4 ms·
Maybe, in the model, the differences persisted but decreased (or increased?). Also, perhaps the model isn't given as much scientific weight as the author of th
by whatshisface 14y ago
Maybe, in the model, the differences persisted but decreased (or increased?).
Also, perhaps the model isn't given as much scientific weight as the author of the article indicated. (It's been something I have wondered about... given how hard it is for us to predict the weather, how can we possibly find long-term climate models useful?)
- Afforess 14y agoClimate models can be useful, especially when you use them to predict events that have already happened (e.g you have the data to either show the model is correctly forecasting the past event, or you have the data to show the model has gone awry.) Theoretically, if your model can predict past events correctly, you can then apply it to present and future events and get reasonable results. Theoretically... PS. Weather forecasting is pretty accurate these days. I'd take bets on 7 day forecasts from the NWS. Source: Meteorology degree.
- fatbob 14y agoThe model forecasting past events doesn't really tell you much about the accuracy of the model, even theoretically. It could be overfitting for example.
- youngerdryas 14y agoWithin 10 degrees F for 10 consecutive hours and precipitation to 1/4" in Boston for even money.