Scientists in Britain and Japan have unveiled a fast-track way towards breeding crops with higher yields or resistance to climate change.

Early beneficiaries should be Japanese farmers who need salt-loving rice plants after their fields were submerged in last year's tsunami.

The technique unveiled Sunday, which does not use genetic modification, pinpoints DNA variants which confer specific qualities in a plant.

Armed with this knowledge, breeders can then use classic methods to splice these genes into an existing strain.

Right now, it can take up to five or even 10 years to develop a strain, which is known as a cultivar. But the "MutMat" approach should speed this marathon to a sprint of little more than a year, say its inventors.

"Essentially, it helps to get to the needle in the haystack faster," Sophien Kamoun, a professor at The Sainsbury Laboratory in Norwich, eastern England, told AFP.

The method, reported in the journal Nature Biotechnology, focussed on a Japanese wild rice cultivar called Hitomebore.

Researchers led by Ryohei Terauchi from the Iwate Biotechnology Research Centre created 1,500 variants of Hitomebore, each of which had different characteristics, using a chemical treatment to speed up the natural rate of mutation.

They honed in on a variant that had a trait for higher yields and crossed it with the original strain of Hitomebore. The resultant plant was then self-pollinated and grown.

The scientists compared the genome of this progeny with that of the original Hitomebore. Like laying one map on top of another, they were swiftly able to spot the genetic telltale for the bigger yield.

The process offers a huge gain in time for plant breeders, say the scientists.

Traditionally, breeders have to cross many generations of plants to ensure that desired genes are anchored in the cultivar and unwanted ones are stripped out.

But the new method quickly highlights the right genes, meaning that it should not take more than a few generations of fine-tuning to come up with the desired outcome.

In their experiment, Terauchi's team identified the traits for semi-dwarfism, which leads to short, stubby plants with a full head of grain.

It was this characteristic that famously unleashed the Green Revolution in the 1960s, boosting rice harvests in China, India and other countries that teetered perpetually on the brink of famine.

The team has since grown a collection of plants from Hitomebore which cope with high salinity.

"Once genes contributing to salt tolerance are identified, they will be used for developing rice cultivars suitable for cultivation in the roughly 20,000 hectares (50,000 acres) of paddy fields of the northern Japan coast that were flooded by the tsunami," the study says.

Kamoun said MutMat was simpler than other gene-spotting methods and was especially promising as it could improve a crop that had already adapted to local conditions.

The right genes are introduced "by classical breeding," he added. "There's no GM [genetic modification] involved in this approach at all."

Other crops with a relatively small and uncomplicated genome are excellent candidates for MutMat but complex species such as wheat and corn would be difficult, he said.

Kamoun said MutMat only became feasible through cheap computing power and low-cost gene sequencing.

"This is what's so exciting for the future," said Kamoun. "These technologies were not available just a few years ago. The full impact on improving crops and on agriculture is going to be tremendous, and it's very timely, given the challenge we have with food security."

Last October, the UN Population Fund (UNFPA) said the global population of seven billion could rise to at least 10 billion by 2100, but could top 15 billion if birth rates are just slightly higher than expected.

That amounts to a major challenge to boost yields, use land and water more sparingly and develop crops that can cope with climate-induced drought and flood.