Researchers have dismissed suggestions that rising temperatures associated with global warming could lead to an increase in the number of malaria cases.
Previous models have predicted that the optimal temperature for transmission is 31 degree centigrade, but the new model suggests this is 25 degrees and that malaria transmission would drastically decrease above 28 degrees.
"Past models showed the whole world was going to light up with malaria - it was quite terrifying," New Scientist quoted senior author Kevin Lafferty, who is based jointly at the University of California, Santa Barbara and the US Geological Survey, as saying.
However, the new model, which takes into account the effects of temperature on a variety of key aspects of insect and parasite physiology, suggests a different pattern.
The new model agrees with past models that cooler areas without malaria like the African highlands, parts of Europe and the US, may warm up to give conditions ripe for transmission.
According to study leader Erin Mordecai from the University of California, Santa Barbara, places currently most favourable for malaria transmission may not remain so in a warmer world.
This model focuses on a key factor used to calculate the risk of spread of a disease called R0, or the Basic Reproduction Number. This measures the number of secondary cases that arise from a single case in a susceptible population.
Past models have assumed a relatively simple relationship between R0 and temperature. The Anopheles mosquito which carries the parasite is cold-blooded, so models have typically assumed that with increase in temperature the propensity for the spread of malaria also goes up.
But it is known that aspects of the life history of the Anopheles mosquito and the malaria-causing parasite it carries increase with temperature to a certain point and then rapidly decline above that temperature.
The new model combines data from other studies on how factors like mosquito bite, parasite development and mosquito egg-laying rates change with temperature.
Crucially, the team found that their model's predictions matched real data from 14 countries in Africa on the rate at which people were bitten by infectious mosquitoes.