanol or bio-diesel, made from processing-plant material or waste oil. In theory, biofuels are a way of reducing greenhouse gas emissions. Burning the fuels releases carbon dioxide, but growing the plants absorbs a comparable amount of the gas from the atmosphere.

A major fear concerning biofuel production is that countries may be tempted to replace rainforest with crops for biofuels, such as sugar cane or palm oil plantations.

Meanwhile, if increased proportions of food crops, such as corn or soy, are used for fuel, that may push prices up, affecting food supplies for less prosperous citizens.

A third concern is the energy which is used in farming and processing the crops. This can make biofuels as polluting as petroleum-based fuels, depending on what is grown and how it is treated.

The great hope for overcoming all these stumbling blocks lies with the so-called second-generation of biofuels, made from breaking down cellulose - a material found in the woody bits of plants. Using cellulose would allow for a much greater range of plants to be converted into biofuel - ie not only food crops.

A major breakthrough was announced this week, with the news that switchgrass, a crop touted by venture capitalists and environmentalists alike as a next-generation ethanol feedstock, yields about five times more energy than is needed to grow it, making the plant a far more efficient fuel source than corn.

In addition, the life cycle of the switchgrass ethanol - which includes growing the crop, making the fuel and burning it in vehicles - emits about 94% less of planet-warming carbon dioxide than the life cycle of gasoline, said the study, published on Tuesday in the Proceedings of the National Academy of Sciences.

US companies are racing to make economical ethanol from non-food sources. Switchgrass, which used to grow naturally across wide swaths of the US, can be grown on marginal crop land using far fewer energy-intensive inputs, such as fertiliser, than corn needs. And since it does not double as a feed crop, it will not lead to higher grain prices.

Cellulosic ethanol currently costs about double the price as making the fuel from corn, the main US ethanol feedstock. But venture capitalists and companies that are making small amounts of cellulosic ethanol say once the industry gets under way, biological advances in the fungi and other organisms used to break down woody plants into fuel will make the process cheaper.