快猫短视频

Some kinds of weather forecasting will be easier, and others will be harder, in a world made warmer by climate change. In particular, global warming will make it harder to predict when torrential downpours will hit the northern hemisphere.

Weather forecasts have been , with forecasters both increasingly confident in their one and two-day forecasts and ever more able to forecast a week ahead, at least in a limited way. While it may be physically impossible to forecast more than two weeks ahead, within the window where forecasts are possible things have got better.

However, while we know the world is getting warmer because of our greenhouse gas emissions, almost nothing is known about how climate change will affect the predictability of the weather.

鈥淲hen you add energy to a system, you might intuitively expect it to become more turbulent and messier,鈥� says Gabriele Messori at Uppsala University in Sweden. This would imply less predictable weather in a warmer world. But Messori says physics does not always work in this intuitive way.

Saw that coming

To work out what will really happen, Messori and his student Sebastian Scher of Stockholm University in Sweden used a climate model to simulate two sets of conditions: the climate as it was from 1976-2005 and the climate as it could be in 2071-2100. Then they took snapshots from both models and fed them into a weather forecasting system to see if it could predict what would happen next.

In both cases they ran many forecasts to see how consistent they were. This 鈥渆nsemble forecasting鈥� allowed them to measure how predictable the weather was. If most of the forecasts were similar, the weather was predictable: if they were wildly different, it was unpredictable.

In the future climate, the most dramatic changes were in the northern hemisphere. The average temperature and air pressure became slightly more predictable. But rainfall and other precipitation became less predictable.

Scher emphasises that the increase in predictability of temperature and air pressure is 鈥渘ot very much鈥�. In the future climate, a forecast six-and-a-half days ahead would be as accurate as a six-day forecast today. 鈥淚 would say the main good news is that it is not becoming more difficult.鈥�

The wind of change

Scher and Messori say the reason for the increased predictability is that the Arctic is warming faster than the equator, so the temperature difference between the two is shrinking. 鈥淭his temperature difference is driving the weather in the mid-latitudes,鈥� says Scher. 鈥淭he warm air from the equator wants to get to the pole, and the cold air at the pole wants to get to the equator. It鈥檚 quite complicated what happens, and that鈥檚 what drives our weather.鈥� Reducing that temperature gradient means a less frantic churn of weather systems.

That makes sense, says Ken Mylne of the UK Met Office. 鈥淚t鈥檚 quite plausible that you鈥檇 see a reduction in ensemble spread, and therefore an increase in predictability to some degree.鈥�

This explains why weather predictability did not change much in the southern hemisphere. Antarctica is only warming slowly, so the temperature gradient is not changing as fast.

Meanwhile, if rainfall becomes less predictable that could cause problems for people who rely on knowing how much rain to expect over the coming week. Farmers and hydroelectric dam operators both need reliable information.

Mylne points out that the biggest change in predictability is in summer rainfall. Because this is driven more by convection above hot ground, instead of large-scale weather systems, it tends to cause sudden and , not widespread and long-lasting floods. If summer rainfall becomes harder to predict, flash floods may take us by surprise even more than they do now.

鈥淚 think the whole subject of how climate change affects the predictability of weather has not been explored very thoroughly,鈥� says Kerry Emanuel at the Massachusetts Institute of Technology. 鈥淎nd it is a consideration, because, of course, weather forecasts are very useful.鈥�

In 2017, Emanuel used climate models to argue that because they will often strengthen rapidly just before striking land. But there has been little follow-up, and he says we do not know how the predictability of other extreme events, such as hailstorms or tornadoes, will be affected.