Scientists: ‘Falsified’ Climate Models ‘Do Not Employ Known Physics Fully’…‘Don’t Agree With Reality’

By Kenneth Richard

Many new scientific papers affirm climate model results conflict with one another, diverge from observations, and aren’t fully rooted in established physics.

Image Source: Essex and Tsonis, 2018


Climate models are predicated on the assumption that greenhouse gases exert fundamental control on the Earth’s climate system.  That’s why for decades it’s been predicted that disaster will befall the planet as a consequence of rising CO2 emissions.

Image Source: Associated Press (1989)
And yet contrary to how they are popularly portrayed, climate models do not fully employ the laws of physics in their representations (Essex and Tsonis, 2018).   This is likely why climate model outputs are (a) often widely different from one another and (b) frequently diverge from real-world observations.

We Lack Understanding of Climate Mechanisms

In contrast with governmental (the United Nations’) manufactured framework of certainty, scientists are increasingly suggesting we have yet to adequately understand fundamental processes and mechanisms in the Earth’s climate system.
“[W]e can build and run complex models of the Earth system, but we do not have adequate enough understanding of the processes and mechanisms to be able to quantitatively evaluate the predictions and projections they produce, or to understand why different models give different answers.” (Collins et al. 2018)
“[C]limate changes in polar areas remain difficult to predict, which indicates that the underlying mechanisms of polar amplification remain uncertain and debatable.”  (Ding et al., 2018)

Unfalsifiable Models

Furthermore, because climate models aren’t subjected to the “hard” science standard of falsification, they are necessarily presented to the non-skeptical public as unfalsifiable.  In other words, when climate models don’t agree with real-world observations, they aren’t assumed to be wrong, or worthy of disposal.  Instead, they are suggested to merely need a little re-tuning.

The refusal to discard climate models that conflict with observations is apparently rooted in politics.  Kundzewicz et al. (2018) point out that the “hard” science standard that says results should be quantitatively validated with a measured degree of certainty before formulating policy initiatives is deemed “unrealistic and counterproductive” today.  That’s why climate modeling thrives in the modern “soft” political world – a realm where the rigors of observation and falsification — the scientific method — need not apply.
“[I]n the past, science was assumed to provide ‘hard’ results in quantitative form, in contrast to ‘soft’ determinants of politics, that were interest-driven and value-laden. Yet, the traditional assumption of the certainty of scientific information is now recognized as unrealistic and counterproductive.” (Kundzewicz et al., 2018)

Climate Models Don’t Agree With Reality

Problematically, even when they are re-tuned, climate models still yield widely divergent outputs both from one another and compared to observational evidence.

Many new scientific papers have been published in recent months that document the failure of climate models to simulate the Earth’s climate.  A sampling of 10 peer-reviewed papers from 2018 are highlighted below.

In several cases, scientists have reported that none of the modern-day climate model results are consistent with real-world observations.  In some cases the models yield opposite results (i.e., warming instead of cooling, rising instead of falling, etc.).

It is increasingly being recognized that climate models “not only don’t agree with each other when it comes to dynamics, they also don’t agree with reality(Essex and Tsonis, 2018).

1. Essex and Tsonis, 2018

• “Climate models do not and cannot employ known physics fullyThus, they are falsified, a priori. Incomplete physics and the finite representation of computers can induce false instabilities.”
The standard model of physics, for example, is subject to falsification. If it fails to make correct predictions in controlled experiments, it is false. Projections are not good enough there. Even in astrophysics, models explain phenomena that are normally subject to falsification through broad questions asked about multiple occurrences of similar physical circumstances, even in highly data-starved contexts. What makes climate models fundamentally different is that they are presented as being unfalsifiable. Even when they deviate from actual observations, they are not superseded by a better competing model. Deviations simply invite some retuning. Moreover instead of replacement by better models retuning leads to all models becoming more alike.”
• “[A]re there propositions that contemporary models make, crucial to their own objectives, that are falsifiable? Is there any physical test possible that would force us to conclude that they are unable to achieve their own objectives, thus requiring a rethinking of basic assumptions? This paper addresses this question. But it is a question that cannot be comprehended in the face of many widely-held misconceptions about the direct meteorologically based projection modeling of climateForemost among these misconceptions is that climate models are full implementations of known, mature physics. This false conception can lead to the conclusion that falsification is irrelevant because models are simply an execution of previously known correct physics.”
• “The empirical nature of large climate models can be clearly seen in their diverse outputs. If they followed the laws of physics in their entirety, they would all produce the same results under the same conditionsBut they do not. In a recent study, the Climate Model Inter-comparison Project phase 3 (CMIP3) models [2] were considered and a detailed comparison at the dynamics level, using an approach involving climate networks [Steinhaeuser and Tsonis, 2013] was performed. It was found that the models not only don’t agree with each other when it comes to dynamics, they also don’t agree with reality.”

2. Collins et al., 2018

• “Here there is a dynamical gap in our understanding. While we have conceptual models of how weather systems form and can predict their evolution over days to weeks, we do not have theories that can adequately explain the reasons for an extreme cold or warm, or wet or dry, winter at continental scales. More importantly, we do not have the ability to credibly predict such states.”
• “Likewise, we can build and run complex models of the Earth system, but we do not have adequate enough understanding of the processes and mechanisms to be able to quantitatively evaluate the predictions and projections they produce, or to understand why different models give different answers.
• “The global warming ‘hiatus’ provides an example of a climate event potentially related to inter-basin teleconnections. While decadal climate variations are expected, the magnitude of the recent event was unforeseen. A decadal period of intensified trade winds in the Pacific and cooler sea surface temperatures (SSTs) has been identified as a leading candidate mechanism for the global slowdown in warming.”

3. Kundzewicz et al., 2018

• “Climate models need to be improved before they can be effectively used for adaptation planning and design. Substantial reduction of the uncertainty range would require improvement of our understanding of processes implemented in models and using finer resolution of GCMs and RCMs. However, important uncertainties are unlikely to be eliminated or substantially reduced in near future (cf. Buytaert et al., 2010). Uncertainty in estimation of climate sensitivity (change of global mean temperature, corresponding to doubling atmospheric CO2 concentration) has not decreased considerably over last decades. Higher resolution of climate input for impact models requires downscaling (statistical or dynamic) of GCM outputs, adding further uncertainty.”
• “[C]limate models do not currently simulate the water cycle at sufficiently fine resolution for attribution of catchment-scale hydrological impacts to anthropogenic climate change. It is expected that climate models and impact models will become better integrated in the future.”
• “Calibration and validation of a hydrological model should be done before applying it for climate change impact assessment, to reduce the uncertainty of results. Yet, typically, global hydrological models are not calibrated and validated. … Model-based projections of climate change impact on water resources can largely differ. If this is the case, water managers cannot have confidence in an individual scenario or projection for the future. Then, no robust, quantitative, information can be delivered and adaptation procedures need to be developed which use identified projection ranges and uncertainty estimates. Moreover, there are important, nonclimatic, factors affecting future water resources.”
• “As noted by Funtowicz and Ravetz (1990), in the past, science was assumed to provide “hard” results in quantitative form, in contrast to “soft” determinants of politics, that were interest-driven and value-laden. Yet, the traditional assumption of the certainty of scientific information is now recognized as unrealistic and counterproductive. Policy-makers have to make “hard” decisions, choosing between conflicting options (with commitments and stakes being the primary focus), using “soft” scientific information that is bound with considerable uncertainty. Uncertainty has been policitized in that policy-makers have their own agendas that can include the manipulation of uncertainty. Parties in a policy debate may invoke uncertainty in their arguments selectively, for their own advantage.”

4. Lacour et al., 2018

• “The representation of clouds over Greenland is a central concern for the models because clouds impact ice-sheet surface melt. We find that over Greenland, most of the models have insufficient cloud cover during summer. In addition, all models create too few non-opaque liquid containing clouds optically thin enough to let direct solar radiation reach the surface (-1% to -3.5% at the ground level). Some models create too few opaque clouds. In most climate models, the cloud properties biases identified over all Greenland also apply at Summit proving the value of the ground observatory in model evaluation.”
• “At Summit, climate models underestimate cloud radiative effect (CRE) at the surface, especially in summer. The primary driver of the summer CRE biases compared to observations is the underestimation of the cloud cover in summer (-46% to -21%), which leads to an underestimated longwave radiative warming effect (CRELW = -35.7 W m-2 to -13.6 W m-2 compared to the ground observations) and an underestimated shortwave cooling effect (CRESW = +1.5 W m-2 to +10.5 W m-2 compared to the ground observations). Overall, the simulated [modeled] clouds do not radiatively warm the surface as much as observed.”
• “Of particular importance, clouds can trigger surface melt over a large portion of the Greenland Ice Sheet (Bennartz et al. 2013; Solomon et al. 2017). Greenland surface melting increases non-linearly with increasing temperatures due to positive feedbacks between cloud microphysics, surface melting and surface albedo (Fettweis et al. 2013) and modulates the ice sheet mass balance (Van Tricht et al. 2016; Hofer et al. 2017).”
• “Every model included in this study underestimates the net cloud radiative surface warming in summer. … [O]nly few general circulation models are able to represent the surface of the Greenland ice sheet (Cullather et al. 2014). … Since the overall cloud radiative warming is underestimated in the models, we may expect an underestimate of Greenland surface melting. However, misrepresentation of clouds is not the only contributor to biases in the modeled surface melting.”

None of the climate models match the observations

5. Kravtsov et al., 2018

• “[D]eviations of the model-simulated climate change from observations, such as a recent “pause” in global warming, have received considerable attention. Such decadal mismatches between model-simulated and observed climate trends are common throughout the twentieth century, and their causes are still poorly understood.”
• “While climate models exhibit various levels of decadal climate variability and some regional similarities to observations, none of the model simulations considered match the observed signal in terms of its magnitude, spatial patterns and their sequential time development. These results highlight a substantial degree of uncertainty in our interpretation of the observed climate change using current generation of climate models.”

6. Kam et al., 2018

• “In summary, there is marginal evidence for an emerging detectable anthropogenic contribution toward earlier WSCT [winter-spring center time] in parts of North America. The regions with strongest relative indication of an anthropogenic contribution in our analysis include: the north-central U.S. (Region 3); the mountainous western U.S./southwestern Canada (Region 1); and extreme northeastern U.S. and Canadian Maritimes (Region 6).”
• “However, in none of the regions examined do a majority of the nine CMIP5 models examined robustly support a detectable attribution of an earlier (decreasing) WSCT trend to anthropogenic forcing. At some level, the difficulty in detecting a climate change signal comes down to low signal to noise ratio (Ziegler et al. 2005). Apparently, for the variable at hand, the climate change influence is not very large compared to interannual/interdecadal variability noise.”

7. Agarwal and Wettlaufer, 2018

• “The fluctuation statistics of the observed sea-ice extent during the satellite era are compared with model output from CMIP5 models using a multifractal time series method. The two robust features of the observations are that on annual to biannual time scales the ice extent exhibits white noise structure, and there is a decadal scale trend associated with the decay of the ice cover.”
• “It is shown that (i) there is a large inter-model variability in the time scales extracted from the models, (ii) none of the models exhibits the decadal time scales found in the satellite observations, (iii) five of the 21 models [24%] examined exhibit the observed white noise structure, and (iv) the multi-model ensemble mean exhibits neither the observed white noise structure nor the observed decadal trend.”

8. Luo et al., 2018

• “Over the recent three decades sea surface temperate (SST) in the eastern equatorial Pacific has decreased, which helps reduce the rate of global warming. However, most CMIP5 model simulations with historical radiative forcing do not reproduce this Pacific La Niña-like cooling. Based on the assumption of ‘perfect’ models, previous studies have suggested that errors in simulated internal climate variations and/or external radiative forcing may cause the discrepancy between the multi-model simulations and the observation.”
• “Based on the total 126 realizations of the 38 CMIP5 model Historical simulations, the results show that none of the 126 model historical realizations reproduce the intensity of the observed eastern Pacific cooling (Fig. 1d) and only one simulation produces a weak cooling (−0.007 °C per decade).”

9. Hanna et al., 2018

• “Recent changes in summer Greenland blocking captured by none of the CMIP5 models
• “Given well-established connections between atmospheric pressure over the Greenland region and air temperature and precipitation extremes downstream … this brings into question the accuracy of simulated North Atlantic jet stream changes and resulting climatological anomalies … as well as of future projections of GrIS mass balance produced using global and regional climate models.”

10. Scanlon et al., 2018

• “The models underestimate the large decadal (2002–2014) trends in water storage relative to GRACE satellites, both decreasing trends related to human intervention and climate and increasing trends related primarily to climate variations. The poor agreement between models and GRACE underscores the challenges remaining for global models to capture human or climate impacts on global water storage trends.”
• “Increasing TWSA [total water storage anomalies] trends are found primarily in nonirrigated basins, mostly in humid regions, and may be related to climate variations. Models also underestimate median GRACE increasing trends (1.6–2.1 km3/y) by up to a factor of ∼8 in GHWRMs [global hydrological and water resource models] (0.3–0.6 km3/y).”
• “Underestimation of GRACE-derived TWSA increasing trends is much greater for LSMs [global land surface models], with four of the five LSMs [global land surface models] yielding opposite trends (i.e., median negative rather than positive trends).”
• “Increasing GRACE trends are also found in surrounding basins, with most models yielding negative trends. Models greatly underestimate the increasing trends in Africa, particularly in southern Africa.”
• “TWSA trends from GRACE in northeast Asia are generally increasing, but many models show decreasing trends, particularly in the Yenisei.”
• “[T]he magnitude of the estimated climate contribution to GMSL [global mean sea level] is twice that of the human contribution and opposite in sign.”

This article appeared on the NoTricksZone website at