By Richard Black
BBC environment correspondent
A major argument used by sceptics of global warming is flawed, a UK Met Office study in Nature magazine says.
Some say any warming that is measured is due to urbanisation
This argument maintains that much recorded climate data is inherently unreliable because of where weather instruments are situated.
Most are in or near cities, which produce their own heat; so the rapid warming measured over the last century could be just a record of urbanisation.
The Met Office believes its study shows this "urban heat island" idea is wrong.
The analysis has been done by Dr David Parker. He used data for the last 50 years to create two separate graphs. One plots temperatures observed on calm nights, the other on windy nights.
If the urban heat island hypothesis is correct, he says, instruments should have recorded a bigger temperature rise for calm nights than for windy ones - because wind blows excess heat away from cities and away from the measuring instruments.
But there is no difference between the curves. "It helps to answer the critics," Dr Parker told BBC News.
"There are other kinds of temperature measurements, too, which could not be influenced by urbanisation, such as warming in the oceans.
"Different methods of measurement can produce different rates of warming but they all point upwards."
Dr Myles Allen, from the atmospheric physics department at Oxford University, agrees: "It's pretty convincing," he said.
"It's a sensible analysis which tests a prediction of the sceptical theory; and if it's right, we should see a greater effect on calm nights.
"But you should never underestimate the ingenuity of the sceptics to come up with a counter-argument."
One of the most prominent scientifically grounded sceptics is Fred Singer, president of the Science and Environmental Policy Project in Arlington, Virginia, US.
"Many people have tried to correct for the urban heat island effect," he told BBC News.
"I'm not sure David Parker has succeeded, but we admire his ingenuity."
Dr Singer said that many weather stations in apparently rural settings might experience a certain amount of heating from nearby settlements or even roads.
"The only true rural records are proxy records, such as the ones from tree rings," he argued.
Other types of proxy records - where natural processes are examined to provide a an indication of past climate - include stalactites, fossil beds, ice cores, ocean sediments and glacial deposits.
The sceptics accuse the "warmers" of being selective in their use of these proxies to show that recent trends go beyond what would be expected from natural variation.
A separate research paper, also published in this week's Nature, has found a different kind of proxy - but definitely of human origin.
A French team, led by Dr Pascal Yiou, from the Climate and Environments Laboratory in Gif-sur-Yvette, examined parish records of grape-harvest dates in Burgundy going back to 1370.
"Grape harvest dates, which are closely related to temperature... may provide one of the longest uninterrupted series of regional temperature anomalies," the writes in the journal.
The study focussed on the Pinot Noir grape; and shows that in Burgundy, temperatures in the 1990s, although higher than average, were not that unusual - some years in the 1400s and 1500s had been warmer.
"This takes us back about as far as the comprehensive tree-ring records," commented Dr Allen, "and the nice thing is, there's an absolute record of what happened in which year.
"The obvious problem would be if there were changes in agricultural practice, or if the Pinot Noir grape had changed significantly over time."
Dr Yiou believes things have not changed: "The farming practices are the same; they still don't use any chemicals or anything like that," he told BBC News.
"Also, we compared our temperature records from the grapes with tree-ring data from nearby places, and they showed the same pattern."
The one year which stands out as being exceptionally warm is 2003 - the year of the French heatwave which claimed thousands of lives.
"It confirms 2003 was an unprecedented year in France," said Dr Allen.
"But this only tells us about one region of France; somewhere in the world, extreme temperatures happen every year."
Dr Yiou now plans to expand his studies to Bordeaux and Italy, where there are similar records.