Because of the huge interest in the paper: “How much CO2 really contributes to global warming? Spectroscopic studies and modelling of the influence of H2O, CO2 and CH4 on our climate” by Herman Harde which comes up with an estimated climate warming of 0.45C (rather than the IPCC 1-1.1C for direct warming or 2-5C with positive feedback), I’ve delved further into the subject to see if I can make more sense of it.
The first clue in understanding this paper was to find that the author Hermann Harde has worked on A New Tool (link) utilising the HITRAN database. This tool uses the database to Compute and Display the Spectra of all relevant gases in the atmosphere with a high resolution over a wide spectral range. So, this appears to be his area of expertise: modelling gas absorption .
So, it seems this author has developed a new tool for assessing the absorption spectrum of atmospheric gases and, it would seem, he decided to see what this new tool would give if it were applied to the problem of CO2 induced warming. But, that still doesn’t explain why this paper is so different to other estimates of warming and produces such a dramatically different result. The next clue is in the first paragraph of the paper:
Based on the actual HITRAN’2008 database  detailed spectroscopic studies on the absorbance of the greenhouse gases water, carbon dioxide and methane in the atmosphere are presented. The line-by-line calculations for sun light … (cont)
To understand this we need to know more about the Hitran 2008 database. At first I thought this was just a database of absorption spectra. And “line-by-line” was a poor translation of “wavelength by wavelength”. But on closer reading of the details of HITRAN2008 it is a lot more complex:
(HITRAN2008) … compilation consists of several parts: (1) the traditional high-resolution, line-by-line portion where fundamental spectroscopic parameters required for calculation of radiative transfer are archived; (2) files of infrared cross-sections primarily for large or heavy polyatomic molecules; (3) UV line-by-line parameters and cross-sections; (4) tables of aerosol refractive indices; and (5) generalized tables and references that relate to HITRAN. (The HITRAN 2008 Molecular Spectroscopic Database)
Immediately it is obvious this is no simple database of spectrum. What the line-by-line refers to is a spectral model where each absorption mode or line is modelled together with detailed curve describing how that single frequency is broadened by other factors such as temperature and presumably pressure. So, this is not an empirically derived continuous curve derived by measuring the absorption spectrum, but a model of the behaviour of the gas using the details of the spectral lines on top of which is added an individual curve for each line describing the way that individual line broadens. This is clearly a highly complex model and it’s worth repeating what they say about it:
In addition to adding some new molecules, many vibration-rotation bands for the previously included species have been updated or extended. One can highlight the vast improvement for H2O, CO2, O3, CH4, O2, and most of the trace-gas species. Emphasis has been on increased accuracy and completeness of line positions, intensities, and line-shape parameters.
Now, that we understand that Hermann Harde had been part of a group creating a brand new tool to utilise the line-by-line spectral model of gases based on the HITRAN2008 database, which provides considerable improvement in the analysis of trace atmospheric gases it is now easier to understand what is new about this model:
The line-by-line calculations for sun light from 0.1 – 8 um (short wavelength radiation) as well as those for the emitted earth radiation from 3 – 60 um (long wavelength radiation) show, that due to the strong overlap of the CO2 and CH4 spectra with the water vapour lines the influence of these gases is significantly reducing with increasing water vapour pressure, and that with increasing CO2-concentration well noticeable saturation effects are observed
limiting substantially the impact of CO2 on the warm-up of the atmosphere.
From this investigation, the main reason for a difference in calculated warming appears to be the more sophisticated modelling of trace gas absorption using the improved ability of the HITRAN database.Moreover as they also model the effect of temperature and pressure(?), the absorption spectra will change for each layer of the atmosphere being modelled even if they contain the same gas proportions.
However, whilst the paper mentions cloud and H2O, there is no explicit mention of feedback such as additional atmospheric H2O and there is no reason to suggest from the known background of the Author that he would have incorporated such feedback which would seem to be outside his field of expertise.
In order to avoid complicating the issue unduly I have used the single term: “absorption spectrum” where strictly speaking I should say “and emission spectrum and other effects”, such as changing refractive index.
So the take home message on this seems to be that the previous versions of HITRAN did not have sufficient resolution, or neglected some important vib/rot bands. I would assume of course that previous models already took band overlap into account so that there would be no double counting of absorption. This paper seems to be saying that better resolution of the absorption bands shows more saturation than previously used (presumably due to sharper resolution. I always thought that the resolution was adequate, and like you I was always under the impression that the spectroscopy side of it was rock solid.
Terry, thanks for the post, with all the searching that hadn’t quite twigged about the resolution. As you say the added precision means that very sharp dips at particular frequency are not smoothed out so they are wide-shallow.
The difference is that you can take out a lot of wide-shallow bites from a spectrum and because each is shallow they (appear) not to bottom out leaving room for further bites, but whilst the size of the area under the bite is the same for a narrow-deep hole, you can take a lot less narrow-deep holes because whilst the bite is narrower, the depth means it quickly bottoms out.
Indeed. The sharper the line res the less that enhanced absorption can take place in the wings due to pressure and temperature broadening. This always means saturation is achieved quicker. Now I dont know that this is the case with this work or not, and I guess that was my query. It will be interesting to see what Eli rabbet has to say about it, since his main field is spectroscopy. Mine is very rusty.
Terry, the enormity of this hit me this morning. Climategate said: “the people aren’t exactly professional” but it did nothing to change the predicted scale. This puny little paper, even at this very advanced stage will wipe out billions worldwide from the value of renewable energy and carbon trading companies.
In fact if I had any money myself I’d be buying into coal and coal electricity producers because they have just become a hell of a lot more attractive as investments.
The stock markets would be foolish not to take this very very seriously!
I suggest you read this paper http://climategate.nl/wp-content/uploads/2011/02/CO2_and_climate_v7.pdf which was the basis of a presentation to KNMI by Dr(Ir) Noor Van Andell. He is saying from look at real climate data that CO2 leads to cooling and not warming. The HITRAN data is irrelevent to what actually occurs.