First Paper to Link CO2 and Global Warming (1856)

Re:First Paper to Link CO2 and Global Warming (1856) by pjc50 on Monday October 14, 2019 @10:47AM (#21248846)

Surely "what would we say had gone wrong" depends entirely on .. what actually happens? This seems like a double counterfactual that's impossible to answer in advance.

Presuming we don't lose key information (e.g. deliberate erasure of the NOAA archives), we'd treat it in the same way as the current investigation; find what changes are required to the model to make it consistent with the data.

[ Parent | on HN ]

Re:First Paper to Link CO2 and Global Warming (1856) by agent008t on Monday October 14, 2019 @12:25PM (#21249984)

And yet it is very useful to know what could go wrong. And that applies to building all kinds of things - software, companies, models, research, systems, etc.

> we'd treat it in the same way as the current investigation; find what changes are required to the model to make it consistent with the data

If that is how it is currently done, I have a big issue with it. It seems like the model is not being treated as a forecasting model but rather as a purely academic one ('if we make these assumptions and have these initial conditions, how does the behavior of the system evolve'?), while being sold to the media, politicians and the general public as a forecasting model that is assumed to have very strong predictive power.

[ Parent | on HN ]