3 ms·
There's a lot of excitement in this space but a lot of weather centers are really struggling to get even a tiny fraction of the resources and talent needed to s
by zulban 2y ago
There's a lot of excitement in this space but a lot of weather centers are really struggling to get even a tiny fraction of the resources and talent needed to start using these results in real operational systems. Everyone is still using fortran physics code from the 90s.
- aeroman 2y agoOne thing I found really interesting abou the Graphcast paper (I appreciate this is not graphcast, but I think it is still relevant) is that it doesn't understand climate change. The model requires the training data to be recent to get the best quality projections. While there are some factors that influence predictability in the weather forecast, as the fortran code is based on physics (at least in a broad sense), it doesn't suffer from those issues in the same way. This doesn't mean that the ML forecasts are wrong (obviously), just different. Given the relative computational simplicity of running them, I wonder if the issue is not just expertise, but also understanding how they can best be used to generate reliable weather forecasts?
- kristopolous 2y agoBut IBM owns weather.com. Oh wait, that's right. IBM owns weather.com.
- jeffbee 2y agoThere's nothing wrong with fortran and doing computational fluid dynamics in fortran is a mature field of endeavor.
- convolvatron 2y agothis is true. and the extreme amount of experience and tuning required to run these models is really not appreciated outside scientific computing but you have to admit, they are really gross to work with