News for hackers. Stuff that matters.

A comment on Warning of forthcoming collapse of Atlantic meridional overturning circulation. Parent comment

Re:Warning of forthcoming collapse of Atlantic meridional ov... by sieste on Tuesday July 25, 2023 @01:15PM (#36865768)
> physical models are the most reliable basis for prediction

Simpler models can yield better predictions than complex ones, even when the complex model is more "realistic". The many tunable parameters and complex feedback loops can increase uncertainty compared to a simple model.

See for example "To explain or to predict?" by Shmueli (2010, https://doi.org/10.1214/10-STS330)

[ Parent | on HN ]
Re:Warning of forthcoming collapse of Atlantic meridional ov... by braaannigan on Tuesday July 25, 2023 @01:29PM (#36866040)
Many of the physical models are of similar complexity to the model in this paper - the sort of analytical model that you run in a few minutes on your laptop, not on a super-computer. It's not a question of complexity it's about whether a statistical model that is not constrained by well-understood constraints is a high-value model
[ Parent | on HN ]
Re:Warning of forthcoming collapse of Atlantic meridional ov... by sieste on Tuesday July 25, 2023 @01:38PM (#36866199)
Agreed. Also I'm not here to defend their particular model. Just saying that "physical" does not always equal "better predictions".

Their model equation (1) has 4 parameters. Do you have any papers at hand that model AMOC in a more physically constrained way, with a similar number of tunable parameters?

[ Parent | on HN ]