There’s been a lot of recent discussion about weather forecasts and loss of accuracy due to the lack of airplanes flying and collecting meteorological data. There are often many contributing sources of data – airplanes being one of them – and depending on the modeling being used, that may or may not be a factor.
Airplanes have been gathering weather data for nearly a century and air travel substantially changed the way the National Weather Bureau, now the National Weather Service, created forecasts. In 1926, the Air Commerce Act directed the agency to provide weather services to all aircraft and this was the beginning of the invaluable connection between weather operations and aviation. During the early to mid-1900s, weather-observing technologies started to grow, and in 1934, the National Weather Bureau started collecting data from twenty daily aircraft observations. This program was ultimately deemed to be too risky and expensive and was replaced by the pilot-balloon program.
However in 1998, the idea of gathering weather data from planes came back to the forefront when the World Meteorological Organization created the Aircraft Meteorology Data Relay Panel. This program led to an automated system for gathering weather data from a variety of aircraft, including commercial, private and military planes. The program uses data that is gathered and transmitted by aircraft to supplement the data gathered by other meteorological instruments and to help improve the accuracy of weather forecasts. This program collects more than 700,000 weather observations of air temperature and wind speed and direction on an average day.
Today, with COVID-19 related travel restrictions and economic impact, The European Centre for Medium-Range Weather Forecasts reports an 80% drop in meteorological readings due to flight cancellations of commercial flights. This situation can have a slight impact on the skill of weather forecasts, as the aircraft provide wind and temperature observations at jet stream level, but this error reduces when the uncertainty of the atmosphere as a whole takes over after 24 hours.
However, some of these ‘data gaps’ can be filled by using a combination of increased weather balloon ascents and also sophisticated new satellites that can measure wind speeds through the depth of the lower atmosphere, so the overall impact on forecast quality isn’t expected to be significant.
Private weather companies, in particular, are using ensembles comprised of multiple models to create forecasts involving numerous sources, which ultimately rely less on the aircraft data. Those companies are looking at an array of global models run by large government agencies and combining them into a single forecast. Although some models may be missing data from fewer aircraft flights right now, there is data from satellite and balloon soundings available which is from similar heights and locations where aircraft fly, and offer an alternative look at what is occurring in the atmosphere.
When combining multiple models into an ensemble fashion, meteorologists are still able to deliver high-quality forecasts. Some private weather companies run their own high-resolution models using proprietary surface observation networks to get better forecasts in areas of the world where they have a large density of customers or difficult places to forecast, for example mountainous terrain or over near-shore areas. When a team of experienced and highly skilled meteorologists curate these forecasts, their experience and knowledge of local weather and how weather impacts customers, enables them to continue providing insights and highly accurate forecasts.
Businesses and consumers alike can still have confidence in weather forecasts and the meteorologists behind them, especially as we continue through the spring severe weather season.