M-ENIAC: A machine learning recreation of the first successful numerical weather forecasts
This work offers a novel approach for solving weather prediction equations, potentially benefiting meteorologists and climate scientists, though it is incremental as it applies an existing ML method to a historical problem.
The authors tackled the problem of recreating the first numerical weather forecasts using modern machine learning instead of traditional numerical methods, showing that physics-informed neural networks provide easier and more accurate solutions for meteorological equations on the sphere compared to the original ENIAC solver.
In 1950 the first successful numerical weather forecast was obtained by solving the barotropic vorticity equation using the Electronic Numerical Integrator and Computer (ENIAC), which marked the beginning of the age of numerical weather prediction. Here, we ask the question of how these numerical forecasts would have turned out, if machine learning based solvers had been used instead of standard numerical discretizations. Specifically, we recreate these numerical forecasts using physics-informed neural networks. We show that physics-informed neural networks provide an easier and more accurate methodology for solving meteorological equations on the sphere, as compared to the ENIAC solver.