Abstract
The recent hype about artificial intelligence has sparked renewed interest in applying the successful deep learning (DL) methods for image recognition, speech recognition, robotics, strategic games and other application areas to the field of meteorology. There is some evidence that better weather forecasts can be produced by introducing big data mining and neural networks into the weather prediction workflow. Here, we discuss the question of whether it is possible to completely replace the current numerical weather models and data assimilation systems with DL approaches. This discussion entails a review of state-of-the-art machine learning concepts and their applicability to weather data with its pertinent statistical properties. We think that it is not inconceivable that numerical weather models may one day become obsolete, but a number of fundamental breakthroughs are needed before this goal comes into reach.This article is part of the theme issue ‘Machine learning for weather and climate modelling’.
Highlights
The history of numerical weather prediction (NWP) and that of machine learning (ML) or artificial intelligence differ substantially
In addition to the improvement of dynamical cores, further gains in accuracy have been achieved by fine-tuning physical parameterizations which are mandatory to represent atmospheric processes that cannot be captured by the grid-scale thermodynamics
We discussed the potential of modern deep learning (DL) approaches to develop purely data-driven end-to-end weather forecast applications
Summary
The history of numerical weather prediction (NWP) and that of machine learning (ML) or artificial intelligence (for the purposes of this paper, the two terms can be used interchangeably) differ substantially. NN, which would take observations as input and generate end-user forecast products directly from the data (figure 1, right column) Such approaches have been investigated for the specific application of wind speed and power predictions (cf [15]), and there is at least one study, which successfully developed an end-to-end workflow to forecast multiple weather variables from the data of the US weather balloon network [16]. Most of these studies were restricted to short-term forecasts and a few individual target sites, and none has yet attempted to explore the wealth of combined meteorological observations from the plethora of instruments and sensors, which is routinely used in operational weather forecasts. We hope that this article will lead to a better understanding between ‘machine learners’ and ‘weather researchers’ and contribute to a more effective development of DL solutions in the field of weather and climate
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
