Warning against excessive hubris, Harries says "we must exercise great caution over the true depth of our understanding, and our ability to forecast future climate trends."  As an example, he states that our knowledge of high cirrus clouds is very poor, noting that "we could easily have uncertainties of many tens of W m^-2 in our description of the radiative effect of such clouds, and how these properties may change under climate forcing."  This state of affairs is extremely disconcerting, especially in light of the fact that the radiative effect of a doubling the air's CO2 content is in the lower single-digit range of W m^-2, and, to quote Harries, that "uncertainties as large as, or larger than, the doubled CO2 forcing could easily exist in our modeling of future climate trends, due to uncertainties in the feedback processes."  Furthermore, because of the vast complexity of the subject, Harries rightly declares that "even if [our] understanding were perfect, our ability to describe the system sufficiently well in even the largest computer models is a problem."

Illustrative of a related problem is the work of Zender (1999), who characterized the spectral, vertical, regional and seasonal atmospheric heating caused by the oxygen collision pairs O2 ^. O2 and O2 ^. N2, which had earlier been discovered to absorb a small but significant fraction of the globally-incident solar radiation.  This work revealed that these molecular collisions lead to the absorption of about 1 Wm^-2 of solar radiation, globally and annually averaged.  This discovery, in Zender's words, "alters the long-standing view that H2O, O3, O2, CO2 and NO2 are the only significant gaseous solar absorbers in Earth's atmosphere," and he suggests that the phenomenon "should therefore be included in ... large-scale atmospheric models used to simulate climate and climate change."  It also raises the possibility there are still other yet-to-be-discovered processes that should be included in the models that are used to simulate earth's climate, and that until we are confident there is little likelihood of further such surprises, we ought not rely too heavily on what the models of today are telling us about the climate of tomorrow.

In another revealing study, Wild (1999) compared the observed amount of solar radiation absorbed in the atmosphere over equatorial Africa with what was predicted by three general circulation models of the atmosphere, finding that the model predictions were much too small.  Indeed, regional and seasonal model underestimation biases were as high as 30 Wm^-2, primarily because the models failed to properly account for spatial and temporal variations in atmospheric aerosol concentrations.  In addition, Wild found that the models likely underestimated the amount of solar radiation absorbed by water vapor and clouds.

Similar large model underestimations were discovered by Wild and Ohmura (1999), who analyzed a comprehensive observational dataset consisting of solar radiation fluxes measured at 720 sites across the earth's surface and corresponding top-of-the-atmosphere locations to assess the true amount of solar radiation absorbed within the atmosphere.  These results were compared with estimates of solar radiation absorption derived from four atmospheric general circulation models (GCMs); and, again, it was shown that "GCM atmospheres are generally too transparent for solar radiation," as they produce a rather substantial mean error close to 20% below actual observations.

Another solar-related deficiency of state-of-the-art GCMs is their failure to properly account for solar-driven variations in earth-atmosphere processes that operate over a range of timescales extending from the 11-year solar cycle to century- and millennial-scale cycles (see several of the subheadings under Solar Effects in our Subject Index).  Although the absolute solar flux variations associated with these phenomena are rather small, there are a number of "multiplier effects" that may significantly amplify their impacts.

According to Chambers et al. (1999), most of the many nonlinear responses to solar activity variability are inadequately represented (in fact, they are essentially ignored) in the global climate models used by the Intergovernmental Panel on Climate Change (IPCC) to predict future greenhouse gas-induced global warming, while at the same time other amplifier effects are used to model past glacial/interglacial cycles and even the hypothesized CO2-induced warming of the future, where CO2 is not the major cause of the predicted temperature increase but rather an initial perturber of the climate system that according to the IPCC sets other more powerful forces in motion that produce the bulk of the ultimate warming.  Hence, there appears to be a double standard within the climate modeling community that may best be described as an inherent reluctance to deal even-handedly with different aspects of climate change.  When multiplier effects suit their purposes, they use them; but when they don't suit their purposes, they don't use them.

In setting the stage for the next study of climate model inadequacies related to radiative forcing, Ghan et al. (2001) state that "present-day radiative forcing by anthropogenic greenhouse gases is estimated to be 2.1 to 2.8 Wm^-2; the direct forcing by anthropogenic aerosols is estimated to be -0.3 to -1.5 Wm^-2, while the indirect forcing by anthropogenic aerosols is estimated to be 0 to -1.5 Wm^-2," so that "estimates of the total global mean present-day anthropogenic forcing range from 3 Wm^-2 to -1 Wm^-2," which implies a climate change somewhere between a modest warming and a slight cooling, which would seem to be a rather shaky justification for mandating draconian measures to combat the first of these possibilities.  Hence, they say that clearly "the great uncertainty in the radiative forcing must be reduced if the observed climate record is to be reconciled with model predictions and if estimates of future climate change are to be useful in formulating emission policies."

Pursuit of this goal, as they describe it, requires achieving "profound reductions in the uncertainties of direct and indirect forcing by anthropogenic aerosols," which is what they set out to do in their analysis of the situation, which consisted of "a combination of process studies designed to improve understanding of the key processes involved in the forcing, closure experiments designed to evaluate that understanding, and integrated models that treat all of the necessary processes together and estimate the forcing."  At the conclusion of this laborious set of operations, Ghan et al. came up with some numbers that considerably reduced the range of uncertainty in the "total global mean present-day anthropogenic forcing," but that still implied a set of climate changes stretching from a small cooling to a modest warming.  Hence, they provided a long list of other things that must be done in order to obtain a more definitive result, after which they acknowledged that even this list "is hardly complete."  In fact, they concluded their analysis by saying "one could easily add the usual list of uncertainties in the representation of clouds, etc."  Consequently, the bottom line, in their words, is that "much remains to be done before the estimates are reliable enough to base energy policy decisions upon," to which we add a loud Amen!

Also studying the aerosol-induced radiative forcing of climate were Vogelmann et al. (2003), who report that "mineral aerosols have complex, highly varied optical properties that, for equal loadings, can cause differences in the surface IR flux between 7 and 25 Wm^-2 (Sokolik et al., 1998)," but who say that "only a few large-scale climate models currently consider aerosol IR effects (e.g., Tegen et al., 1996; Jacobson, 2001) despite their potentially large forcing."  Because of these facts, and in an attempt to persuade climate modelers to rectify the situation, Vogelmann et al. used high-resolution spectra to calculate the surface IR radiative forcing created by aerosols encountered in the outflow of air from northeastern Asia, based on measurements made by the Marine-Atmospheric Emitted Radiance Interferometer aboard the NOAA Ship Ronald H. Brown during the Aerosol Characterization Experiment-Asia.  In doing so, they determined, in their words, that "daytime surface IR forcings are often a few Wm^-2 and can reach almost 10 Wm^-2 for large aerosol loadings," which values they say "are comparable to or larger than the 1 to 2 Wm^-2 change in the globally averaged surface IR forcing caused by greenhouse gas increases since pre-industrial times."  In a massive understatement of fact, the researchers thus concluded that their results "highlight the importance of aerosol IR forcing which should be included in climate model simulations," causing us to wonder that if a forcing of this magnitude is not included in current state-of-the-art climate models, what other major forcings are they ignoring?

Shifting gears just a bit, two papers published one year earlier in the same issue of Science (Chen et al., 2002; Wielicki et al., 2002) revealed what Hartmann (2002) called a pair of "tropical surprises."  The first of the seminal discoveries was the common finding of both groups of researchers that the amount of thermal radiation emitted to space at the top of the tropical atmosphere increased by about 4 Wm^-2 between the 1980s and the 1990s, while the second was that the amount of reflected sunlight decreased by 1 to 2 Wm^-2 over the same period, with the net result that more total radiant energy exited the tropics in the latter decade.  In addition, the measured thermal radiative energy loss at the top of the tropical atmosphere was of the same magnitude as the thermal radiative energy gain that is generally predicted to result from an instantaneous doubling of the air's CO2 content.  Yet as Hartman correctly notes, "only very small changes in average tropical surface temperature were observed during this time."  So what went wrong?  Or, as we probably more correctly should phrase the question, what went right?

One thing was the change in solar radiation reception that was driven by changes in cloud cover, which allowed more solar radiation to reach the surface of the earth's tropical region and warm it.  These changes were produced by what Chen et al. determined to be "a decadal-time-scale strengthening of the tropical Hadley and Walker circulations."  Another helping-hand was likely provided by the past quarter-century's slowdown in the meridional overturning circulation of the upper 100 to 400 meters of the tropical Pacific Ocean (McPhaden and Zhang, 2002), which circulation slowdown also promotes tropical sea surface warming by reducing the rate-of-supply of relatively colder water to the region of equatorial upwelling.

So what do these observations have to do with evaluating the ability of climate models to correctly predict the future?  For one thing, they provide several new phenomena for the models to replicate as a test of their ability to properly represent the real-world.  In the words of McPhaden and Zhang, the time-varying meridional overturning circulation of the upper Pacific Ocean provides "an important dynamical constraint for model studies that attempt to simulate recent observed decadal changes in the Pacific."  If the climate models can't reconstruct this simple wind-driven circulation, for example, why should we believe anything else they tell us?

In an eye-opening application of this principle, Wielicki et al. tested the ability of four state-of-the-art climate models and one weather assimilation model to reproduce the observed decadal changes in top-of-the-atmosphere thermal and solar radiative energy fluxes that occurred over the past two decades.  The results were truly pathetic.  No significant decadal variability was exhibited by any of the models; and they all failed to reproduce even the cyclical seasonal change in tropical albedo.  The administrators of the test thus kindly concluded that "the missing variability in the models highlights the critical need to improve cloud modeling in the tropics so that prediction of tropical climate on interannual and decadal time scales can be improved."  Hartmann, on the other hand, was considerably more candid in his scoring of the test, saying that the results indicated "the models are deficient."  Expanding on this assessment, he further noted that "if the energy budget can vary substantially in the absence of obvious forcing," as it did over the past two decades, "then the climate of earth has modes of variability that are not yet fully understood and cannot yet be accurately represented in climate models," which leads us to wonder why anyone would put any faith in them.  To do so is simply illogical.

Also concentrating on the tropics, Bellon et al. (2003) note that "observed tropical sea-surface temperatures (SSTs) exhibit a maximum around 30°C," and that "this maximum appears to be robust on various timescales, from intraseasonal to millennial."  Hence, they say that "identifying the stabilizing feedback(s) that help(s) maintain this threshold is essential in order to understand how the tropical climate reacts to an external perturbation," which knowledge is needed for understanding how the global climate reacts to perturbations such as those produced by solar variability and the ongoing rise in the air's CO2 content.  This contention is further substantiated by the study of Pierrehumbert (1995), which "clearly demonstrates," in the words of Bellon et al., "that the tropical climate is not determined locally, but globally."  Also, they note that Pierrehumbert's work demonstrates that interactions between moist and dry regions are an essential part of tropical climate stability, which hearkens back to the adaptive infrared iris concept of Lindzen et al. (2001).

Noting that previous box models of tropical climate have shown it to be rather sensitive to the relative areas of moist and dry regions of the tropics, Bellon et al. analyzed various feedbacks associated with this sensitivity in a four-box model of the tropical climate "to show how they modulate the response of the tropical temperature to a radiative perturbation."  In addition, they investigated the influence of the model's surface-wind parameterization in an attempt to shed further light on the nature of the underlying feedbacks that help define the global climate system that is responsible for the tropical climate observations of constrained maximum SSTs.

Bellon et al.'s work, as they describe it, "suggests the presence of an important and as-yet-unexplored feedback in earth's tropical climate, that could contribute to maintain the 'lid' on tropical SSTs," much like the adaptive infrared iris concept of Lindzen et al. does.  They also say that the demonstrated "dependence of the surface wind on the large-scale circulation has an important effect on the sensitivity of the tropical system," specifically stating that "this dependence reduces significantly the SST sensitivity to radiative perturbations by enhancing the evaporation feedback," which injects more heat into the atmosphere and allows the atmospheric circulation to export more energy to the subtropical free troposphere, where it can be radiated to space.  Clearly, therefore, the case is not closed on either the source or the significance of the maximum "allowable" SSTs of tropical regions; and, hence, neither is the case closed on the degree to which the planet may warm in response to continued increases in the atmospheric concentrations of carbon dioxide and other greenhouse gases, in stark contrast to what is suggested by the climate models promoted by the IPCC.

In conclusion, there appear to be a number of major inadequacies in the ways in which several aspects of earth's radiative energy balance are treated in contemporary general circulation models of the atmosphere, as well as numerous other telling inadequacies stemming from the non-treatment of pertinent phenomena that are nowhere to be found in the models.  Hence, there is no rational basis for any of the IPCC-inspired predictions of catastrophic climatic changes due to continued anthropogenic CO2 emissions.  The scary scenarios they promulgate are simply unwarranted projections that have far outpaced what can be soundly supported by the current state of the climate modeling enterprise.

References
Bellon, G., Le Treut, H. and Ghil, M.  2003.  Large-scale and evaporation-wind feedbacks in a box model of the tropical climate.  Geophysical Research Letters 30: 10.1029/2003GL017895.

Chambers, F.M., Ogle, M.I. and Blackford, J.J.  1999.  Palaeoenvironmental evidence for solar forcing of Holocene climate: linkages to solar science.  Progress in Physical Geography 23: 181-204.

Chen, J., Carlson, B.E. and Del Genio, A.D.  2002.  Evidence for strengthening of the tropical general circulation in the 1990s.  Science 295: 838-841.

Ghan, S.J., Easter, R.C., Chapman, E.G., Abdul-Razzak, H., Zhang, Y., Leung, L.R., Laulainen, N.S., Saylor, R.D. and Zaveri, R.A.  2001.  A physically based estimate of radiative forcing by anthropogenic sulfate aerosol.  Journal of Geophysical Research 106: 5279-5293.

Harries, J.E.  2000.  Physics of the earth's radiative energy balance.  Contemporary Physics 41: 309-322.

Hartmann, D.L.  2002.  Tropical surprises.  Science 295: 811-812.

Jacobson, M.Z.  2001.  Global direct radiative forcing due to multicomponent anthropogenic and natural aerosols.  Journal of Geophysical Research 106: 1551-1568.

Lindzen, R.S., Chou, M.-D. and Hou, A.Y.  2001.  Does the earth have an adaptive infrared iris?  Bulletin of the American Meteorological Society 82: 417-432.

McPhaden, M.J. and Zhang, D.  2002.  Slowdown of the meridional overturning circulation in the upper Pacific Ocean.  Nature 415: 603-608.

Pierrehumbert, R.T.  1995.  Thermostats, radiator fins, and the local runaway greenhouse.  Journal of the Atmospheric Sciences 52: 1784-1806.

Sokolik, I.N., Toon, O.B. and Bergstrom, R.W.  1998.  Modeling the radiative characteristics of airborne mineral aerosols at infrared wavelengths.  Journal of Geophysical Research 103: 8813-8826.

Tegen, I., Lacis, A.A. and Fung, I.  1996.  The influence on climate forcing of mineral aerosols from disturbed soils.  Nature 380: 419-422.

Vogelmann, A.M., Flatau, P.J., Szczodrak, M., Markowicz, K.M. and Minnett, P.J.  2003.  Observations of large aerosol infrared forcing at the surface.  Geophysical Research Letters 30: 10.1029/2002GL016829.

Wielicki, B.A., Wong, T., Allan, R.P., Slingo, A., Kiehl, J.T., Soden, B.J., Gordon, C.T., Miller, A.J., Yang, S.-K., Randall, D.A., Robertson, F., Susskind, J. and Jacobowitz, H.  2002.  Evidence for large decadal variability in the tropical mean radiative energy budget.  Science 295: 841-844.

Wild, M.  1999.  Discrepancies between model-calculated and observed shortwave atmospheric absorption in areas with high aerosol loadings.  Journal of Geophysical Research 104: 27,361-27,371.

Wild, M. and Ohmura, A.  1999.  The role of clouds and the cloud-free atmosphere in the problem of underestimated absorption of solar radiation in GCM atmospheres.  Physics and Chemistry of the Earth 24B: 261-268.

Zender, C.S.  1999.  Global climatology of abundance and solar absorption of oxygen collision complexes.  Journal of Geophysical Research 104: 24,471-24,484.