In the future, weather forecasts that provide storm alerts and help us plan our daily lives can come up to five days before reaching the limits of numerical weather prediction, researchers say.
"The obvious question raised from the beginning of our entire area is, what is the ultimate limit for predicting the daily weather in the future," said Fuqing Zhang, professor of meteorology and atmospheric science and director of the Center for Advanced Data Assimilation and Predictability Techniques at Penn State. "We think we have found that limit and on average it is about two weeks."
Reliable forecasts are now possible nine to 1
The research confirms a long hypothesized predictability limit for weather forecasts first proposed in the 1960s by Edward Lorenz, a mathematician from the Massachusetts Institute of Technology, meteorologist and pioneer in chaos theory, researchers said.
"Edward Lorenz showed that one cannot predict the weather beyond any time horizon, even in principle" Kerry Emanuel, professor of atmospheric knowledge at MIT and co-author of the study. "Our research shows that this weather forecast horizon is about two weeks, remarkable near Lorenz's estimate."
Unpredictability in how weather develops means that even with perfect models and understanding of the original conditions, there is a limit to how far in advance accurate forecasts are possible, researchers say.
"We used state-of-the-art models to answer the most basic question," said Zhang, lead author of the study. "I think we will refine this answer in the future, but our study finally shows that there is a limit, but we still have a big room to improve the forecast before the limit is reached."
To test the boundary, Zhang and his team used the world's two most advanced numerical weather forecasting modeling systems – the European Center for Medium Range Weather Forecasts and the United States next-generation global predictive system.
They gave an almost perfect picture of the original conditions and tested how the models could recreate two real weather events, a cold rise in northern Europe and flood inducing rain in China. The simulations could predict the weather patterns with reasonable accuracy up to about two weeks, the researchers say.
Improvements to daily weather forecasts have implications for things like storm evacuation, energy supply, agriculture and wild fires.
"We have made significant progress in the weather forecast in recent decades, and we can predict weather five days in advance with great confidence now," Zhang said. "If we can predict further days of high confidence in the future, it would have great economic and social benefits."
Researchers said better data collection, algorithms for integrating data into models and improved computing power to run experiments are all needed In order to further improve our understanding of the original conditions.
"Achieving this additional predictability limit requires concerted efforts by the entire community to design better numerical weather models, improve observations, and better utilize advanced data assimilation and computing techniques observations," Zhang said.
Bridging the gap between radar meteorology / hydrology / technology and weather forecast
Fuqing Zhang et al., What is the predictability limit for midlatitude Weather ?, Journal of the Atmospheric Sciences (2019). DOI: 10,1175 / JAS-D-18-0269.1
Predictability limit: Researchers find limits for weather forecasts (2019, April 15)
downloaded April 16, 2019
This document is subject to copyright. Except for any fair trade for private study or research, no
Some may be reproduced without written permission. The content is provided for informational purposes only.