Refuting IPCC's claims on climate change, by showing how science basis has been used in an inappropriate way

Antonio Sesé, M.Sc. in Physics antonio_sese_@ yahoo.com, v.2, March 2014.

The author encourages the free distribution of this document. Except for third party materials, (i.e. figures: from IPCC's reports and from NOAA/ NCDC), content on this document is made available under a Creative Commons Attribution-NonCommercial 3.0 License.

Abstract

Intergovernmental Panel on Climate Change in its fifth assessment report (WGI AR5)1 claims2:(A.1) Based on the combined evidence from observed climate change including the observed 20th century

warming, climate models, feedback analysis and paleoclimate, [...], [equilibrium climate sensitivity] ECS is likely in the range 1.5°C to 4.5°C with high confidence {TFE.6 p.83 (99/1552)}. No best estimate for ECS is given because of a lack of agreement on [that estimation] {TFE.6 p.85 (101/1552)}.

(A.2) It is extremely likely that human influence has been the dominant cause of the observed warming since the mid-20th century {D.3 p.17 (33/1552)}. Human influence on the climate system is clear. This is evident from the increasing greenhouse gas concentrations in the atmosphere, positive radiative forcing, observed warming, and understanding of the climate system {D. p.15 (31/1552)}.

(A.3) Global surface temperature change for the end of the 21st century is likely to exceed 1.5°C relative to 1850 to 1900 for all [representative concentration pathway] RCP scenarios except RCP2.6 {E.1 p.20 (36/1552)}. This claim is better explained three paragraphs below: Relative to the average from year 1850 to 1900, global surface temperature change by the end of the 21st century is projected to likely exceed 1.5°C for RCP4.5, RCP6.0 and RCP8.5 (high confidence) {E.1 p.20 (36/1552)}.

IPCC's working group I is made up of 259 authors/editors, collecting more than 600 contributions from external experts. WGI seem to have based these claims in: mathematics (statistics, control theory, Monte Carlo simulations, ...) and thermodynamics (conservation of energy, heat transfer, ...).

I intend to refute these claims, reached by IPCC's WGI in their AR5, by showing how mathematics and thermodynamics have been used in an inappropriate way. It might seem quixotic that a person, on its own, dares to challenge what hundreds of those WGI experts agreed; but I will explain that:

(A.a) deduction of climate sensitivity value (due to CO2 doubling) is basically science fiction because, when getting the ECS value (in that range [1.5 – 4.5] ºC) through any method: estimations from models, feedback analysis, instrumental or proxy derivations; there are either inaccuracies or subjective settings that imply tuning. Thus, ECS ends up being a fictitious value.

(A.b) attributing the observed increase in global surface temperature as being man-made is incorrect. WGI AR5 bases his attribution in figure 8.16, but this is not correct because Monte Carlo techniques cannot be used if the uncertainty value comes from having a lack of knowledge.

(A.c) CMIP5 models, in WGI AR5, are not reliable and its projections have no predictive capacity (i.e., the multimodel method do not provide statistical predictability). An appropriate statistical treatment implies a deep paradigm shift in climatic science: only by accumulating climatic data for +900 years, climate could be predicted in the following 30 to 90 years.

Key words: IPCC, AR5, climate, sensitivity, CMIP5, statistics, timescales, proxy, attribution.

1 "WGI AR5" is the document: "Climate Change 2013: The Physical Science Basis. Working Group I Contribution to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press".

2 Along this paper: WGI AR5 words would be cited in italics. Brackets "{}" say where to find each cite in WGI AR5.

– 1 –

1.- Glossary. 1.1.- Climate: is the statistical description in terms of the mean and variability of relevant quantities over a period of time ranging from months to thousands or millions of years. The classical period for averaging these variables is 30 years, as defined by the World Meteorological Organization. The relevant quantities are most often surface variables such as temperature, precipitation and wind. {AIII, p.1450 (1466/1552)}. 1.2.- Weather: is the state of the atmosphere at a given time and place. It can change hourly or dayly. 1.3.- Climate model (spectrum or hierarchy): a numerical representation of the climate system based on the physical, chemical and biological properties of its components, their interactions and feedback processes, and accounting for all or some of its known properties. The climate system can be represented by models of varying complexity, that is, for any one component or combination of components a spectrum or hierarchy of models can be identified, differing in such aspects as the number of spatial dimensions, the extent to which physical, chemical or biological processes are explicitly represented, or the level at which empirical parametrizations are involved. Coupled atmosphere-ocean general circulation models (AOGCMs) provide a representation of the climate system that is near the most comprehensive end of the spectrum currently available. There is an evolution towards more complex models with interactive chemistry and biology. Climate models are applied as a research tool to study and simulate the climate, and for operational purposes, including monthly, seasonal and interannual climate predictions. {AIII, p.1450 (1466/1552)}. 1.4.- Global mean surface temperature (GST): GST is an estimate of the global mean surface air temperature. However, for changes over time, only anomalies, as departures from a climatology, are used, most commonly based on the area-weighted global average of the sea surface temperature anomaly and land surface air anomaly. {AIII, p.1455 (1471/1552)}. 1.5.- Extreme weather event: is an event that is rare at a particular place and time of year. Definitions of rare vary, but an extreme weather event would normally be as rare as or rarer than the 10th or 90th

percentile of a probability density function estimated from observations. By definition, the characteristics of what is called extreme weather may vary from place to place in an absolute sense. When a pattern of extreme weather persists for some time, such as a season, it may be classed as an extreme climate event, especially if it yields an average or total that is itself extreme (e.g., drought or heavy rainfall over a season).{AIII, p.1454 (1470/1552)}. 1.6.- Paleoclimate: Climate during periods prior to the development of measuring instruments, including historic and geologic time, for which only proxy climate records are available. {AIII, p.1459 (1475/1552)}. 1.7.- Proxy: A proxy climate indicator is a record that is interpreted, using physical and biophysical principles, to represent some combination of climate-related variations back in time. Climate-related data derived in this way are referred to as proxy data. Examples of proxies include pollen analysis, tree ring records, speleothems, characteristics of corals and various data derived from marine sediments and ice cores. Proxy data can be calibrated to provide quantitative climate information. {AIII, p.1460 (1476/1552)}. 1.8.- Radiative forcing [R]: is the change in the net, downward minus upward, irradiance (expressed in W·m–2) at the tropopause (or top of atmosphere, TOA) due to a change in an external driver of climate change, such as for example, a change in the concentration of carbon dioxide or the output of the Sun. Sometimes internal drivers are still treated as forcings even though they result from the alteration in climate, for example aerosol or greenhouse gas (GHG) changes in paleoclimates. The traditional radiative forcing is computed with all tropospheric properties held fixed at their unperturbed values, and after allowing for stratospheric temperatures, if perturbed, to readjust to radiative-dynamical equilibrium. (For the purposes of this report, radiative forcing is further defined as the change relative to the year 1750 and, unless otherwise noted, refers to a global and annual average value). {AIII, p.1460 (1476/1552)}. Global radiative forcing from doubling the concentration of carbon dioxide is in [Ar13] represented by: R2x. 1.9.- Climate feedback: an interaction in which a perturbation in one climate quantity causes a change in a second, and the change in the second quantity ultimately leads to an additional change in the first. A negative feedback is one in which the initial perturbation is weakened by the changes it causes; a positive feedback is one in which the initial perturbation is enhanced. In this Assessment Report, [...] the climate quantity that is perturbed is the global mean surface temperature, which in turn causes changes in the global radiation budget. In either case, the initial perturbation can either be externally forced or arise as part of internal variability. {AIII, p.1450 (1466/1552)}.

– 2 –

1.10.- Climate feedback parameter [α with units W·m–2·°C–1]: a way to quantify the radiative response of the climate system to a global mean surface temperature change induced by a radiative forcing. It varies as the inverse of the effective climate sensitivity. Formally, it is defined as: α = (ΔQ – ΔF)/ΔT, where Q is the global mean radiative forcing, T is the global mean air surface temperature (GST), F is the heat flux into the ocean and Δ represents a change with respect to an unperturbed climate. {AIII, p.1450 (1466/1552)}. In [Ar13], effective climate feedback parameter (λeff) is obtained through equation (2): H = λeff T + R (where H is the global mean energy imbalance given by the net radiation flux at the top of atmosphere, T is the global mean surface temperature anomaly and R is the radiative forcing), or equation (3): Teff = – R2x/λeff

(where Teff is the effective climate sensitivity, R2x is the global radiative forcing from doubling the concentration of carbon dioxide; so Teff may be viewed as the climate sensitivity implied by λeff), and also the climate feedback at equilibrium could be set as: λeq = – R2x/T2x. Along this document, we will label equation: H = R + λeff T as [eq.1] and ΔF = ΔQ – α·ΔT as [eq.2]. As they are equivalent, we could say that R is equivalent to ΔQ, H is equivalent to ΔF (i.e., the global mean energy imbalance given by the net radiation flux at the top of atmosphere is equivalent to the heat flux into the ocean) and α = – λeff. So then, α = R2x/Teff that “varies as the inverse of the effective climate sensitivity”. 1.11.- Climate sensitivity: In IPCC reports, equilibrium climate sensitivity [ECS] refers to the equilibrium (steady state) change in the annual global mean surface temperature following a doubling of the atmospheric equivalent carbon dioxide concentration [T2x]. Owing to computational constraints, the equilibrium climate sensitivity in a climate model is sometimes estimated by running an atmospheric general circulation model coupled to a mixed-layer ocean model, because equilibrium climate sensitivity is largely determined by atmospheric processes. Efficient models can be run to equilibrium with a dynamic ocean. The climate sensitivity parameter (units: °C (W m–2)–1) refers to the equilibrium change in the annual global mean surface temperature following a unit change in radiative forcing. Effective climate sensitivity, [Teff, in °C]: is an estimate of the global mean surface temperature response to doubled carbon dioxide concentration that is evaluated from model output or observations for evolving non-equilibrium conditions. It is a measure of the strengths of the climate feedbacks at a particular time and may vary with forcing history and climate state, and therefore may differ from equilibrium climate sensitivity. Transient climate response [TCR] (units: °C) is the change in the global mean surface temperature, averaged over a 20-year period, centred at the time of atmospheric carbon dioxide doubling, in a climate model simulation in which CO2 increases at 1% yr–1. It is a measure of the strength and rapidity of the surface temperature response to greenhouse gas forcing. {AIII, p.1451 (1467/1552)}.

2.- Comments related to these definitions. 2.1.- Differences in 'weather, by measurement' vs. 'climate, by estimation'. Meteorological models are built to predict weather in a region of our planet: usually within daily timescales. Although they can be expanded (with less accuracy) up to monthly timescales. Climate models reach the global Earth scale and they span (see §1.3): from a few months, until many years. Let's notice that, for example, while meteorologist can measure the temperature of a place in a given time: climate temperatures (like GST) are not measured, but statisticaly estimated (see §1.4). In general, values from all climatic parameters (e.g.: precipitation, wind, a radiative forcing, climate sensitivity, a climate feedback, ...) must be obtained through statistical estimations. That is why WMO set those 30 years (see §1.1) for averaging variables: because, by statistical convention, a minimum of 30 samples is required to apply the central limit theorem and to accurately obtain the statistical inference: meaning that inferences from less samples might be inaccurate.

2.2.- Climatic value comparisons are only valid between independent baseline averages. National oceanic and atmospheric administration / National climatic data center (NOAA / NCDC), in its annual report of the State of the Climate, estimates global surface temperature (GST) by averaging: land and ocean's temperature. In 2012's report the temperature anomalies, [Sa13], were displayed as in figure 1. Notice that the averaged anomaly in the base period 1981-2010, (that gray dotted line), has been set to 0ºC; so any averaged anomalies have to be compared with the one in that period.

– 3 –

Fig.1.- Global average surface temperature annual anomalies (ºC, 1981-2010 base period).

Just by looking at the graph anyone can understand that, despite it is true, it is not statistically correct to say that between 1944 and 2012 there has been a global mean surface temperature (GST) increment of 0.2ºC, or that between 1950 and 2010 the GST increment was of around 0.75ºC. It is statistically correct to say that in the base period 1951-1980 respect to 1981-2010, GST increased in around 0.25ºC; or that from the period 1891-1920 to 1921-1950, GST increased nearly in 0.15ºC. In general, the values of any climatic parameters can be only appropriately compared if we use independent baseline averages (the word 'independent' means that it is not appropriate to compare an averaged value in the period 1891-1920 with another in the period 1901-1930 because, see §1.1, less than 30 years in comparison might provide inaccurate inferences; disabling statistical method).

2.3.- Climate feedback definition requires a linear and independent climate system. 'Climate feedback' definition comes from control theory under the fundamental assumption that climate system behaves linearly and independently. About climate forcing mechanisms, in [Cu99] Ch. 13, it is said: “In principle, the contribution of each mechanism to the total feedback could be individually determined and ranked. In a nonlinear system, however, the feedbacks are not independent and addition of the individual terms will not give the true feedback of the nonlinear climate system. Applications of this type of linear feedback analysis have been made to the climate system, justified by considering only small perturbations to the radiative flux and surface temperature”. Thus, as our climate system is nonlinear and the feedbacks are not independent, the only way to appropriately apply the control theory to climate feedbacks is with that requirement of “small perturbations”.

2.4.- No scientific evidence supports the idea of climate change causing extreme weather. Definition §1.5 says that extreme weather event, if “persists for some time, such as a season, it may be classed as an extreme climate event”. So then, does climate change cause extreme weather or extreme climate events?. WGI AR5, in p.121 (137/1552), answers: “Climate change, whether driven by natural or human forcing, can lead to changes in the likelihood of the occurrence or strength of extreme weather and climate events or both”. But in p.928 (944/1552) the answer is: “In the present climate, individual extreme weather events cannot be unambiguously ascribed to climate change, since such events could have happened in an unchanged climate [...] it is incorrect to ascribe every new weather record to climate change”. So it is difficult to attribute climate change (and even more difficult to attribute humans) any pattern's modification of extreme weather. Anyhow, the book “Coles, S., 2001: An Introduction to Statistical Modeling of Extreme Values” has been referenced by the authors of the “Special Report. Managing the risks of extreme events and disasters to advance climate change adaptation (SREX)” as a way to study extreme weather. Using extreme value statistical theory to extrapolate the historical data is useful, for example, in structural engineering. But it is inappropriate for studying extreme weather because, even with that requirement for the climate of small perturbations, seen in §2.3, many parameters in weather events usually vary non-linearly and non-independently (see §2.1, and compare §1.1 vs. §1.2).

– 4 –

2.5.- Different radiative forcing sources, have different level of understanding (LOSU). There are many radiative forcing (RF) sources: solar irradiance, cloud albedo effect, greenhouse gases, ozone, … Figure 2, (modified from fig. TS.5 or fig. 2.20a of the WGI AR4 [IP07]), contains some RF sources. In fig. 2 we see that the level of scientific understanding for greenhouse gases is much higher than the one for cloud albedo effect. This LOSU difference is maintained, in WGI AR5, at Table 8.5 {p. 694 (710/1552)}. Table 8.6 says that aerosol-cloud interactions are “not estimated”, while, in figure 8.16, that RF value is estimated up to nearly: –0.45 (–1.20 to +0.00) Wm-2.

Fig. 2.- Global mean radiative forcings (values, spatial scale and level of understanding)

As we have seen, in §2.1, climate RFs (or any other climatic parameter) cannot be measured: only estimated. The estimation of each of these climatic parameters should consist in averaging, on a 30 (or 50, or 100, ...) year period (e.g, 1981-2010), the radiative forcing corresponding to each related source. But in IPCC reports these estimations are given for a period that always start in 1750 and ends in the previous years to the publication of each report.

2.6.- Spectroscopic analysis allow to model the radiative forcing of greenhouse gases. Radiative forcing for each greenhouse gas can be modeled. For example, carbon dioxide model is:

ΔQCO2 = a·ln(C/C0) where C0 = 279.00 ppmv is the reference concentration (of the year 1750), and radiative forcing variation of CO2 (ΔQCO2) has a logarithmic dependence with CO2

tropospheric concentration C. The value of the 'a' coefficient is deduced from spectroscopic analysis of CO2 absorption bands in HITRAN database. This is an example of a reliable model.

In reference [My98] this coefficient is deduced and set to: a = 5.35, so if doubling CO2 concentration at the tropopause, radiative forcing would be: ΔQ2xCO2 = 5.35·ln2 = 3.701 W·m–2 [eq.3] The rest of greenhouse gases (CH4, N2O, ...) can be modeled as equivalent CO2 (eq-CO2) RFs.

2.7.- Hierarchy of climate models: how complex models are validated by simpler ones. Complex climatic models like coupled atmosphere-ocean general circulation models (AOGCM) try to represent physical processes in Earth by combining atmospheric models with other model types. General circulation models (GCM) can be based in 3-dimensional atmosphere circulation models (ACM), all them based in one-dimensional radiative convective models (RCM): where it is modeled how radiation properties of tropospheric greenhouse gases might affect Earth's surface temperature. RCM assume that there is a linear relation between global surface temperature and the radiative forcing corresponding to doubling the atmospheric CO2 concentration (a parameter that has been called R2x in §1.10 and) that, from Charney's report [Ch79], can be called ΔQ2x. So then, global mean surface temperature variation ΔT was modeled by: ΔT2x = ΔQ2x/α [eq. 4] Despite [eq. 4] is an old assumption it is not a scientific fact as nobody has proven, scientifically, that linear relation between GST and ΔQ2x: the most plausible attempt being refuted in §3.1.1.

– 5 –

3.- Refuting IPCC's claims on climate change.3.1.- Deduction of climate sensitivity value, due to CO2 doubling, is basically science fiction. From definitions, §1.10 and §1.11, climate sensitivity inversely changes with 'climate feedbacks'. ECS is then proportional to 1/α (being α = – λeff, from [eq.1] & [eq.2] comparison). Climate sensitivity value, due to CO2 doubling, has not been logically deduced along climatic literature because when ECS value, in that range [1.5 – 4.5] ºC, has been obtained through any method: estimations from models, feedback analysis, instrumental or proxy derivations; there has always been, or a subjective setting or inaccuracies, implying tuning. Thus, ECS becomes a fictitious value.

3.1.1.- “Derivation” of climate sensitivity value with no feedbacks via Stefan-Boltzmann equation From Stefan-Boltzmann equation (Q = σ·T4), climate sensitivity value with no feedbacks (ΔT0) is obtained (dQ/dT = 4·σ·T3 = α0, σ = 5.67·10–8 Wm−2K−4) by setting T=Ta ≈ 255 K (Ta being TOA temperature): α0 = 3.76 Wm-2K-1; then, (by eq.3 & 4, ΔQ2x = α0·ΔT0 = 5.35·ln 2): ΔT0 ≈ 1K. Another tuning including variations on latitudes/seasons sets: ΔT0 ≈ 1.1237 K (α0 ≈ 3.3 Wm-2K-1). In [Ce76] 'derivation': emissivity ε is set to ε ≈ 0.6 and T=Ts ≈ 288 K (TS being the actual GST), to provide the usual tuned value for α0 (α0 = 4·ε·σ·Ts

3 ≈ 3.3 W·m-2·K-1). A 2006 thesis inappropriately uses the equation: Q = σTs

4– εa·σ·(Ts4–Ta

4), [as: 240=390–150], because the atmospheric emissivity (εa = 0.4–0.8), is set to 1 in this calculation to match the usual 150 Wm-2 value for the greenhouse effect. All these “deductions” via Stefan-Boltzmann do not make sense. Too many thermodynamic assumptions would be needed for translating appropriately ~5.35·ln 2 ≈ 3.7 Wm-2, of more radiative forcing of equivalent CO2 doubling, into a ΔT0 ~1K temperature increase in Earth's surface.

3.1.2.- “Derivation” of climate sensitivity value with no feedbacks via the 1-D atmospheric model Climate sensitivity value with no feedbacks could be also “derived” from a one dimensional radiative-convective model, (1D-RCM), which computes temperature as a function of altitude. In WGI AR5 this 1D-RCM is used, in figure 8.1 {p.669 (685/1552)}, only in a pedagogical way. In [Ha81], the mean surface temperature (Ts) is set in terms of the effective Earth's radiating temperature (Te) in this 1D-RCM: Ts ≈ Te + Γ·He; where He is the flux-weighted mean altitude of the emission to space, and Γ is the mean temperature gradient (lapse rate) between surface and He. Any atmospheric book, e.g. [Cu99], shows that average lapse rate in troposphere is Γ ≈ 6.5ºC/km. RCM seems to deduce climate sensitivity value in a proper way: as infrared absorbers increase in concentration, He increases (by a set ΔH0,e ≈150 m, if CO2 concentration doubles), and Ts increases proportionally (ΔT0,s ~ 150m·6.5K/km ≈ 1 K). That ΔH0,e ≈ 150 m seems to be justified by some kind of atmospheric opacity but, in fact, it is another tuning of parameters that do not make sense.

3.1.3.- Climate sensitivity feedback values obtained from that one dimensional atmospheric model Many feedbacks can be added to this 1D-RC model. In [Ma67] setting a feedback related to average cloudiness and fixed absolute humidity, drives climate sensitivity to ΔTc+ah,s ≈ 1.33 K; and to fixed relative humidity, to: ΔTrh,s ≈ 2.36 K. In [Ha81], there are set six different feedback types, giving: ΔT1,s ≈ 1.2 K, ΔT2,s ≈ 1.9 K, ΔT3,s ≈ 1.37 K, ΔT4,s ≈ 2.7 K, ΔT5,s ≈ 2.6 K, ΔT6,s ≈ 3.5 K. But again, as 1D-RCM allow tuning, these climate sensitivity feedback values are not deduced appropriately.

3.1.4.- Climate sensitivity values based in instrumental data are inaccurate mainly due to timescales Climate sensitivity is derived from instrumental data through compared measurements of energy flux vs. surface temperatures: α·ΔT = Q – N, [eq.5] where N is the net energy imbalance at TOA (top of atmosphere), Q is the global mean radiative forcing (mainly from greenhouse forcings and also from volcanic eruptions) and ΔT: either the GST or the surface sea temperature variation. Many inaccuracies have been detected in this Earth radiation budget “experiment” (ERBE) approach: in [Fo06], timescales span only for 5 years and that correlation (between Q – N and ΔT) diverges too much from 1. And [Li09] has the same problems: much less samples than 30 and correlations too different from 1. This ERBE's approach is so inaccurate, that same analysis leads to opposite estimations: in [Fo06], ECS is in the [1.0, 4.1] K range; while in [Li09], ECS = 0.5 K.

– 6 –

3.1.5.- Climate sensitivity values based in paleoclimate are inaccurate due to uncertainties In IPCC's reports, paleoclimate means: “climate during periods prior to the development of measuring instruments” (see §1.6). In WGI AR5, 'paleo' climate sensitivity is studied: either through a model approach (trying, with some models, to constraint climate sensitivity parameters) or by reconstructing the global mean temperature and the radiative forcings. A paleoclimate model's study, of late Paleocene era (about 55 million years ago), concludes: “increasing CO2 concentrations from 560 to 840 ppm yields an additional surface warming of 3.8K that translates into an ECS of 6.5K”. WGI AR5, in chapter 5.3.3.2 'LGM constraints on ECS', concludes: “New estimates of the equilibrium climate sensitivity based on reconstructions and simulations of the Last Glacial Maximum (21,000 years to 19,000 years ago) show that values below 1°C as well as above 6°C for a doubling of atmospheric CO2 concentration are very unlikely” {p. 385 (401/1552)}. But no matter what paleo-time is studied (millions, thousands or hundreds of years ago): uncertainties in estimating radiative forcings and in estimating global temperature are too big. Error bars (see, for example, WGI AR5 Figure 5.2 {p.395 (411/1552)}) tend to grow as we move to the past; spanning not only in the vertical axis (in CO2 RF or GST), but in the time axis. Thus, reconstructing CO2 RF, or GST, vs. time: becomes a highly inaccurate issue.

WGI AR5, see (A.1), estimates equilibrium climate sensitivity (ECS) feedback values: likely in the range 1.5°C to 4.5°C with high confidence {TFE.6 p.83 (99/1552)}. But along this section, §3.1, it has been shown that no clear logical deductive estimation of ECS values can be made. In all studies, from models or feedback analysis, seen: there is always a subjective setting that implies tuning. And that, in instrumental or proxy studies, there are too big inaccuracies involved. Thus, no plausible climate sensitivity value can be obtained: ECS ends up being a fictitious value range.

3.2.- Ascribing, as man-made, the observed increase in global surface temperature is incorrect The main message of WGI AR5 is: “it is virtually certain that human influence has warmed the global climate system [...] There is strong evidence that excludes solar forcing, volcanoes and internal variability as the strongest drivers of warming since 1950” {p. 73 (89/1552)}. But the same technique that uses IPCC to justify that human influence has warmed the Earth; can be used to show that solar forcing is the main driver of climate change. As a same technique can not provide two opposite results, its abuse is a non sense. Thus, we conclude that it is incorrect to attribute global warming to humans.

3.2.1.- Monte Carlo technique: a statistical tool for attributing humans Earth's temperature increase In documents AR4 and AR5, of IPCC's working group 1 (WGI), Monte Carlo technique is applied in order to add uncertainties from different radiative forcings. Initially this procedure seems correct as statistical techniques, based in Monte Carlo, allow to add asymmetric uncertainties. In Figure 3 we copy: the original WGI AR5's figure 8.16 and, on the left, its corresponding sidenote. WGI AR5's figure 8.16: Probability density function (PDF) of ERF due to total GHG, aerosol forcing, and total anthropogenic forcing. The GHG consists of WMGHG, ozone, and stratospheric water vapour. The PDFs are generated based on uncertainties provided in Table 8.6. The combination of the individual RF agents to derive total forcing over the industrial era are done by Monte Carlo simulations and based on the method in Boucher and Haywood (2001) [Bo01]. PDF of the ERF from surface albedo changes and combined contrails and contrail induced cirrus are included in the total anthropogenic forcing, but not shown as a separate PDF. We currently do not have ERF estimates for some forcing mechanisms: ozone, land use, solar etc. For these forcings we assume that the RF is representative of the ERF and for the ERF uncertainty an additional uncertainty of 17% has been

Fig. 3.- Total anthropogenic RF as an addition of aerosols and greenhouse gases

included in quadrature to the RF uncertainty. Lines at the top of the figure compare the best estimates and uncertainty ranges (5–95% confidence range) with RF estimates from AR4.

– 7 –

The 'theorem of simulation', used in Monte Carlo, allows to apply some statistical techniques (e.g., maximum likelihood, Bayesian method, ...) to sequences of random parameters and to probability-density-functions (PDFs). It also allows to add appropriately any asymmetric uncertainties.According to fig.3, total anthropogenic effective RF is +2.3 [1.1 - 3.3] W·m–2, the result of adding well-mixed greenhouse gases (GHG) effective RF (+3.2 [2.6 - 3.8] W·m–2) to aerosol effective RF (–0.9 [–1.9 - –0.1] W·m–2). An appropriately added value despite of the PDF-shape's adding terms.

3.2.2.- Showing the inappropriate use of science when attributing humans the temperature increase In §3.2.1, it was explained that Monte Carlo techniques allow to add appropriately asymmetric uncertainties. But Monte Carlo techniques cannot be used if the uncertainty value comes from having a lack of knowledge (see LOSU in §2.5); so then, fig. 3 (WGI AR5's figure 8.16) is not appropriate for attributing humans the observed increment in global surface temperature (GST). In fact, IPCC seems to admit this mistake in Table 8.6; because, in there, RF estimates of aerosol-cloud interactions are “not estimated”, as well as, RF estimates of total anthropogenic. And as total anthropogenic RF is not estimated, then we cannot attribute humans the observed global warming. Another odd issue was also mentioned in §2.5. In Table 8.6, RFs seem to be estimated since the year 1750 but, as we saw in §2.2, the estimation of climatic parameters should average +30 year periods in order to be able to compare different base periods and evaluating how climate changes. About reconstructing GHG RF between 1750 and 1900: we saw in §3.1.5 that it is inaccurate. Even worst is trying to reconstruct aerosols RF between 1750 and 2011 because, as we do not have a consistent model to evaluate the aerosol effective RF (see again the LOSU in §2.5), aerosol effective RF's PDF is a fictitious to-work-with-scenario: not an actual value obtained from measurements. A much lower scientific understanding of aerosol RF implies that it may have another central value or that its PDF may have a different shape (the blue line in fig.3 could become the one in fig.4): thus, there is no point in adding RF with largely different LOSU.

3.2.3.- We use also Monte Carlo, as in §3.2.1, to show how solar RF would cause global warmingAfter the issues seen, in §3.2.2, let us invent another to-work-with-scenario. If we believed that solar irradiance radiative forcing causes global warming, we could just invent the data of aerosol effective RF (setting it, for example, to: –2.8 [–3.8 - –2.2] W·m–2), and “obtain” the figure 4, beside, with an effective total anthropogenic RF of: 0.1 [–0.8 - 0.5] W·m–2. In this scenario, as all RF that IPCC's calls “anthropogenic” add up to a negligible value, we might conclude that only solar irradiance contributes to climatic change. Fig. 4.- RFs with an invented Aerosol PDFBut this conclusion is not valid as we are basing it in our previous belief that “only solar irradiance causes climate to change”. This type of reasoning is not scientifically correct.

In summary: – “Estimations” from a lack of knowledge are not the same as estimations from measurements of certain parameters belonging to a reliable model. Monte Carlo techniques cannot be used if uncertainty value is due to having a lack of knowledge (a Lack Of Scientific Understanding). – As seen in §2.2, it has been observed an increase in global surface temperatures (GST) during certain periods of the 20th century. But there is no scientific criteria, based on statistical climate parameter estimations, that allow us to ascribe these global surface temperatures (GST) increments as caused by humans. – As seen in §3.1, climate sensitivity values are not justified by any logical scientific deduction. Thus, climate sensitivity values cannot correlate RF sources with GST.

– 8 –

3.3.- WGI AR5's CMIP5 models are not reliable and its projections have no predictive capacity IPCC, see (A.3), claims that: Global surface temperature change for the end of the 21st century is likely to exceed 1.5°C relative to 1850 to 1900 for all Representative Concentration Pathways (RCPs) scenarios except RCP2.6. This claim is supported by a group of general circulation models (GCMs), called by the IPCC, "Coupled Model Intercomparison Project 5 (CMIP5) concentration-driven experiments". But CMIP5 models are not reliable and their projections have no predictive capacity.

3.3.1.- CMIP5 models relate greenhouse gases RF with Earth's surface temperature Coupled Model Intercomparison Project 5 (CMIP5) general circulation models (GCMs), relate global surface temperature to anthropogenic radiative forcing and to four arbitrarily chosen well mixed greenhouse gases concentration levels (or eq-CO2) in the following way: Representative concentration pathways (RCP) are four arbitrarily chosen scenarios where the concentration level of the well mixed greenhouse gases concentrations (in units of parts per million by volume) has been set, for the next centuries, as in figure 5 (adapted from WGI AR5's Figure 12.42.a. The dashed line indicates the pre-industrial CO2 concentration). Fig. 5.- RCP scenarios of atmospheric CO2

Notice that in figure 5 only atmospheric CO2 is represented, but that the rest of well mixed greenhouse gases can be modeled to contribute as equivalent CO2 so then the model (seen in §2.6): ΔQCO2 = a·ln(C/C0) [a = 5.35 W·m–2, C0 =280 ppmv, ΔQCO2 is the equivalent CO2 radiative forcing and C is time dependent eq-CO2 concentration value]; is applied to get the anthropogenic RF in the figure 6, adapted from IPCC's WGI AR5 Figure 12.3.a, with the adapted footnote: Time evolution of the total anthropogenic radiative forcing relative to preindustrial (~1765) between 2000 and 2300 for RCP scenarios as computed by the integrated assessment models (IAMs) used to develop those scenarios. The four RCP scenarios used in CMIP5 are: RCP2.6 (dark blue), RCP4.5 (light blue), RCP6.0 (orange) and RCP8.5 (red). The total radiative forcing of the RCP family is computed taking into account the efficacy of the various forcings (Meinshausen, 2011a). {p. 1046 (1062/1552)} Fig. 6.- RCP scenarios of anthropogenic RFs CMIP5 models are then built to follow the linear equation, ΔF = ΔQ – α·ΔT, where ΔF is the top of atmosphere (TOA) change in energy imbalance, ΔQ is a change in a climate forcing component, ΔT is the averaged GST change and α are climate feedback components (with units: W·m–2·K–1). This linear model, [eq.2], is equivalent to eq. (1) in [Fo13] and allows to show ΔT in terms of ΔQ in the figure 7, copied from IPCC's WGI AR5 Figure 12.5, with the adapted footnote: Time series of global annual mean surface air temperature anomalies (relative to 1986–2005) from CMIP5 concentration-driven experiments. Projections are shown for each RCP for the multi model mean (solid lines) and the 5–95% range (±1.64 standard deviation) across the distribution of individual models (shading). Discontinuities at 2100 are due to different numbers of models performing the extension runs beyond the 21st century and have no physical meaning. [...] No ranges are given for RCP6.0 projections beyond 2100 as only two models are available. {p. 1054 (1070/1552)}

Fig. 7.- Global surface temperature projected from the above four RCP scenarios

– 9 –

3.3.2.- Showing the inappropriate use of science in CMIP5 models It is scientifically inappropriate to use CMIP5 models for justifying IPCC's claims like (A.3). In §3.3.1: fig. 6 was obtained from fig. 5, by using a model where that 'a' coefficient was deduced scientifically from spectroscopic analysis of the CO2 absorption bands in the HITRAN database. But this is the unique normal use of science in this CMIP5 model issue. Inappropriate use is listed:

– The well mixed greenhouse gases RF (or in §3.3.1 also labeled as equivalent CO2 radiative forcing: eq-CO2 RF) and the total aerosol RF are, see §3.2.2, the main contributors to anthropogenic radiative forcing. In the original fig. 6 (IPCC's WGI AR5 Figure 12.3.a) aerosol RF tends to be less than 0.2 W·m–2; but, unless IPCC gets a reliable model for aerosol RF, it is inappropriate to use this invented value. Thus, fig. 6 is not appropriately derived from fig. 5.– Figure 7 is not appropriately derived from figure 6 because CMIP5 models use invented climate feedbacks, α, with no logical deduction of their values. Let's explain that solid lines in fig. 7, assume that the value of climate sensitivity is around 2 ºC, (i.e., that for a doubling eq-CO2

concentration, global surface temperature will increase in 2 ºC). Increments in climate sensitivity values (inversely proportional, see §1.10, to climate feedbacks) will elevate those solid lines. In these CMIP5 models: climate sensitivity values are, thus, used inappropriately. – Figure 7 is also not correctly derived from figure 6 because the linear relation: ΔF = ΔQ – α·ΔT, see §3.3.1, is a non-scientifically demonstrated hypothesis. CMIP5 linearity guarantees that each GST variation output follows the related RCPs input; but it is inappropriate to oversimplify complexity like this, as in many cases: oversimplification leads to getting huge errors. Anyhow, if according to §2.3 IPCC wishes to keep this linear relation by using only small perturbations, then it is a nonsense to set scenarios like 6.0 or 8.0 where concentrations and GST grow rapidly. – In this same linear relation [eq.5] the value for the top of atmosphere (TOA) change in energy imbalance, ΔF, is also inaccurate (as in §3.1.4) implying that big errors will be obtained.– From the different statistical tools available: stochasticity, perturbations or multi-model methods; CMIP5 models are based in the multi-model approach. (CMIP5 models seem to include also stochasticity: as ripples in solid lines must be linked to random, i.e.: stochastic, uncertainties). But the footnote in figure 12.5 (fig. 7) “projections are shown for each RCP for the multi model mean (solid lines) and the 5–95% range (±1.64 standard deviation)” is a nonsense. Because CMIP5 models, despite using these statistical tools, do not provide any statistical predictability. As scenarios are arbitrarily chosen and all the key values in CMIP5 models (ΔF, ΔQ and α) are fictitious: it is inappropriate to extract conclusions from these models. IPCC labels their predictions as “projections”, but we know that their models are not reliable. CMIP5 multi-models, based in fictitious values and invented scenarios, have no predictive capacity. – WGI AR5 says: Projections of changes in the climate system are made using a hierarchy of climate models ranging from simple climate models, to (...) Earth System Models. These models simulate changes based on a set of scenarios of anthropogenic forcings {p.19 (35/1552)}. Hierarchy of models, defined in §1.3, do not seem to be important as IPCC uses, in WGI AR5, CMIP5 models. But it is a key concept that explains, see §2.7 & §3.1.2, why anthropogenic climate change theory is a non sense. IPCC's WGI in AR4, [Ip07] in chapter 1.5, explained why hierarchy (see §2.7) had to be used in 'climate models': With the development of computer capacities, simpler models have not disappeared; on the contrary, a stronger emphasis has been given to the concept of a ‘hierarchy of models’ as the only way to provide a linkage between theoretical understanding and the complexity of realistic models. This is an euphemistic way to say that models validate models (as seen in §2.7); but this not how scientific method should work. In normal science, what it really matters is that models should only be validated by observation, of course, in the appropriate climate change timescale.

IPCC believes that atmospheric emissions of equivalent CO2 cause global warming. IPCC models set different scenarios of eq-CO2 emissions (RCP scenarios) projecting different effects in global surface temperatures. But the methodology used: conflicting assumptions and arbitrarily preset values; invalidate such models, leaving them with no predictive ability.

– 10 –

3.3.3.- Hiatus in global warming and how WGI AR5 explains it WGI AR5's Figure 11.25.a is, in this document, copied into figure 8 with the adapted footnote: Simulations and projections of annual mean GMST 1986–2050 (anomalies relative to 1986–2005). Projections under all RCPs from CMIP5 models (grey and coloured lines, one ensemble member per model), with four observational estimates [...] for the period 1986–2012 (black lines) {p. 1011 (1027/1552)}.Notice that observational estimates in fig. 8, are different from the historical estimates in fig. 7.

Fig. 8.- Simulations and projections of global mean surface temperatures with observations

Observational estimates in global mean surface temperatures between the years 1998 and 2012, seem to be flat: neither increasing, nor decreasing. This is what it has been called 'hiatus': a period of time when GST seem to have paused, despite CO2 concentrations have kept rising as usual. The explanation, in Chapter 9 of WGI AR5, for this hiatus is: this difference between simulated and observed trends could be caused by some combination of (a) internal climate variability, (b) missing or incorrect radiative forcing and (c) model response error. {p.769 (785/1552)}:

(a) Internal Climate Variability: Overall, there is medium confidence that […] the hiatus is in part a consequence of internal variability that is predictable on the multi-year time scale. {p.770 (786/1552)}.(b) Radiative Forcing: Although the forcing uncertainties are substantial, there are no apparent incorrect or missing global mean forcings in the CMIP5 models over the last 15 years that could explain the model–observations difference during the warming hiatus. {p.770 (786/1552)}.(c) Model Response Error: The discrepancy between simulated and observed GMST trends during 1998–2012 could be explained in part by a tendency for some CMIP5 models to simulate stronger warming in response to increases in greenhouse gas (GHG) concentration than is consistent with observations {p.771 (787/1552)}.

You can check how (b) agrees with sub-chapters §3.2.2 and §3.2.3; or how (c) agrees with sub-chapters §3.3.2 and §3.1. That 'internal variability' makes us wonder about appropriate timescales.

3.3.4.- Showing how science can be used appropriately in climate change predictions WGI AR5 models, in chapter 11 (Near Term Climate Change: Projections and Predictability) and in chapter 12 (Long-term Climate Change: Projections, Commitments and Irreversibility), span along different timescales: nearly 60 years, around 65 years, 450 years and over a thousand years; seen, respectively, in WGI AR5 figures: Box 11.1 Fig.1, Fig.11.9, Fig.12.5 and Fig.12.43.b. Climate seems to be influenced by that internal variability (seen in §3.3.3) in the multi-year time scale. But, which are the appropriate timescales for climate change models?. As the minimum number of years/samples to obtain accurate statistical inferences is 30 (as a convention, see §2.1). Then, every 30 years climatic parameters can be averaged. Mankind has measured climatic parameters accurately only from around the year 1950. Thus, we can only evaluate (see §2.2) two data: belonging to the base periods, 1951-1980 and 1981-2010. But the minimum number of base periods to evaluate how climate actually changes is not 2: it should be 30 (see §2.1). Then, by the year 2850 (30·30=900, 900+1950=2850) mankind would have measured enough data to establish accurate correlations between climatic parameters.It is hilarious what IPCC does in WGI AR5 fig. 12.43.b: running a climate model “according” to the last 50 years of observed values and inferring the climate behavior of the next 1000 years. Statistics dictates that, instead, we need to observe (to measure climatic parameters) for 900 years and only then, by the year 2850, we will be capable of predicting climate change trends and establishing appropriate attributions for the following decades (up to around the year 2940). Notice that by setting around 1000 years as the appropriate timescale in climate change models: climate change debate has to be rethought. This is a deep paradigm shift in climatic science: as all climatic speculations dealing with decadal timescales models might be useless; including those against IPCC's claims ([Fy13]), those in favor or those new ones related to internal variability.

– 11 –

4.- Concluding remarks. IPCC has never justified why if doubling eq-CO2 concentration, global mean surface temperature (GST) will increase in a range of 1.5 to 4.5 ºC. These values appeared historically as the result of an inappropriate use of thermodynamics (Stefan-Boltzmann equation & radiative-convective models). WGI AR5 does not justify why humans are mainly responsible of climate change: Monte Carlo simulations have been inappropriately used to “justify” that humans cause climate change and substantial uncertainties in radiative forcings disable any reliable calculation of Anthropogenic RF. WGI AR5 does not offer any reliable prediction on climate change. Their “projections” have no predictive abilities. Many scientist (may be all of them!) do not use appropriate timescales in the climatic change debate: only after hundreds of years of climatic analysis, mankind will be statistically capable to accurately predict future trends and to address appropriate attributions.

5.- Acknowledgments. From the people I contacted with, in blogs or by email, I would like to thank those who helped me in improving this document. I also want to thank you, if you free distribute this document.

6.- References.[Ar13] Armour K.C. et al.; 2013, Time varying climate sensitivity from regional feedbacks, AMS.[Bo01] Boucher, O. & Haywood, J; 2001, On summing the components of radiative forcing of climate change, Climate dynamics, vol. 18, p. 297-302.[Ce76] Cess, R.D.; 1976, Climate change: an appraisal of atmospheric feedback mechanisms employing zonal climatology, Journal of atmospheric sci., vol. 33, nº 10.[Ch79] Charney, J.G. et al.; 1979, Carbon dioxide and climate, National Academy of Sci. Washington D.C.[Cu99] Curry, J.A & Webster, P.J; 1999, Thermodyna-mics of atmospheres and oceans, Academic Press, V. 65[Fo06] Forster P.M & Gregory J.M, 2006, Climate sen-sitivity & its components diagnosed from ERBE, AMS.[Fo13] Forster, P.M., et al., 2013, Evaluating adjusted forcing & model spread for historical & future scenarios in the CMIP5 generation of climate models, J. Geoph. Res. Atm., 118, 1139–1150, doi:10.1002/jgrd.50174

[Fy13] Fyfe et...; 2013, Overestimated global warming over the past 20 years. Nature Climate Change, 3, 767-9[Ha81] Hansen, et al.; 1981, Climate impact of increa-sing atmospheric carbon dioxide. Science, V.213 nº4511 [IP07] Climate Change 2007: the physical science basis. WGI contribution to AR4 of IPCC, fig.2.20, CUP[Li09] Lindzen R., & Choi Y., 2009, On the determi-nation of climate feedbacks from ERBE, Geoph. R. Let. [Ma67] Manabe, S. & Wetherald, R.T., 1967, Thermal equilibrium of the atmosphere with a given distribution of relative humidity, Journal of Atm. Sci., vol.24,nº3[My98] Myhre, et al, 1998b: New estimates of radiative forcing due to well mixed greenhouse gases. Geophys. Res. Lett., 25, 2715-2718. [Sa13] Sánchez-Lugo, et al., 2012, [Global climate] Surface temperature [in 'State of the Climate in 2012']. Bull. Amer. Meteor. Soc., Vol. 94 (Nº.8), S11, fig. 2.

7.- Appendix. After applying ethics to science: IPCC must justify or rectify. Intergovernmental Panel on Climate Change (IPCC), is a scientific body established by the World Meteorological Organization (WMO) and the United Nations Environment Programme (UNEP). IPCC has a millionaire yearly budget and a global influence in our society. Ethics in science imply that, as science is a difficult task (not everybody can, for example, read and fully understand this document), society trust in scientist to develop science in its appropriate way. But, as it has been shown along this document, WGI AR5 has used science in an inappropriate way. An ethical mandate is: our privileges become an abuse when we do not assume our responsibilities. IPCC cannot accommodate science to their conveniences, so they must either justify scientifically their claims or make public rectification of these scientifically inappropriate claims (I have been pointing out in WGI AR5). It is mandatory that IPCC assumes their responsibilities on WGI AR5 document. The main responsibles of the whole content are Rajendra Pachauri, Qin Dahe and Thomas Stocker. The main responsibles of chapter 12 in WGI AR5 are Matthew Collins and Reto Knutti. The main responsibles of chapter 8 are Gunnar Myhre and Drew Shindell.

– 12 –