by Judith Curry
The calculation of atmospheric radiative fluxes is central to any argument related to the atmospheric greenhouse/Tyndall gas effect. Atmospheric radiative transfer models rank among the most robust components of climate model, in terms of having a rigorous theoretical foundation and extensive experimental validation both in the laboratory and from field measurements. However, I have not found much in the way of actually explaining how atmospheric radiative transfer models work and why we should have confidence in them (at the level of technical blogospheric discourse). In this post, I lay out some of the topics that I think need to be addressed in such an explanation regarding infrared radiative transfer. Given my limited time this week, I mainly frame the problem here and provide some information to start a dialogue on this topic, I hope that other experts participating can fill in (and I will update the main post).
Atmospheric radiative transfer models
The Wikipedia provides a succint description of radiative transfer models:
An atmospheric radiative transfer model calculates radiative transfer of electromagnetic radiation through a planetary atmosphere, such as the Earth’s. At the core of a radiative transfer model lies the radiative transfer equation that is numerically solved using a solver such as a discrete ordinate method or a Monte Carlo method. The radiative transfer equation is a monochromatic equation to calculate radiance in a single layer of the Earth’s atmosphere. To calculate the radiance for a spectral region with a finite width (e.g., to estimate the Earth’s energy budget or simulate an instrument response), one has to integrate this over a band of frequencies (or wavelengths). The most exact way to do this is to loop through the frequencies of interest, and for each frequency, calculate the radiance at this frequency. For this, one needs to calculate the contribution of each spectral line for all molecules in the atmospheric layer; this is called a line-by-line calculation. A faster but more approximate method is a band transmission. Here, the transmission in a region in a band is characterised by a set of coefficients (depending on temperature and other parameters). In addition, models may consider scattering from molecules or particles, as well as polarisation.
If you don’t already have a pretty good understanding of this, the Wikipedia article is not going to help much. There are a few good blog posts that I’ve spotted that explain aspects of this (notably scienceofdoom):
- Theory and experiment – atmospheric radiation
- CO2: An insignificant trace gas Part III
- CO2: An insignificant trace gas Part IV
You find scienceofdoom’s treatments to be beyond your capability to understand? Lets try more of a verification and validation approach to assessing whether we should have confidence in the radiation transfer codes used in climate models.
History of atmospheric (infrared) radiative transfer modeling
I don’t recall ever coming across a history on this subject? Here are a few pieces of that history that I know of (I hope that others can fill in the holes in this informal history).
Focusing on infrared radiative transfer, there is some historical background in the famous Manabe and Wetherald 1967 paper on early calculations of infrared radiative transfer in the atmosphere. As a graduate student in the 1970’s, I recall using the Ellsasser radiation chart.
The first attempt to put a sophisticated radiative transfer model into a climate model was made by Fels and Kaplan 1975, who used a model that divided the infrared spectrum into 19 bands. I lived a little piece of this history, when I joined Kaplan’s research group in 1975 as a graduate student.
In the 1980’s, band models began to be incorporated routinely in climate models. An international program of Intercomparison of Radiation Codes in Climate Models (ICRCCM) was inaugurated for clear sky infrared radiative transfer, with results described by Ellingson et al. 1991 and Fels et al. 1991 (note Andy Lacis is a coauthor):
During the past 6 years, results of calculations from such radiation codes have been compared with each other, with results from line-by-line models and with observations from within the atmosphere. Line by line models tend to agree with each other to within 1%; however, the intercomparison shows a spread of 10-20% in the calculations by less detailed climate model codes. When outliers are removed, the agreement between narrow band models and the line-by-line models is about 2% for fluxes.
Validation and improvement of atmospheric radiative transfer models
In 1990, the U.S. Department of Energy initiated the Atmospheric Radiation Measurement Program (ARM) program targeted at improving the understanding of the role and representation of atmospheric radiative processes and clouds in models of the earth’s climate (see here for a history).
A recent summary of the objectives and accomplishments is provided in the 2004 ARM Science Plan. A list of measurements (and instruments) made by ARM at its sites in the tropics, midlatitudes and the arctic is very comprehensive. Of particular relevance to evaluating infrared radiative transfer codes is the Atmospheric Emitted Radiance Interferometer. For this of you who want empirical validation, the ARM program provides this in spades.
The ARM measurements have become the gold standard for validating radiative transfer models. For line-by-line models, see this closure experiment described by Turner et al. 2004 (press release version here). More recently, see this evaluation of the far infrared part of the spectra by Kratz et al. (note: Miskolczi is a coauthor).
For a band model (used in various climate models), see this evaluation of the RRTM code:
Mlawer, E.J., S.J. Taubman, P.D. Brown, M.J. Iacono and S.A. Clough: RRTM, a validated correlated-k model for the longwave. J. Geophys. Res., 102, 16,663-16,682, 1997 link
This paper is unfortunately behind paywall, but it provides an excellent example of the validation methodology.
The most recent intercomparison of climate model radiative transfer codes against line-by-line calculations is described by Collins et al. in the context of radiative forcing.
There is a new international program (the successor to ICRCCM) called the Continual Comparison of Radiation Codes (CIRC), which established benchmark observational case studies and coordinates intercomparison of models.
The problem of infrared atmospheric radiative transfer (clear sky, no clouds or aerosols) is regarded as a solved problem (with minimal uncertainties), in terms of the benchmark line-by-line calculations. Deficiencies in some of the radiation codes used in certain climate models have been identified, and these should be addressed if these models are to be included in the multi-model ensemble analysis.
The greater challenges lie in modeling radiative transfer in an atmosphere with clouds and aerosols, although these challenges are greater for modeling solar radiation fluxes than for infrared fluxes. The infrared radiative properties of liquid clouds are well known; some complexities are introduced for ice crystals owing to their irregular shapes (this issue is much more of a problem for solar radiative transfer than for infrared radiative transfer). Aerosols are a relatively minor factor in infrared radiative transfer owing to the typically small size of aerosol particles.
However, if you can specify the relevant conditions in the atmosphere that provide inputs to the radiative transfer model, you should be able to make accurate calculations using state-of-the art models. The challenge for climate models is in correctly simulating the variations in atmospheric profiles of temperature, water vapor, ozone (and other variable trace gases), clouds and aerosols.
And finally, for calculations of the direct radiative forcing associated with doubling CO2, atmospheric radiative transfer models are more than capable of addressing this issue (this will be the topic of the next greenhouse post).
Note: this is a technical thread, and comments will be moderated for relevance.