Further Feedback on the Frank (2019) “Propagation of Error” Paper
From Dr Roy Spencer’s Weblog
September 12th, 2019 by Roy W. Spencer, Ph. D.
NOTE: This put up has undergone just a few revisions as I attempt to be extra exact in my wording. The newest revision was at 0900 CDT Sept. 12, 2019.
If this put up is re-posted elsewhere, I ask that the above time stamp be included.
Yesterday I posted an prolonged and significant evaluation of Dr. Pat Frank’s latest publication entitled Propagation of Error and the Reliability of World Air Temperature Projections. Dr. Frank graciously offered rebuttals to my factors, none of which have modified my thoughts on the matter. I’ve made it clear that I don’t belief local weather fashions’ long-term forecasts, however that’s for various causes than Pat supplies in his paper.
What follows is the crux of my primary drawback with the paper, which I’ve distilled to its essence, under. I’ve averted my earlier mistake of paraphrasing Pat, and as an alternative I’ll quote his conclusions verbatim.
In his Conclusions part, Pat states “As famous above, a GCM simulation might be in good exterior vitality stability on the TOA whereas nonetheless expressing an incorrect inside local weather energy-state.”
This I agree with, and I consider local weather modelers have admitted to this as nicely.
However, he then additional states, “LWCF [longwave cloud forcing] calibration error is +/- 144 x bigger than the annual common enhance in GHG forcing. This truth alone makes any attainable international impact of anthropogenic CO2 emissions invisible to current local weather fashions.”
Whereas I agree with the primary sentence, I completely disagree with the second. Collectively, they symbolize a non sequitur. All the fashions present the impact of anthropogenic CO2 emissions, regardless of recognized errors in elements of their vitality fluxes (akin to clouds)!
Why?
If a mannequin has been compelled to be in international vitality stability, then vitality flux part biases have been cancelled out, as evidenced by the management runs of the assorted local weather fashions of their LW (longwave infrared) conduct:
Determine 1. Yearly- and global-average longwave infrared vitality flux variations at top-of-atmosphere from 10 CMIP5 local weather fashions within the first 100 years of their pre-industrial “management runs”. Knowledge out there from https://climexp.knmi.nl/
Importantly, this forced-balancing of the worldwide vitality price range just isn’t carried out at each mannequin time step, or yearly, or each 10 years. If that was the case, I might agree with Dr. Frank that the fashions are ineffective, and for the rationale he offers. As an alternative, it’s carried out as soon as, for the typical conduct of the mannequin over multi-century pre-industrial management runs, like these in Fig. 1.
The ~20 completely different fashions from world wide cowl a WIDE number of errors within the part vitality fluxes, as Dr. Frank reveals in his paper, but all of them mainly behave the identical of their temperature projections for a similar (1) local weather sensitivity and (2) charge of ocean warmth uptake in response to anthropogenic greenhouse gasoline emissions.
Thus, the fashions themselves display that their international warming forecasts don’t rely on these bias errors within the elements of the vitality fluxes (akin to international cloud cowl) as claimed by Dr. Frank (above).
That’s partly why completely different modeling teams world wide construct their very own local weather fashions: to allow them to take a look at the impression of various assumptions on the fashions’ temperature forecasts.
Statistical modelling assumptions and error evaluation don’t change this truth. A local weather mannequin (like a climate forecast mannequin) has time-dependent differential equations protecting dynamics, thermodynamics, radiation, and vitality conversion processes. There are bodily constraints in these fashions that result in internally compensating behaviors. There isn’t any approach to symbolize this conduct with a easy statistical evaluation.
Once more, I’m not defending present local weather fashions’ projections of future temperatures. I’m saying that errors in these projections should not resulting from what Dr. Frank has offered. They’re primarily as a result of processes controlling local weather sensitivity (and the speed of ocean warmth uptake). And local weather sensitivity, in flip, is a operate of (for instance) how clouds change with warming, and apparently not a operate of errors in a selected mannequin’s common cloud quantity, as Dr. Frank claims.
The same conduct of the wide range of various fashions with differing errors is proof of that. All of them reply to rising greenhouse gases, opposite to the claims of the paper.
The above represents the crux of my primary objection to Dr. Frank’s paper. I’ve quoted his conclusions, and defined why I disagree. If he needs to dispute my reasoning, I might request that he, in flip, quote what I’ve stated above and why he disagrees with me.
Like this:
Loading…
