The severe disruption to commercial air traffic caused by travel bans and national lockdowns imposed by countries to counter the Covid-19 pandemic has had an unexpected side-effect: a decline in the accuracy of weather forecasts. It is not generally realised that airliners collect and relay data on air pressure, air temperature, relative humidity and wind as the fly their routes, amounting to some 700 000 meteorological reports every day. Meteorological agencies then feed this data into their weather forecasts.
But the grounding of so many airliners owing to Covid-19 resulted in the loss of 50% to 75% of aircraft weather observations, worldwide, in the period from March to May this year. This has been revealed by a new study published in the American Geophysical Union’s Geophysical Research Letters. The loss of this data reduced the accuracy of weather forecasts and, the longer the term of the forecasts, the greater the inaccuracies became.
Study lead author and (UK) University of Lancaster Environment Centre senior research associate Dr Ying Chen pointed out that inaccurate weather forecasts could directly affect the economy. While the impact on agriculture was obvious, inaccurate weather forecasts could affect the stability of electrical grids. In particular, wind turbine operators needed accurate windspeed forecasts and electricity companies in general relied on temperature forecasts to be able to predict daily energy loads – would people switch on air conditioning or not?
“If this uncertainty goes over a threshold, it will introduce unstable voltage for the electrical grid,” he explained. “That could lead to a blackout, and I think this is the last thing we want to see in this pandemic.”
The parts of the globe that have been most affected were, with one exception, those that normally experienced heavy air traffic – Australia, south-east China and the US – plus remote regions such as Antarctica, Greenland and the Sahara Desert. The exception was Western Europe, where the accuracy of the weather forecasts was unaffected despite a decline in air traffic of 80% to 90%.
Chen suggested that this was because Western Europe had a very dense network of ground-based weather stations (numbering more than 1 500) and frequent balloon measurements, which filled the gap left by the aircraft. However, UK University of Exeter atmospheric science Professor Jim Haywood, who did not participate in the study, pointed out that Western European weather was notably stable during the March to May 2020 period, making accurate forecasts easier with less data. This could also have been a factor, along with the dense weather station network, in the maintenance of accuracy in European weather forecasts.
Important exceptions to this global decline in weather forecast accuracy were precipitation (rain and snow) forecasts. This was because these forecasts could depend on weather satellite observations.