I feel like many are unable to accept that perfectly good science is being done in this field.
I think many are unable to accept they've been duped.
No simulations perfectly keep track of all variables with the highest accuracy, that's computationally impossible.
Right. That's what sensitivity and residual analyses are for. Now go ask for those from a climate "scientist" and see what happens. Papers are routinely submitted "peer reviewed" and declared gospel for the acolytes without any of this data. In fact... If you require it for peer review, the Priesthood will boycott your journal. They will threaten your editor. And if that doesn't work, they'll try to get the editorial board fired. You should look into these things before you declare all this "good science in the field". Go read the emails. It's all there.
You seem to believe that if we can't model an input then the input doesn't contribute. That's bassackwards. We didn't have workable models for nuclear physics 1000 years ago. But the sun still shined did it not?
What matters is that you incorporate the most pertinent factors with a high enough degree of accuracy so that the simulation is able to accurately reproduce what's observed.
Would you consider energy balance "pertinent factor"? Do you not believe that clouds have a direct effect on energy into and out of the climate system? It's pretty obvious they do.
But who cares? Right? Several IPCC models don't balance heat correctly. So they have a bit more heat coming in than leaving. Does that sound like good modeling of a thermodynamic system to you?
There are a few very complicated climate models that do quite well in predicting temperatures and variations in climate over timescales of thousands to millions of years.
Huh? How do we know they "predict" temperatures thousands to millions of years in the future? Or do you mean they calibrate to a dataset of manipulated proxy data that is both sparse and extremely low resolution, and is often based on single trees to guess at an average temperature of the entire globe? Google Yamal.
And before you even make that claim they calibrate to those datasets correctly... you should read the HarryRead.Me file that was recovered in the Climategate I data dump.
Simulation of shorter timescales can be difficult, because more factors that affect variations come into play that are generally washed out over longer periods.
NNNOOOOOOOO! NO. No. NO.
Just because you don't have the resolution to capture events... does NOT mean the events don't happen. NOR does it mean they don't have an effect. It just means you're ignorant of them happening, and you're unaware of any contribution they provide.
Ignorance of a state variable is the most basic of all modeling errors. Not a "feature" of long time steps.
You're assuming that since we only have low resolution data, that only low resolution events contribute. That's silly. Not to mention arrogant.
Clearly short term discrete events have played a VERY significant role in our climate. The dinosaurs didn't die off over billions of years. They died damn quick. And we know the climate changed even quicker.
There are extremely fast transients to our climate. Anyone saying otherwise is lying. Take a look at historical datasets? Do they have a sampling period low to satisfy the Nyquist theorem? Not even a fraction!
However, we can be fairly confident in the predictions of long term trends that will occur due to the input of greenhouse gases.
Really? How so? Based on models that show increasing temperatures even when the CO2 inputs were negative? Look into MacEntyer's work. After he started his verification projects, all of the leaders in the field (UEA-CRU, UCAR, PENN) stopped releasing their methodologies, data, residual analyses. That's not science. That hiding.
How quickly these changes will manifest is not entirely known, though.
If a broker told you "I predict the stock market will go up by 5%. I'm just not sure when
." would you give him your money?
This post was edited on 4/7 at 1:55 am