Jovial Monk wrote on Aug 11
th, 2018 at 9:08pm:
ou are the only one babbling about parameters.
They used to need lots of parameters, assumptions. Partly because the computers were just too slow and limited to run the models. But they learned and tried new things and now we have models that accurately predict climate.
And they STILL use assumptions (parameters).
From AR5WG1_Chapter06_final.pdf -
"A challenge is that limitations of the satellite sensors demand various assumptions in order to convert a satellite measurement into a ‘model equivalent’ climate variable"
"An alternative approach is to calculate ‘observation-equivalents’ from models using radiative transfer calculations to simulate what the satellite would provide if the satellite system were ‘observing’ the model. This approach is usually referred to as an instrument simulator’. Microphysical assumptions (which differ from model to model)
can be included in the simulators, avoiding inconsistencies"
"In addition to mesoscale eddies, there has been a growing awareness of the role that sub-mesoscale eddies and fronts play in restratifying the mixed layer (Boccaletti et al., 2007; Fox-Kemper et al., 2008; Klein and Lapeyre, 2009), and the parameterization of Fox-Kemper et al. (2011) is now used in some CMIP5 models."
So we have a mixture of assumptions, simulations and parameterization. None of which exist in the real world.