Lane et al. (2000), for example, evaluated the sensitivities of cloud-radiation parameterizations utilized in the GCMs of that era to changes in vertical model resolution, varying the latter from 16 to 60 layers in increments of four and comparing the results to observed values. This effort revealed that (1) cloud fraction varied by approximately 10% over the range of resolutions tested, which corresponded to about 20% of the observed cloud cover fraction. Similarly, (2) outgoing longwave radiation varied by 10 to 20 Wm-2 as model vertical resolution was varied, amounting to approximately 5 to 10% of observed values, while (3) incoming solar radiation experienced similar significant variations across the range of resolutions tested. What is more, (4) the model results did not converge, even at a resolution of 60 layers.

In an analysis of the multiple roles played by cloud microphysical processes in determining tropical climate, Grabowski (2000) found much the same thing, noting that (1) there were serious problems related to the degree to which computer models failed to correctly incorporate cloud microphysics. And these observations led him to conclude that (2) "it is unlikely that traditional convection parameterizations can be used to address this fundamental question in an effective way." He also was convinced that (3) "classical convection parameterizations do not include realistic elements of cloud physics and (4) they represent interactions among cloud physics, radiative processes, and surface processes within a very limited scope." Consequently, he but stated the obvious when he concluded that (5) "model results must be treated as qualitative rather than quantitative."

Reaching similar conclusions were Gordon et al. (2000), who determined that (1) many GCMs of the late 1990s tended to under-predict the presence of subtropical marine stratocumulus clouds, and that (2) they failed to simulate the seasonal cycle of the clouds. These deficiencies were extremely important, because the particular clouds they studied exerted a major cooling influence on the surface temperatures of the sea below them. In the situation Gorden and his colleagues investigated, for example, the removal of the low clouds, as occurred in the normal application of their model, led to sea surface temperature increases on the order of 5.5°C.

Further condemnation of turn-of-the-century model treatments of clouds came from Harries (2000), who wrote that our knowledge of high cirrus clouds was very poor and that (1) "we could easily have uncertainties of many tens of Wm-2 in our description of the radiative effects of such clouds, and how these properties may change under climate forcing." This problem was especially noteworthy in light of the fact that the radiative effect of a doubling of the air's CO2 content is only on the order of low single-digit Wm-2. And, therefore, it was truly an understatement to say, as Harries did, that (2) "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."

Moving into the 21st century, Lindzen et al. (2001) analyzed cloud cover and sea surface temperature (SST) data over a large portion of the Pacific Ocean, finding a strong inverse relationship between upper-level cloud area and mean SST, such that the area of cirrus cloud coverage normalized by a measure of the area of cumulus coverage decreased by about 22% for each degree C increase in cloudy region SST. Essentially, as the three researchers thus described it, "the cloudy-moist region appears to act as an infrared adaptive iris that opens up and closes down the regions free of upper-level clouds, which more effectively permit infrared cooling, in such a manner as to resist changes in tropical surface temperature."

The sensitivity of this negative feedback was calculated by Lindzen et al. to be substantial. In fact, they estimated it would "more than cancel all the positive feedbacks in the more sensitive then-current climate models," which were being used at that time to predict the consequences of projected increases in the atmosphere's CO2 concentration. And, as one might have expected, evidence for this potential impediment to global warming was nowhere to be seen back then, just as it is nowhere to be seen now, even in today's most advanced GCMs.

Clearly, this challenge to climatic political correctness could not go uncontested; and Hartmann and Michelsen (2002) quickly claimed that the correlation noted by Lindzen et al. resulted from variations in subtropical clouds that are not physically connected to deep convection near the equator, and that it was thus "unreasonable to interpret these changes as evidence that deep tropical convective anvils contract in response to SST increases." Fu et al. (2002) also chipped away at the adaptive infrared iris concept, arguing that "the contribution of tropical high clouds to the feedback process would be small since the radiative forcing over the tropical high cloud region is near zero and not strongly positive," while additionally claiming to show that water vapor and low cloud effects were overestimated by Lindzen et al. by at least 60% and 33% respectively." And as a result, they obtained a feedback factor in the range of -0.15 to -0.51, compared to Lindzen et al.'s much larger negative feedback factor of -0.45 to -1.03.

In a simultaneously published reply to this critique, Chou et al. (2002) stated that Fu et al.'s approach to specifying longwave emission and cloud albedos "appears to be inappropriate for studying the iris effect," and that since "thin cirrus are widespread in the tropics and ... low boundary clouds are optically thick, the cloud albedo calculated by [Fu et al.] is too large for cirrus clouds and too small for boundary layer clouds," so that "the near-zero contrast in cloud albedos derived by [Fu et al.] has the effect of underestimating the iris effect." In the end, however, Chou et al. agreed that Lindzen et al. "may indeed have overestimated the iris effect somewhat, though hardly by as much as that suggested by [Fu et al.]."

Although there has thus been some convergence in the two extreme views of the subject, the debate over the reality and/or magnitude of the adaptive infrared iris effect continued apace; and when some of the meteorological community's best minds continued to clash over the nature and magnitude of the phenomenon, it was amazing that climate alarmists continued to clamor for actions to reduce anthropogenic CO2 emissions at almost all costs, as if the issue were settled when it clearly was not.

This situation is illustrative of the importance of the advice given two years earlier by Grassel (2000), who in a review of the then-current status of the climate modeling enterprise noted that changes in many climate-related phenomena, including cloud optical and precipitation properties caused by changes in the spectrum of cloud condensation nuclei, were insufficiently well known to provide useful insights into future conditions. His advice in the light of this knowledge gap was that "we must continuously evaluate and improve the GCMs we use," although he was forced to acknowledge that contemporary climate model results were already being "used by many decision-makers, including governments."

This state of affairs has continued to the present day and is very disturbing, as national and international policy is being made on the basis of vastly imperfect mathematical representations of a whole host of physical, chemical and biological phenomena, many of which involve clouds. Although some may think that what we currently know about the subject is sufficient for predictive purposes, a host of questions posed by Grassl -- for which we still lack definitive answers -- demonstrates that this assumption is erroneous.

As but a single example, Charlson et al. (1987) described a negative feedback process that links biologically-produced dimethyl sulfide (DMS) in the oceans with climate. The basic tenant of this hypothesis derives from the fact that the global radiation balance is significantly influenced by the albedo of marine stratus clouds, and that the albedo of these clouds is a function of cloud droplet concentration, which is dependent upon the availability of condensation nuclei that have their origin in the flux of DMS from the world's oceans to the atmosphere.

Acknowledging that the roles played by DMS oxidation products within the context described above are indeed "diverse and complex" and in many instances "not well understood," Ayers and Gillett (2000) summarized empirical evidence supporting Charlson et al.'s hypothesis that was derived from data collected at Cape Grim, Tasmania, and from reports of other pertinent studies in the peer-reviewed scientific literature. And in light of their findings, they reported that (1) the "major links in the feedback chain proposed by Charlson et al. (1987) have a sound physical basis," and that there is thus (2) "compelling observational evidence to suggest that DMS and its atmospheric products participate significantly in processes of climate regulation and reactive atmospheric chemistry in the remote marine boundary layer of the Southern Hemisphere."

The empirical evidence analyzed by Ayers and Gillett highlights an important suite of negative feedback processes that act in opposition to model-predicted CO2-induced global warming over the world's oceans; and these processes are not fully incorporated into even the very best of the current crop of climate models, nor are analogous phenomena that occur over land included in them, such as those discussed by Idso (1990).

Further to this point, O'Dowd et al. (2004) measured size-resolved physical and chemical properties of aerosols found in northeast Atlantic marine air arriving at the Mace Head Atmospheric Research station on the west coast of Ireland during phytoplanktonic blooms at various times of the year. And in doing so, they found that in the winter, when biological activity was at its lowest, the organic fraction of the sub-micrometer aerosol mass was about 5%. During the spring through autumn, however, when biological activity was high, they found that "the organic fraction dominates and contributes 63% to the sub-micrometer aerosol mass (about 45% is water-insoluble and about 18% water-soluble)." And based on these findings, they performed model simulations that indicated that (1) the marine-derived organic matter "can enhance the cloud droplet concentration by 15% to more than 100% and is therefore an important component of the aerosol-cloud-climate feedback system involving marine biota."

As for the significance of their findings, O'Dowd et al. stated that their data "completely change the picture of what influences marine cloud condensation nuclei given that water-soluble organic carbon, water-insoluble organic carbon and surface-active properties, all of which influence the cloud condensation nuclei activation potential, are typically not parameterized in current climate models," or as they stated in another place in their paper, "an important source of organic matter from the ocean is omitted from current climate-modeling predictions and should be taken into account."

Another perspective on the cloud-climate conundrum was provided by Randall et al. (2003), who stated at the outset of their review of the subject that "the representation of cloud processes in global atmospheric models has been recognized for decades as the source of much of the uncertainty surrounding predictions of climate variability." They reported, however, that "despite the best efforts of [the climate modeling] community ... the problem remains largely unsolved." In addition, they said that "at the current rate of progress, cloud parameterization deficiencies will continue to plague us for many more decades into the future."

So what's the problem? "Clouds are complicated," Randall et al. declared, as they began to describe what they called the "appalling complexity" of the cloud parameterization situation. For starters, they stated that (1) "our understanding of the interactions of the hot towers [of cumulus convection] with the global circulation is still in a fairly primitive state." And not knowing all that much about what goes up, it's not surprising that we didn't know all that much about what comes down, as they reported that (2) "downdrafts are either not parameterized or only crudely parameterized in large-scale models."

With respect to stratiform clouds, the situation was no better, as their parameterizations were described by Randall et al. as "very rough caricatures of reality." As for interactions between convective and stratiform clouds, forget about it! ... which is pretty much what many scientists themselves did during the 1970s and 80s, when Randall et al. were reporting that "cumulus parameterizations were extensively tested against observations without even accounting for the effects of the attendant stratiform clouds." Even at the time of their study, in fact, they had to report that the concept of detrainment was "somewhat murky," and that (1) the conditions that trigger detrainment were "imperfectly understood." Hence, it should again come as no surprise that "at this time," as they put it, (2) "no existing GCM includes a satisfactory parameterization of the effects of mesoscale cloud circulations."

On top of these problems, Randall et al. additionally indicated that (3) "the large-scale effects of microphysics, turbulence, and radiation should be parameterized as closely coupled processes acting in concert." But they had to report that (4) only a few GCMs had even attempted to do so. And why? Because, as they continued, "the cloud parameterization problem is overwhelmingly complicated," and "cloud parameterization developers," as they called them, were still "struggling to identify the most important processes on the basis of woefully incomplete observations." To drive this point home, they also said "there is little question why the cloud parameterization problem is taking a long time to solve: it is very, very hard." In fact, the four scientists concluded that (5) "a sober assessment suggests that with current approaches the cloud parameterization problem will not be 'solved' in any of our lifetimes." That's right - in any of our lifetimes!

With such a bleak assessment of where the climate-modeling community stood at that time with respect to just the single issue of cloud parameterization, it might be well to pause and ask ourselves how anyone could possibly feel confident about what even the best climate models of today are predicting about CO2-induced global warming, where proper cloud responses are critical to reaching a correct conclusion. The answer is so obvious it need not even be stated.

But wait! There appeared to be a glimmer of light at the end of the climate-modeling tunnel. It was a long way off ... and it looked to be incredibly expensive ... but it was there. And it beckoned, ever so enticingly.

This shining hope of the climate-modeling community of today, as forseen by Randall et al., resides in something called "cloud system-resolving models" or CSRMs, which can be compared with single-column models or SCMs that can be "surgically extracted from their host GCMs." These advanced models, as they described them, "have resolutions fine enough to represent individual cloud elements, and space-time domains large enough to encompass many clouds over many cloud lifetimes." Of course, these improvements would mean that "the computational cost of running a CSRM is hundreds or thousands of times greater than that of running an SCM." But in a few more decades, according to Randall et al., it should become possible to use such global CSRMs "to perform century-scale climate simulations, relevant to such problems as anthropogenic climate change."

Though normally less than a lifetime, a few more decades is a little long to have to wait to address an issue that climate alarmists have long been prodding the world to confront. Hence, Randall et al. suggested that an approach that could be used very soon (to possibly determine whether or not there even is a problem) would be to "run a CSRM as a 'super-parameterization' inside a GCM," which configuration they called a "super-GCM."

So it all comes down to this: either we know enough about how the world's climate system works, so that we don't need the postulated super-GCMs, or we don't know enough about how it works and we do need them. Unfortunately, the cloud parameterization problem by itself is so complex that no one can validly claim, at this point in time, that humanity's continued utilization of fossil-fuel energy will result in massive counter-productive climatic changes. There is absolutely no justification for that conclusion in the output of reliable super-GCMs, simply because there are no such models. And that the basis for this conclusion is robust, and cannot be said to rely upon the less-than-enthusiastic remarks of a handful of exasperated climate model belivers, we report the results of two additional studies of the subject that were published subsequent to the analysis of Randall et al.

In the first of these studies, which was conducted by seventeen other climate modelers, Siebesma et al. (2004) reported that "simulations with nine large-scale models [were] carried out for June/July/August 1998 and the quality of the results [was] assessed along a cross-section in the subtropical and tropical North Pacific ranging from (235°E, 35°N) to (187.5°E, 1°S)," in order to "document the performance quality of state-of-the-art GCMs in modeling the first-order characteristics of subtropical and tropical cloud systems." And the main conclusions of this study, according to Siebesma et al., were that (1,2) "almost all models strongly under-predicted both cloud cover and cloud amount in the stratocumulus regions," while (3,4) "the situation is opposite in the trade-wind region and the tropics," where (5,6) "cloud cover and cloud amount are over-predicted by most models." In fact, they reported that (7) "these deficiencies result in an over-prediction of the downwelling surface short-wave radiation of typically 60 W m-2 in the stratocumulus regimes," and (8,9) "a similar under-prediction of 60 W m-2 in the trade-wind regions and in the intertropical convergence zone (ITCZ)," which discrepancies are to be compared with a radiative forcing of only a couple of W m-2 for a 300-ppm increase in the atmosphere's CO2 concentration. In addition, they stated that (10, 11) "similar biases for the short-wave radiation were found at the top of the atmosphere, while discrepancies in the outgoing long-wave radiation are most pronounced in the ITCZ."

The seventeen scientists, who hailed from nine different countries, also stated that (12) "the representation of clouds in general-circulation models remains one of the most important as yet unresolved issues in atmospheric modeling," which was partially due, they said, "to the overwhelming variety of clouds observed in the atmosphere, but even more so due to the large number of physical processes governing cloud formation and evolution as well as the great complexity of their interactions." Hence, they concluded that through repeated critical evaluations of the type they conducted, "the scientific community will be forced to develop further physically sound parameterizations that ultimately result in models that are capable of simulating our climate system with increasing realism," which suggests that it would not be wise to put much credence in what admittedly inadequate current state-of-the-art GCMs suggest about the future, nor to actually mandate drastic reductions in fossil-fuel energy use on the basis of what the predictions of these models currently suggest.

Moving forward in time a bit more, Zhang et al. (2005) compared basic cloud climatologies derived from ten atmospheric GCMs with satellite measurements obtained from the International Satellite Cloud Climatology Project (ISCCP) and the Clouds and Earth's Radiant Energy System (CERES) program, where ISCCP data were available from 1983 to 2001, while data from the CERES program were available for the winter months of 2001 and 2002 and for the summer months of 2000 and 2001, the purpose of their analysis being two-fold: (1) to assess the current status of climate models in simulating clouds so that future progress can be measured more objectively, and (2) to reveal serious deficiencies in the models so as to improve them.

The work of the 20 additional climate modelers involved in this exercise revealed a huge list of major model imperfections. First of all, Zhang et al. reported there was (1) a four-fold difference in high clouds among the models, and that (2) the majority of the models only simulated 30-40% of the observed middle clouds, with (3) some models simulating less than a quarter of observed middle clouds. For low clouds, they additionally reported that (4) half the models underestimated them, such that (5) the grand mean of low clouds from all models was only 70-80% of what was observed. Furthermore, when stratified in optical thickness ranges, (6) the majority of the models simulated optically thick clouds more than twice as frequently as was found to be the case in the satellite observations, while (7,8) the grand mean of all models simulated about 80% of optically intermediate clouds and 60% of optically thin clouds. And (9) in the case of individual cloud types, the group of researchers reported that "differences of seasonal amplitudes among the models and satellite measurements can reach several hundred percent."

Two years later, in their introduction to a group of eight research papers published in the International Journal of Remote Sensing, Muller and Fischer (2007) reported some of the highlights of the 1997-2000 EU-CLOUDMAP project. Originally designed to improve the measurement and characterization of cirrus and contrail cloud properties, the two researchers described how the project was ultimately broadened "to include properties of clouds at all altitudes, as Cess et al. (1993) had shown that depending on how cloud processes are parameterized can lead to an order of magnitude difference in predictions of surface temperature due to changes in CO2 radiative forcing," which they noted was "by far the largest uncertainty in making accurate forecasts of global warming."

Conducted as a collaborative effort of five university and government research groups in the UK, Germany, Switzerland and the Netherlands, Muller and Fischer noted that "the primary technological motivation of the project was to develop new techniques for deriving cloud-top properties (cloud-top height, amount, microphysics and winds) from a new series of meteorological sensors," and to apply these properties "to the generation of new cloud climatology products," while they wrote that "a secondary goal was to develop an automated technique, based on fuzzy logic, to detect contrails in non-thermal imagery where contrails can only be detected through their unique spatial characteristics." So how did the project fare in terms of contributing to its ultimate goals?

Muller and Fischer generously concluded that "the principal scientific goals to improve the measurement and characterization of cirrus and contrail cloud properties as a first priority as well as clouds in general were attained." However, they went on to say that (1) "in the future, more extensive investigations on clouds are necessary with respect to global observations to reach the essential knowledge on clouds required for significant improvements in ... climate modeling."

In a contemporary study that sought out some of that "essential knowledge," Zhou et al. (2007) compared the cloud and precipitation properties observed from the Clouds and the Earth's Radiant Energy System (CERES) and Tropical Rainfall Measuring Mission (TRMM) instruments against simulations obtained from the three-dimensional Goddard Cumulus Ensemble (GCE) model during the South China Sea Monsoon Experiment (SCSMEX) field campaign of 18 May-18 June 1998. And this work revealed, as they described it, that (1) "the GCE rainfall spectrum includes a greater proportion of heavy rains than PR (Precipitation Radar) or TMI (TRMM Microwave Imager) observations," that (2) "the GCE model produces excessive condensed water loading in the column, especially the amount of graupel as indicated by both TMI and PR observations," that (3) "the model also cannot simulate the bright band and the sharp decrease of radar reflectivity above the freezing level in stratiform rain as seen from PR," that (4) "the model has much higher domain-averaged OLR (outgoing longwave radiation) due to smaller total cloud fraction," that (5) "the model has a more skewed distribution of OLR and effective cloud top than CERES observations, indicating that the model's cloud field is insufficient in area extent," that (6) "the GCE is ... not very efficient in stratiform rain conditions because of the large amounts of slowly falling snow and graupel that are simulated," and finally, in summation, that (7) "large differences between models and observations exist in the rain spectrum and the vertical hydrometeor profiles that contribute to the associated cloud field." And in light of these several significant findings, it was made quite clear that cloud resolving models still had a long way to go before they would be ready for "prime time" in mankind's complex quest to properly assess the roles of various types of clouds and forms of precipitation in the future evolution of earth's climate in response to variations in numerous anthropogenic and background forcings.

Jumping ahead five years -- and while noting that climate modelers had long struggled to adequately represent the sensitivity of convective cloud systems to tropospheric humidity in their mathematical representations of earth's climate system -- Del Genio (2012) reviewed the rate of progress in this important area. And in doing so, he found that a number of important problems related to this particular field of study had yet to be adequately resolved. He noted, for example, that (1) many parameterizations of convective cloud variability "are not sufficiently sensitive to variations in tropospheric humidity," which "lack of sensitivity," as he described it, "can be traced in part to [2] underestimated entrainment of environmental air into rising convective clouds and [3] insufficient evaporation of rain into the environment." And as a result of these deficiencies, he further noted that (4,5) "the parameterizations produce deep convection too easily while stabilizing the environment too quickly to allow the effects of convective mesoscale organization to occur."

To be fair, Del Genio did note that "recent versions of some models have increased their sensitivity to tropospheric humidity and improved some aspects of their variability," but he said that (6) "a parameterization of mesoscale organization is still absent from most models," while stating that (7) "adequately portraying convection in all its realizations remains a difficult problem."

On another note, Del Genio wrote that "to date, metrics for model evaluation have focused almost exclusively on time mean two-dimensional spatial distributions of easily observed parameters," and he indicated that "it has become clear that such metrics have no predictive value for climate feedbacks and climate sensitivity (e.g., Collins et al., 2011)," while adding that those metrics "are also probably not helpful for assessing most other important features of future climate projections, because temporal variability gives greater insight into the physical processes at work." And so it was that Del Genio concluded by opining that (8) "given the insensitivity of these models to tropospheric humidity and [9,10] their failure to simulate the Madden-Julian Oscillation and diurnal cycle, ... it seems unlikely that it will ever be possible to establish a general set of metrics that can be used to anoint one subset of models as our most reliable indicators of all aspects of climate change."

Also publishing a pertinent paper in the same year were Li et al. (2012), who introduced the report of their study by noting that representing clouds and cloud climate feedback in global climate models (GCMs) remained "a pressing challenge" that needed to be overcome in order to "reduce and quantify uncertainties associated with climate change projections." And two of the primary parameters that needed to be accurately modeled, in this regard, were cloud ice water content (CIWC) and cloud ice water path (CIWP).

Consequently, Li et al. went on to perform "an observationally based evaluation of the cloud ice water content and path of present-day GCMs, notably 20th century CMIP5 simulations," after which they compared the results to a pair of climatic reanalyses. This they did using "three different CloudSat + CALIPSO ice water products and two methods to remove the contribution from the convective core ice mass and/or precipitating cloud hydrometeors with variable sizes and falling speeds so that a robust observational estimate can be obtained for model evaluations."

Unfortunately, the eleven U.S. scientists found that (1,2) "for annual mean CIWP, there are factors of 2-10 in the differences between observations and models for a majority of the GCMs and for a number of regions," additionally noting that (3) "systematic biases in CIWC vertical structure occur below the mid-troposphere where the models overestimate CIWC." And in light of these and other shortcomings they identified, they ultimately concluded that (4) "neither the CMIP5 ensemble mean nor any individual model performs particularly well," adding that (5) "there are still a number of models that exhibit very large biases" and noting that they do so "despite the availability of relevant observations." What is more, even in situations where they felt the models might be providing roughly the correct radiative energy budget," they found that (6) "many are accomplishing it by means of unrealistic cloud characteristics of cloud ice mass at a minimum, which in turn likely indicates [7,8] unrealistic cloud particle sizes and cloud cover."

In a concomitant study, Cesana and Chepfer (2012) made a point of noting that "clouds are the primary modulators of the Earth's radiation budget" and that they therefore constitute "the main source of uncertainty in model estimates of climate sensitivity," citing Randall et al. (2007). And as a result of this fact, they further stated that (1) the modeling of cloud properties represents "a major limitation to the reliability of climate change projections," additionally citing Dufresne and Bony (2008) in this regard.

Faced with this problem, Cesana and Chepfer indicated that in order "to improve the reliability of climate change projections, it is therefore imperative to improve the representation of cloud processes in models." So how much improving did the models need?

In broaching this important question, the two French researchers compared the then-most-recent cloud representations of five of the climate models involved in the Coupled Model Intercomparison Project Phase 5 (CMIP5) effort -- which had just recently been described by Taylor et al. (2012) -- with real-world satellite-derived observations obtained from the GCM-Oriented CALIPSO Cloud Product (GOCCP), which had earlier been described by Chepfer et al. (2010). And what did they learn from this exercise?

In the words of Cesana and Chepfer, they learned that (1) "low- and mid-level altitude clouds are underestimated by all the models (except in the Arctic)," that (2) "high altitude cloud cover is overestimated by some models," that (3) "some models shift the altitude of the clouds along the ITCZ by 2 km (higher or lower) compared to observations," that (4) "the models hardly reproduce the cloud free subsidence branch of the Hadley cells," that (5) "the high-level cloud cover is often too large," that (6) "in the tropics, the low-level cloud cover (29% in CALIPSO-GOCCP) is underestimated by all models in subsidence regions (16% to 25%)" and, last of all, that (7) "the pronounced seasonal cycle observed in low-level Arctic clouds is hardly simulated by some models."

Working contemporaneously, Cesana et al. (2012) reported that "low-level clouds frequently occur in the Arctic and exert a large influence on Arctic surface radiative fluxes and Arctic climate feedbacks," noting that during winter, in particular, surface net longwave radiation (FLW,NET) has a bi-modal distribution, with extremes that have been termed "radiatively clear" and "radiatively opaque." And in further discussing these clouds, they said that Arctic ice clouds "tend to have small optical depths and a weak influence on FLW,NET," which explains the "radiatively clear" condition, while adding that Arctic liquid-containing clouds "generally have large optical depths and a dominant influence on FLW,NET (Shupe and Intrieri, 2004)," which helps explain the "radiatively opaque" condition, as discussed by Doyle et al. (2011).

Getting back to their own study, Cesana et al. employed real-world Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observation (CALIPSO) data to document cloud phases over the Arctic basin (60-82°N) during the five-year period 2006-2011, after which they used the results they obtained "to evaluate the influence of Arctic cloud phase on Arctic cloud radiative flux biases in climate models." This work revealed, as they reported, that their evaluation of climate models participating in the most recent Coupled Model Inter-comparison Project (Taylor et al., 2012) indicated that (1) "most climate models are not accurately representing the bimodality of FLW,NET in non-summer seasons." In fact, they found that even when advanced microphysical schemes that predict cloud phase had been used -- such as those employed in the fifth version of the Community Atmosphere Model (CAM5, Neale et al., 2010) -- (2) "insufficient liquid water was predicted." And so it was that Cesana et al. concluded from what they had learned that (3) "the simple prescribed relationships between cloud phase and temperature that have historically been used in climate models are incapable of reproducing the Arctic cloud phase observations described here," which finding must inevitably lead to (4,5) similarly inaccurate values of "Arctic surface radiative fluxes and Arctic climate feedbacks," when employed in current state-of-the-art climate models.

In yet another contemporaneous study, Nam et al. (2012) wrote that the response of low-level clouds had long been identified as "a key source of uncertainty for model cloud feedbacks under climate change," citing the work of Bony and Dufresne (2005), Webb et al. (2006), Wyant et al. (2006) and Medeiros et al. (2008). And they further stated that "the ability of climate models to simulate low-clouds and their radiative properties" plays a huge role in assessing "our confidence in climate projections."

In studying this important unresolved dilemma, Nam et al. analyzed "outputs from multiple climate models participating in the Fifth phase of the Coupled Model Intercomparison Project (CMIP5) using the Cloud Feedback Model Intercomparison Project Observations Simulator Package (COSP), and compared them with different satellite data sets," including "CALIPSO lidar observations, PARASOL mono-directional reflectances, and CERES radiative fluxes at the top of the atmosphere." And what did they thereby learn?

In the words of the four French researchers, "the current generation of climate models still experiences difficulties in predicting the low-cloud cover and its radiative effects." In particular, they reported that the models: (1) "under-estimate low-cloud cover in the tropics," (2) "over-estimate optical thickness of low-clouds, particularly in shallow cumulus regimes," (3) "poorly represent the dependence of the low-cloud vertical structure on large-scale environmental conditions," and (4) "predict stratocumulus-type of clouds in regimes where shallow cumulus cloud-types should prevail." However, they said that "the impact of these biases on the Earth's radiation budget ... is reduced by compensating errors," including (5-7) "the tendency of models to under-estimate the low-cloud cover and to over-estimate the occurrence of mid- and high-clouds above low-clouds."

In a relevant study published a year later, Lauer and Hamilton (2013) reported that numerous previous studies from the Coupled Model Intercomparison Project phase 3 (CMIP3) showed (1) quite large biases in the simulated cloud climatology affecting all GCMs (Global Climate Models), as well as (2) "a remarkable degree of variation among the models that represented the state of the art circa 2005." So what was the case in 2013?

The two researchers provided an update by describing the progress that had been made in the intervening years by comparing mean cloud properties, their interannual variability and the climatological seasonal cycle -- as derived from CMIP5 models -- with results from comparable CMIP3 experiments, as well as with actual satellite observations. And these analyses revealed, in Lauer and Hamilton's words, that (1) "the simulated cloud climate feedbacks activated in global warming projections differ enormously among state-of-the-art models," informing us that (2) "this large degree of disagreement has been a constant feature documented for successive generations of GCMs from the time of the first Intergovernmental Panel on Climate Change assessment through the CMIP3 generation models used in the fourth IPCC assessment." And they added that (3) "even the model-simulated cloud climatologies for present-day [2013] conditions are known to depart significantly from observations and, once again, [4] the variation among models is quite remarkable," citing the studies of Weare (2004), Zhang et al. (2005), Waliser et al. (2007, 2009), Lauer et al. (2010) and Chen et al. (2011).

As for some other specifics, the two researchers determined that (5) "long-term mean vertically integrated cloud fields have quite significant deficiencies in all the CMIP5 model simulations," that (6) "both the CMIP5 and CMIP3 models display a clear bias in simulating too high LWP [liquid water path] in mid-latitudes," that (7) "this bias is not reduced in the CMIP5 models," that (8,9) there have been "little to no changes in the skill of reproducing the observed LWP and CA [cloud amount]," that (10) "inter-model differences are still large in the CMIP5 simulations," and that (11) "there is very little to no improvement apparent in the tropical and subtropical regions in CMIP5."

In closing, therefore, Lauer and Hamilton indicated there was "only very modest improvement in the simulated cloud climatology in CMIP5 compared with CMIP3," and they sadly stated that even this slightest of improvements was "mainly a result of careful model tuning rather than an accurate fundamental representation of cloud processes in the models."

Writing concurrently in the Journal of Geophysical Research (Atmospheres), Wang and Su (2013) noted that "coupled general circulation models (GCMs) are the major tool to predict future climate change, yet cloud-climate feedback constitutes the largest source of uncertainty in these modeled future climate projections." Consequently, they correctly stated that "confidence in the future climate change projections by the coupled GCMs to a large extent depends on how well these models simulate the observed present-day distribution of clouds and their associated radiative fluxes." And in describing how they made this determination, they wrote that in their particular study, "the annual mean climatology of top of the atmosphere (TOA) shortwave and longwave cloud radiative effects in 12 Atmospheric Model Intercomparison Project (AMIP)-type simulations participating in the Coupled Model Intercomparison Project Phase 5 (CMIP5) [were] evaluated and investigated using satellite-based observations, with a focus on the tropics."

At the conclusion of this undertaking, the two researchers reported that (1) the CMIP5 AMIPs "produce considerably less cloud amount [than what is observed], particularly in the middle and lower troposphere," that there are "good model simulations in tropical means," but they are (2) "a result of compensating errors over different dynamical regimes," that "over the Maritime Continent, [3] most of the models simulate moderately less high-cloud fraction, leading to [4-6] weaker shortwave cooling and longwave warming and a larger net cooling," that "over subtropical strong subsidence regimes, most of the CMIP5 models [7] strongly underestimate stratocumulus cloud amount and [8] show considerably weaker local shortwave cloud radiative forcings," that "over the transitional trade cumulus regimes, a notable feature is that while at varying amplitudes, [9,10] most of the CMIP5 models consistently simulate a deeper and drier boundary layer, [11] more moist free troposphere, and [12] more high clouds and consequently [13-14] overestimate shortwave cooling and longwave warming effects there," such that, in the final analysis, (15) "representing clouds and their TOA radiative effects remains a challenge in the CMIP5 models."

Also publishing in the same timeframe were Evan et al. (2013), who wrote as background for their study that "stratocumulus (Sc) cloud cover is a persistent feature of the subtropical North and South Atlantic," further noting that "it is well known that Sc cloud cover increases with decreasing temperatures of the underlying sea surface and that an increase in cloud cover will cool the surface temperatures via increasing the local albedo, otherwise known as the Sc feedback." And, therefore, they used real-world observations to "quantify the magnitude and spatial structure of the Sc feedback in the tropical-extratropical Atlantic Ocean," as well as to "investigate the role of the Sc feedback in shaping the evolution of coupled modes of variability there," especially when utilizing CMIP3 models.

This work revealed, as the four researchers reported, that (1,2) "most models have negative biases in the mean state of Sc cloud cover and do not reproduce the observed spatial structure of Atlantic Sc clouds." In addition, they found that "while the majority of models exhibit some agreement with observations in the meridional structure of the Sc feedback, [3] the vast majority of models underestimate the dependence of Sc cloud cover on the underlying sea surface temperature." So once again, there is another situation where important aspects of both cloud type and cloud cover are simply not portrayed to an acceptable degree of real-world faithfulness in the vast majority of CMIP3 models.

In another same-year publication, Suzuki et al. (2013) wrote that "climate models contain various uncertain parameters in the formulations of parameterizations for physical processes," but they additionally noted that "these parameters represent 'tunable knobs' that are typically adjusted to let the models reproduce realistic values of key-observed climate variables." And, therefore, they felt it important to examine "the validity of a tunable cloud parameter," namely, "the threshold particle radius triggering the warm rain formation in a climate model."

The model they chose for this particular purpose was the Geophysical Fluid Dynamics Laboratory (GFDL) Coupled Climate Model version 3 (CM3), because it was known that alternate values of that model's tunable cloud parameter that fall within its real-world range of uncertainty "have been shown to produce severely different historical temperature trends due to differing magnitudes of aerosol indirect forcing."

The results of the three researchers' analysis indicated that (1) "the simulated temperature trend best matches [the] observed trend when the model adopts the threshold radius that worst reproduces satellite-observed microphysical statistics and vice versa." And in light of this finding, the three researchers wrote that (2) "this inconsistency between the 'bottom-up' process-based constraint and the 'top-down' temperature trend constraint implies the presence of compensating errors in the model." And they thus concluded that "if this behavior is not a peculiarity of the GFDL CM3, the contradiction may be occurring in other climate models as well," which is not what one would want to see happen.

About this same time, Karlsson and Svensson (2013) introduced their most recent study of the subject by writing that "clouds significantly influence the Arctic surface energy budget and a realistic representation of this impact is a key for proper simulation of the present-day and future climate," while further indicating that "considerable across-model spread in cloud variables remains in the fifth phase of the Coupled Model Intercomparison Project ensemble and partly explains the substantial across-model spread in the surface radiative effect of the clouds," which further impacts sea-ice extent and albedo. And, therefore, they focused their attention primarily "on how model differences in the parameterization of sea-ice albedo in the fifth phase of the Coupled Model Intercomparison Project (CMIP5) influence the cloud radiative effect on the surface energy budget and the annual cycle of sea-ice concentration."

This work revealed, in the words of the two researchers, that (1,2) "the across-model spread in Arctic cloud cover and cloud condensates is substantial, and no improvement is seen from previous model intercomparisons," citing their own earlier study (Karlsson and Svenson, 2011) in this regard and noting that "this diversity of simulated Arctic clouds in the CMIP5 ensemble contributes to a spread in the models' cloud influence on the surface energy budget." And so it was that in the concluding sentence of their most recent paper, the two Stockholm (Sweden) University scientists reiterated the fact that (3) present-day sea-ice albedo is so badly constrained in GCMs that it "impacts the fidelity of future scenario assessments of the Arctic region and should therefore be a concern for the modeling community."

Still stuck in the same -- but scientifically productive - year, Huang (2013) addressed a pair of issues that concerned the longwave climate feedbacks in transient climate change assessments. The first of these issues was the fact that (1) "the radiative forcing of greenhouse gases, as measured by their impact on the outgoing longwave radiation (OLR), may vary across different climate models even when the concentrations of these gases are identically prescribed," which forcing variation, as Huang continued, (2) "contributes to the discrepancy in these models' projections of surface warming." The second issue was that "the stratosphere is an important factor that affects the OLR in transient climate change," in that stratospheric water vapor and temperature changes may both act as positive feedbacks during global warming and, therefore, "cannot be fully accounted as a 'stratospheric adjustment' of radiative forcing."

And as the Canadian researcher went on to demonstrate in the body of his paper, (3) "neglecting these two issues may cause a bias in the longwave cloud feedback diagnosed as a residual term in the decomposition of OLR variations." And he further noted, in this regard, that his results "and the recent results of others [e.g., the estimate of Zelinka et al. (2012) based on cloud property histograms] indicate that (4) there is, in fact, no consensus in terms of the sign of the longwave cloud feedback among the GCMs."

Finally moving on another year, Rosenfeld et al. (2014) wrote that "aerosols counteract part of the warming effects of greenhouse gases, mostly by increasing the amount of sunlight reflected back to space." However, they also noted that (1) "the ways in which aerosols affect climate through their interaction with clouds are complex and incompletely captured by climate models." And as a result, the four researchers further acknowledged that (2) "the radiative forcing (that is, the perturbation to Earth's energy budget) caused by human activities is highly uncertain, making it difficult to predict the extent of global warming," while also citing, in this regard, the studies of Anderson et al. (2003) and Stocker et al. (2013).

Delving still deeper into the subject, Rosenfeld et al. further reported that (3) "recent advances have revealed a much more complicated picture of aerosol-cloud interactions than considered previously," while also stating that (4) "further progress is hampered by limited observational capabilities and coarse-resolution models." In addition, they acknowledged that (5) "little is known about the unperturbed aerosol level that existed in the preindustrial era," noting that "this reference level is very important for estimating the radiative forcing from aerosols," citing Carslaw et al. (2013).

Also holding up progress was the fact, as Rosenfeld et al. put it, that (6,7) our "understanding of the formation of ice and its interactions with liquid droplets is even more limited, mainly due to poor ability to measure the ice-nucleating activity of aerosols and the subsequent ice-forming processes in clouds." And in this regard they said that "improved observational tests are essential for validating the results of simulations and ensuring that modeling developments are on the right track." But they also indicated that (8) what they called a "major challenge" in this area was the fact that "the most important aerosol nucleation region is at the bottom of a cloud, which is obscured by the rest of the cloud if measured from above."

Consequently, it was no surprise that Rosenfeld et al. concluded that (9) "fully resolved, global, multi-year simulations are not likely to become feasible for many decades." Yes, that's right -- many decades. And this analysis of the situation can only make one wonder if the world's climate alarmists are not putting the cart way before the horse, when it comes to pushing for such drastic actions as they continually promote to prevent what may eventually be found to actually be no problem at all.

In another contemporary and intriguing paper, Lin et al. (2014) wrote that stratocumulus clouds in the tropics and subtropics have come to be known as "climate refrigerators," in light of the likelihood that "a 5% increase of their coverage would be sufficient to offset the global warming induced by doubling CO2," due to the clouds' reflecting of an enhanced amount of incoming sunlight back to space, as suggested by the studies of Randall et al. (1984), Slingo (1990), Bretherton et al. (2004) and Wood (2012). And in light of these several earlier findings, Lin et al. went on to examine stratocumulus clouds and associated cloud feedback in the southeast Pacific (SEP), as simulated by eight global climate models that participated in phase 5 of the Coupled Model Intercomparison Project (CMIP5), as well as the Cloud Feedback Model Intercomparison Project (CFMIP), based on "long-term observations of clouds, radiative fluxes, cloud radiative forcing (CRF), sea surface temperature (SST), and [the] large-scale atmospheric environment."

This work revealed, in the words of the three researchers, that "state-of-the-art global climate models still have significant difficulty in simulating the SEP stratocumulus clouds and associated cloud feedback." More specifically, and compared with observations, they reported that (1) "the models tend to simulate significantly less cloud cover," (2) "higher cloud tops," (3) "a variety of unrealistic cloud albedos," (4) "overly weak shortwave cloud radiative forcing," (5) "biases in large-scale temperature structure," including (6) "lack of temperature inversion," (7) "insufficient lower troposphere stability [LTS]," and (8) "insufficient reduction of LTS with local SST warming," along with (9) "improper model physics" as it pertains to (10) "insufficient increase of low cloud cover associated with larger LTS." And these findings represent but one group of a host of global climate model failures to adequately "predict the past," which fact raises serious questions about mankind's ability to ever correctly predict earth's future climate.

Also contributing to the evaluation of modern-day climate models was Pithan et al. (2014), who introduced their study of the subject by noting that "temperature inversions are a common feature of the Arctic wintertime boundary layer," and by going on to say that "they have important impacts on both radiative and turbulent heat fluxes and partly determine local climate-change feedbacks," which led them to further state that "understanding the spread in inversion strength modelled by current global climate models is thus an important step in better understanding Arctic climate and its present and future changes." And in a quest to help obtain that "better understanding," Pithan et al. went on to show "how the formation of Arctic air masses leads to the emergence of a cloudy and a clear state of the Arctic winter boundary layer," after which they also described the different climatic implications of each of these states.

In the Arctic's cloudy state, the three researchers found "little to no surface radiative cooling occurs and inversions are elevated and relatively weak," whereas in the Arctic's clear state they found that "surface radiative cooling leads to strong surface-based temperature inversions." And when comparing specific aspects of model output to real-world observations, they determined that (1) the "freezing of super-cooled water at too warm temperatures that occurs in many CMIP5 models leads to a lack of high-emissivity mixed-phase clouds and thus of a cloudy state in these models," and that (2) "models lacking a cloudy state display excessive surface radiative cooling in Arctic winter, which tends to produce strong low-level stability and temperature inversions."

In addition, Pithan et al. reported that (3) "few models that allow for cloud liquid water at very low temperatures reproduce both the clear and cloudy state of the boundary layer," that (4) "a second group of models lacks the cloudy state and exhibits strong stability and strong long-wave cooling, that (5) "other models also lack the cloudy state, but generate weak stability despite strong long-wave cooling," that (6) the CMIP5 inter-model spread of typical monthly-mean low-level stability over sea ice in winter is about 10 K, which is similar to that in CMIP3 models," that (7) "15 out of 21 CMIP5 models overestimate low-level stability over sea ice compared to reanalysis data," that (8) "this wide-spread model bias is linked to shortcomings in the representation of mixed-phase cloud microphysics," and that (9) "differences in cloud properties, energy fluxes and inversion strengths between land and sea ice domains remain to be investigated."

In light of these several findings, Pithan et al. stated, in the concluding sentence of their paper, that "in order to better represent the Arctic winter boundary layer and surface energy budget in climate models, an important step would be to improve the mixed-phase cloud microphysics and to obtain an adequate representation of the cloudy state." And so we have yet another glaring example of the fact that today's CMIP5 models are simply not up to the task of adequately portraying Earth's current climate, which must surely be done before we can rely on them to provide valid portrayals of Earth's future climatic state.

Another recent stab at dealing with current climate model complexities was taken by Park et al. (2014), who wrote that "clouds cool the Earth-atmosphere system by reflecting incoming shortwave (SW) radiation and warm it by absorbing outgoing longwave (LW) radiation from the surface," while further noting that "satellite observations reveal that the net radiative effect of clouds on the Earth-atmosphere system is a cooling of 20-24 Wm-2 in the global average," which they note is "about six times larger than the radiative forcing associated with doubled CO2," citing Ramanathan et al. (1989) and Loeb et al. (2009). And, hence, it can be appreciated that properly modeling cloud processes is an important aspect of ongoing efforts to predict the future course of Earth's climate.

As for their contribution to this important task, the three US researchers devised several unique adjustments to previous versions of the Community Atmosphere Models (CAMs) versions 3 and 4, which are now found in the new-and-improved CAM5 set of models. And "compared with the previous versions," as they wrote, "the cloud parameterizations in CAM5 are more consistent and physically based, due to inclusion of more realistic and complex parameterizations and much attention given to the interactions among them within a more consistent framework."

However, they went on to acknowledge that even with these improvements, "several systematic biases were also identified in the simulated cloud fields in CAM5," which they grouped into three different categories: (1) deficient regional tuning, (2) inconsistency between various physics parameterizations, and (3) incomplete modeled physics. And in the case of the latter category, they listed the following problems: (4) "underestimation of LW CRF [cloud radiative forcing] due to the horizontal heterogeneity assumption of water vapor within each grid layer in the radiation scheme," (5) "overly strong SW CRF and LW CRF in the tropics due to the use of a single-type cloud within the radiation scheme," and (6) "under-frequent shallow convective activity over summer continents due to the neglect of forced convection." And in light of these several remaining problems, they ultimately concluded that "while substantially improved from its predecessors [CAM3/CAM4], many aspects of CAM5 can and should be improved in the future," which they described as something "upon which we are continuously working with collaborators." But until such improvements are made, model treatments of clouds should be treated with a healthy dose of skepticism.

Filling out this year of research, Lacagnina and Selten (2014) wrote that "despite the importance of clouds, their representation in general circulation models (GCMs) continues to account for much of the uncertainties in climate projections," citing Cess et al. (1996), Stocker et al. (2001) and Solomon et al. (2007), while noting that "the spread associated with inter-model differences is roughly three times larger than that associated with other main feedbacks," citing Dufresne and Bony (2008). And six years after the last of these assessments, they have found little that encourages them.

Working with the EC-Earth GCM version 2.3 that they coupled to an ocean GCM based on version 2 of the Nucleus for European Modeling of the Ocean (NEMO) model, the Dutch duo from the Royal Netherlands Meteorological Institute compared the married models' outputs to a wealth of real-world observational data obtained from numerous satellite and land-based sensors that they described in detail and to which they provided profuse references.

These efforts revealed, as they described them, that (1) "EC-Earth exhibits the largest cloud biases in the tropics," where it (2) "underestimates the total cloud cover," but that it (3) "overestimates the optically thick clouds," with the net effect that (4) "clouds exert an overly strong cooling effect in the model," that (5) "the magnitude of the cooling due to the shortwave cloud radiative effect is underestimated for the stratiform low-clouds," because (6) "the model simulates too few of them," that (7) the "shortwave cloud radiative effect is overestimated for trade wind cumulus clouds," because (8) "in the model these are too thick," that (9) "the clouds in the deep convection regions also tend to overestimate the shortwave cloud radiative effect," because (10) "these clouds are generally too thick" and that (11) "there are too few mid and high thin clouds."

As for the ultimate take-home message of these several findings, Lacagnina and Selten (2014) concluded that "the model weaknesses discussed above indicate that more effort is needed to improve the physical parameterizations employed." And that may well be a huge understatement of the current status of climate modeling; for after several generations of "improvements," costing multiple hundreds of millions of dollars, these theoretical constructs prove yet again that they are still not ready for dependable usage!

At long last, arriving in the year in which this analysis and writing was conducted, Rapp (2015) introduced her study of the subject by writing that (1) "the sensitivity of regimes dominated by low clouds has been identified as the largest contributor to uncertainties in tropical cloud feedback estimates in climate models." And she thus went on to describe how Atmospheric Model Intercomparison Project [AMIP] simulations of low cloud responses to sea surface temperature (SST) compare with satellite observations in the southeastern Pacific subsidence region, beginning with the (2) "too few, too bright" problem identified by Nam et al. (2012).

As for what she further learned, Rapp reports that the AMIP models have considerable difficulty in simulating (1) the annual cycle in the cloud radiative effect (CRE), (2) cloud fraction, and (3) liquid water path (LWP), likely due in part to (4,5) "underestimation of the strength of lower tropospheric stability and the depth of the boundary layer," but also noting that (6) stratocumulus clouds "are not sensitive enough." In addition, Rapp notes that (7,8) "not only do the seasonal amplitudes disagree but also some models have an annual cycle that is nearly 180 degrees out of phase with the observations."

"Ultimately," therefore, Rapp wrote that the results she reported "show that climate models still have difficulty reproducing the observed cloud and radiative sensitivities in a low cloud regime, even when forced with climatological SSTs."

References
Anderson, T.L., Charlson, R.J., Schwartz, S.E., Knutti, R., Boucher, O., Rodhe, H. and Heintzenberg, J. 2013. Climate forcing by aerosols - a hazy picture. Science 300: 1103-1104.

Ayers, G.P. and Gillett, R.W. 2000. DMS and its oxidation products in the remote marine atmosphere: implications for climate and atmospheric chemistry. Journal of Sea Research 43: 275-286.

Bony, S. and Dufresne, J. 2005. Marine boundary layer clouds at the heart of tropical cloud feedback uncertainties in climate models. Geophysical Research Letters 32: 10.1029/2005GL023851.

Bretherton, C.S., Uttal, T., Fairall, C.W., Yuter, S., Weller, R., Baumgardner, D., Comstock, K., Wood, R. and Raga, G. 2004. The EPIC 2001 stratocumulus study. Bulletin of the American Meteorological Society 85: 967-977.

Carslaw, K.S., Lee, L.A., Reddington, C.L., Pringle, K.J., Rap, A., Forster, P.M., Mann, G.W., Spracklen, D.V., Woodhouse, M.T., Regayre, L.A. and Pierce, J.R. 2013. Nature 503: 67-71.

Cesana, G. and Chepfer, H. 2012. How well do climate models simulate cloud vertical structure? A comparison between CALIPSO-GOCCP satellite observations and CMIP5 models. Geophysical Research Letters 39: 10.1029/2012GL053153.

Cesana, G., Kay, J.E., Chepfer, H., English, J.M. and de Boer, G. 2012. Ubiquitous low-level liquid-containing Arctic clouds: New observations and climate model constraints from CALIPSO-GOCCP. Geophysical Research Letters 39: 10.1029/2012GL053385.

Cess, R.D., Nemesure, S., Dutton, E.G., Deluisi, J.J., Potter, G.L. and Morcrette, J. 1993. The impact of clouds on the shortwave radiation budget of the surface atmosphere system - Interfacing measurements and models. Journal of Climate 6: 308-316.

Cess, R.D., Zhang, M.H., Ingram, W.J., Potter, G.L., Alekseev, V., Barker, H.W., Cohen?Solal, E., Colman, R.A., Dazlich, D.A., Del Genio, A.D., Dix, M.R., Dymnikov, V., Esch, M., Fowler, L.D., Fraser, J.R., Galin, V., Gates, W.L., Hack, J.J., Kiehl, J.T., Le Treut, H., Lo, K.K.-W., McAvaney, B.J., Meleshko, V.P., Morcrette, J.-J., Randall, D.A., Roeckner, E., Royer, J.-F., Schlesinger, M.E., Sporyshev, P.V., Timbal, B., Volodin, E.M., Taylor, K.E., Wang, W. and Wetherald, R.T. 1996. Cloud feedback in atmospheric general circulation models: an update. Journal of Geophysical Research 101: 12,791-12,794.

Charlson, R.J., Lovelock, J.E., Andrea, M.O. and Warren, S.G. 1987. Oceanic phytoplankton, atmospheric sulfur, cloud albedo and climate. Nature 326: 655-661.

Chen, W.-T., Woods, C.P., Li, J.-L.F., Waliser, D.E., Chern, J.-D., Tao, W.-K., Jiang, J.H. and Tompkins, A.M. 2011. Partitioning CloudSat ice water content for comparison with upper tropospheric ice in global atmospheric models. Journal of Geophysical Research 116: 10.1029/2010JD015179.

Chepfer, H., Bony, S., Winker, D., Cesana, G., Dufresne, J.L., Minnis, P., Stubenrauch, C.J. and Zeng, S. 2010. The GCM Oriented CALIPSO Cloud Product (CALIPSO-GOCCP). Journal of Geophysical Research 115: 10.1029/2009JD012251.

Chou, M.-D., Lindzen, R.S. and Hou, A.Y. 2002. Reply to: "Tropical cirrus and water vapor: an effective Earth infrared iris feedback?" Atmospheric Chemistry and Physics 2: 99-101.

Collins, M., Booth, B.B.B., Bhaskaran, B., Harris, G.R., Murphy, J.M., Sexton, D.M.H. and Webb, M.J. 2011. Climate model errors, feedbacks and forcings: a comparison of perturbed physics and multi-model ensembles. Climate Dynamics 36: 1737-1766.

Del Genio, A.D. 2012. Representing the sensitivity of convective cloud systems to tropospheric humidity in general circulation models. Surveys in Geophysics 33: 637-656.

Doyle, J.G., Lesins, G., Thakray, C.P., Perro, C., Nott, G.J., Duck, T.J., Damoah, R. and Drummond, J.R. 2011. Water vapor intrusions into the High Arctic during winter. Geophysical Research Letters 38: 10.1029/2011GL047493.

Dufresne, J.-L. and Bony, S. 2008. An assessment of the primary sources of spread of global warming estimates from coupled atmosphere-ocean models. Journal of Climate 21: 5135-5144.

Evan, A.T., Allen, R.J., Bennartz, R. and Vimont, D.J. 2013. The modification of sea surface temperature anomaly linear damping time scales by stratocumulus clouds. Journal of Climate 26: 3619-3630.

Fu, Q., Baker, M. and Hartmann, D.L. 2002. Tropical cirrus and water vapor: an effective Earth infrared iris feedback? Atmospheric Chemistry and Physics 2: 31-37.

Gordon, C.T., Rosati, A. and Gudgel, R. 2000. Tropical sensitivity of a coupled model to specified ISCCP low clouds. Journal of Climate 13: 2239-2260.

Grabowski, W.W. 2000. Cloud microphysics and the tropical climate: Cloud-resolving model perspective. Journal of Climate 13: 2306-2322.

Grassl, H. 2000. Status and improvements of coupled general circulation models. Science 288: 1991-1997.

Groisman, P.Ya., Bradley, R.S. and Sun, B. 2000. The relationship of cloud cover to near-surface temperature and humidity: Comparison of GCM simulations with empirical data. Journal of Climate 13: 1858-1878.

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

Hartmann, D.L. and Michelsen, M.L. 2002. No evidence for IRIS. Bulletin of the American Meteorological Society 83: 249-254.

Huang, Y. 2013. On the longwave climate feedbacks. Journal of Climate 26: 7603-7610.

Idso, S.B. 1990. A role for soil microbes in moderating the carbon dioxide greenhouse effect? Soil Science 149: 179-180.

Karlsson, J. and Svensson, G. 2011. The simulation of Arctic clouds and their influence on the winter surface temperature in present-day climate in the CMIP3 multi-model dataset. Climate Dynamics 36: 623-635.

Karlsson, J. and Svensson, G. 2013. Consequences of poor representation of Arctic sea-ice albedo and cloud-radiation interactions in the CMIP5 model ensemble. Geophysical Research Letters 40: 4374-4379.

Lane, D.E., Somerville, R.C.J. and Iacobellis, S.F. 2000. Sensitivity of cloud and radiation parameterizations to changes in vertical resolution. Journal of Climate 13: 915-922.

Lauer, A. and Hamilton, K. 2013. Simulating clouds with global climate models: A comparison of CMIP5 results with CMIP3 and satellite data. Journal of Climate 26: 3823-3845.

Lauer, A., Hamilton, K., Wang, Y., Phillips, V.T.J. and Bennartz, R. 2010. The impact of global warming on marine boundary layer clouds over the eastern Pacific - A regional model study. Journal of Climate 23: 5844-5863.

Li, J.-L.F., Waliser, D.E., Chen, W.-T., Guan, B., Kubar, T., Stephens, G., Ma, H.-Y., Deng, M., Donner, L., Seman, C. and Horowitz, L. 2012. An observationally based evaluation of cloud ice water in CMIP3 and CMIP5 GCMs and contemporary reanalyses using contemporary satellite data. Journal of Geophysical Research 117: 10.1029/2012JD017640.

Lin, J.-L., Qian, T. and Shinoda, T. 2014. Stratocumulus clouds in Southeaster Pacific simulated by eight CMIP5-CFMIP global climate models. Journal of Climate 27: 3000-3022.

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.

Loeb, N., Wielicki, D., Doelling, D., Smith, G., Keyes, D., Kato, S., Manalo-Smith, N. and Wong, T. 2009. Toward optimal closure of the earth's top-of-atmosphere radiation budget. Journal of Climate 22: 748-766.

Medeiros, B., Stevens, B., Held, I., Zhao, M., Williamson, D., Olson, J. and Bretherton, C. 2008. Aquaplanets, climate sensitivity, and low clouds. Journal of Climate 21: 4974-4991.

Muller, J.-P. and Fischer, J. 2007. The EU-CLOUDMAP project: Cirrus and contrail cloud-top maps from satellites for weather forecasting climate change analysis. International Journal of Remote Sensing 28: 1915-1919.

Nam, C., Bony, S., Dufresne, J.-L. and Chepfer, H. 2012. The 'too few, too bright' tropical low-cloud problem in CMIP5 models. Geophysical Research Letters 39: 10.1029/2012GL053421.

Neale, R.B., et al. 2010. Description of the NCAR Community Atmosphere Model (CAM 5.0). Technical Note 486+STR. National Center for Atmospheric Research, Boulder, Colorado, USA.

O'Dowd, C.D., Facchini, M.C., Cavalli, F., Ceburnis, D., Mircea, M., Decesari, S., Fuzzi, S., Yoon, Y.J. and Putaud, J.-P. 2004. Biogenically driven organic contribution to marine aerosol. Nature 431: 676-680.

Park, S., Bretherton, C.S. and Rasch, P.J. 2014. Integrating cloud processes in the Community Atmosphere Model, Version 5. Journal of Climate 27: 6821-6856.

Pithan, F., Medeiros, B. and Mauritsen, T. 2014. Mixed-phase clouds cause climate model biases in Arctic wintertime temperature inversions. Climate Dynamics 43: 289-303.

Ramanathan, V., Cess, R., Harrison, E.,Minnis, P., Barkstrom, B., Ahmad, E. and Hartmann, D. 1989. Cloud-radiative forcing and climate: Results from the Earth Radiation Budget Experiment. Science 243: 57-63.

Randall, D.A., Coakley, J.A., Fairall, C.W., Knopfli, R.A. and Lenschow, D.H. 1984. Outlook for research on marine subtropical stratocumulus clouds. Bulletin of the American Meteorological Society 65: 1290-1301.

Randall, D., Khairoutdinov, M. Arakawa, A. and Grabowski, W. 2003. Breaking the cloud parameterization deadlock. Bulletin of the American Meteorological Society 84: 1547-1564.

Randall, D.A., Wood, R.A., Bony, S., Coleman, R., Fichefet, T., Fyfe, J., Kattsov, V., Pitman, A., Shukla, J., Srinivasan, J., Stouffer, R.J., Sumi, A. and Taylor, K.E. 2007. Chapter 8: Climate Models and Their Evaluation. In: Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, United Kingdom.

Rapp, A.D. 2015. Cloud responses in the AMIP simulations of CMIP5 models in the southeastern Pacific marine subsidence region. International Journal of Climatology 35: 2908-2921.

Rosenfeld, D., Sherwood, S., Wood, R. and Donner, L. 2014. Climate effects of aerosol-cloud interactions. Science 343: 379-380.

Shupe, M.D. and Intrieri, J.M. 2004. Cloud radiative forcing of the Arctic surface: The influence of cloud properties, surface albedo, and solar zenith angle. Journal of Climate 17: 616-628.

Siebesma, A.P., Jakob, C., Lenderink, G., Neggers, R.A.J., Teixeira, J., van Meijgaard, E., Calvo, J., Chlond, A., Grenier, H., Jones, C., Kohler, M., Kitagawa, H., Marquet, P., Lock, A.P., Muller, F., Olmeda, D. and Severijns, C. 2004. Cloud representation in general-circulation models over the northern Pacific Ocean: A EUROCS intercomparison study. Quarterly Journal of the Royal Meteorological Society 130: 3245-3267.

Slingo, A. 1990. Sensitivity of the earth's radiation budget to changes in low clouds. Nature 343: 49-51.

Solomon, S., Qin, D., Manning, M., Chen, Z., Marquis, M., Averyt, K.B., Tignor, M. and H.L. Miller, H.L. (Eds.) 2007. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, New York, New York, USA.

Stocker, T.F. et al. 2001. Physical climate processes and feedbacks. In: Houghton, J.T. et al. (Eds), Climate Change 2001: The Scientific Basis, Contributions of Working Group I to the Third Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, United Kingdom.

Stocker, T.F., Qin, D., Plattner, G.K., Tignor, M.M.B., Allen, S.K., Boschung, J., Nauels, A., Xia, Y., Bex, V. and Midgley, P.M. (Eds.). Summary for Policymakers. In: Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, UK.

Suzuki, K., Golaz, J.-C. and Stephens, G.L. 2013. Evaluating cloud tuning in a climate model with satellite observations. Geophysical Research Letters 40: 4464-4468.

Taylor, K.E., Stouffer, R.J. and Meehl, G.A. 2012. An overview of CMIP5 and the experimental design. Bulletin of the American Meteorological Society 93: 485-498.

Waliser, D.E., Li, J.-L.F., Woods, C.P., Austin, R.T., Bacmeister, J., Chern, J., Del Genio, A., Jiang, J.H., Kuang, Z., Meng, H., Minnis, P., Platnick, S., Rossow, W.B., Stephens, G.L., Sun-Mack, S., Tao, W.-K., Tompkins, A.M., Vane, D.G., Walker, C. and Wu, D. 2009. Cloud ice: A climate model challenge with signs and expectations of progress. Journal of Geophysical Research 114: 10.1029/2008JD010015.

Waliser, D.E., Seo, K.-W., Schubert, S. and Njoku, E. 2007. Global water cycle agreement in the climate models assessed in the IPCC AR4. Geophysical Research Letters 34: 10.1029/2007GL030675.

Wang, H. and Su, W. 2013. Evaluating and understanding top of the atmosphere cloud radiative effects in Intergovernmental Panel on Climate Change (IPCC) Fifth Assessment Report (AR5) Coupled Model Intercomparison Project Phase 5 (CMIP5) models using satellite observations. Journal of Geophysical Research: Atmospheres 118: 683-699.

Weare, B.C. 2004. A comparison of AMIP II model cloud layer properties with ISCCP D2 estimates. Climate Dynamics 22: 281-292.

Webb, M.J., Senior, C.A., Sexton, D.M.H., Ingram, W.J., Williams, K.D., Ringer, M.A., McAvaney, B.J., Colman, R., Soden, B.J., Gudgel, R., Knutson, T., Emori, S., Ogura, T., Tsushima, Y., Andronova, N., Li, B., Musat, I., Bony, S. and Taylor, K.E. 2006. On the contribution of local feedback mechanisms to the range of climate sensitivity in two GCM ensembles. Climate Dynamics 27: 17-38.

Wood, R. 2012. Stratocumulus clouds. Monthly Weather Review 140: 2373-2423.

Wyant, M., Khairoutdinov, M. and Bretherton, C. 2006. Climate sensitivity and cloud response of a GCM with a superparameterization. Geophysical Research Letters 33: 10.1029/2005GL025464.

Zelinka, M.D., Klein, S.A. and Hartmann, D.L. 2012. Computing and partitioning cloud feedbacks using cloud property histograms. Part I.: Cloud radiative kernels. Journal of Climate 25: 3715-3735.

Zhang, M.H., Lin, W.Y., Klein, S.A., Bacmeister, J.T., Bony, S., Cederwall, R.T., Del Genio, A.D., Hack, J.J., Loeb, N.G., Lohmann, U., Minnis, P., Musat, I., Pincus, R., Stier, P., Suarez, M.J., Webb, M.J., Wu, J.B., Xie, S.C., Yao, M.-S. and Yang, J.H. 2005. Comparing clouds and their seasonal variations in 10 atmospheric general circulation models with satellite measurements. Journal of Geophysical Research 110: D15S02, doi:10.1029/2004JD005021.

Zhou, Y.P., Tao, W.-K., Hou, A.Y., Olson, W.S., Shie, C.-L., Lau, K.-M., Chou, M.-D., Lin, X. and Grecu, M. 2007. Use of high-resolution satellite observations to evaluate cloud and precipitation statistics from cloud-resolving model simulations. Part I: South China Sea monsoon experiment. Journal of the Atmospheric Sciences 64: 4309-4329.