We would like to continue to take advantage of the brain trust in our audience to refine and harden the articles on the Everything Climate , one article at a time.
Please continue to contribute input to improve and tighten up both the Pro and Con sections, etc.)
We will start with one article at at time and if this works well, it will become a regular feature.
So here’s the Next one. Please give us your input. If you wish to email marked up word or PDF documents, use the information on this page to submit.
Pro: Climate Models Have Been Demonstrated to be Accurate.
By Alan Buis,
NASA’s Jet Propulsion Laboratory
An animation of a GISS (Goddard Institute for Space Studies) climate model simulation made for the United Nations’ Intergovernmental Panel on Climate Change Fourth Assessment Report, showing five-year averaged surface air temperature anomalies in degrees Celsius from 1880 to 2100. The temperature anomaly is a measure of how much warmer or colder it is at a particular place and time than the long-term mean temperature, defined as the average temperature over the 30-year base period from 1951 to 1980. Blue areas represent cool areas and yellow and red areas represent warmer areas. The number in the upper right corner represents the global mean anomaly.
Credit: NASA’s Goddard Institute for Space Studies
There’s an old saying that “the proof is in the pudding,” meaning that you can only truly gauge the quality of something once it’s been put to a test. Such is the case with climate models: mathematical computer simulations of the various factors that interact to affect Earth’s climate, such as our atmosphere, ocean, ice, land surface and the Sun.
For decades, people have legitimately wondered how well climate models perform in predicting future climate conditions. Based on solid physics and the best understanding of the Earth system available, they skillfully reproduce observed data. Nevertheless, they have a wide response to increasing carbon dioxide levels, and many uncertainties remain in the details. The hallmark of good science, however, is the ability to make testable predictions, and climate models have been making predictions since the 1970s. How reliable have they been?
Now a new evaluation of global climate models used to project Earth’s future global average surface temperatures over the past half-century answers that question: most of the models have been quite accurate.
In a study accepted for publication in the journal Geophysical Research Letters, a research team led by Zeke Hausfather of the University of California, Berkeley, conducted a systematic evaluation of the performance of past climate models. The team compared 17 increasingly sophisticated model projections of global average temperature developed between 1970 and 2007, including some originally developed by NASA, with actual changes in global temperature observed through the end of 2017. The observational temperature data came from multiple sources, including NASA’s Goddard Institute for Space Studies Surface Temperature Analysis (GISTEMP) time series, an estimate of global surface temperature change.
The results: 10 of the model projections closely matched observations. Moreover, after accounting for differences between modeled and actual changes in atmospheric carbon dioxide and other factors that drive climate, the number increased to 14. The authors found no evidence that the climate models evaluated either systematically overestimated or underestimated warming over the period of their projections.
“The results of this study of past climate models bolster scientists’ confidence that both they as well as today’s more advanced climate models are skillfully projecting global warming,” said study co-author Gavin Schmidt, director of NASA’s Goddard Institute of Space Studies in New York. “This research could help resolve public confusion around the performance of past climate modeling efforts.”
Con: Global Ocean Temperatures are Warming at Only ~50% the Rate of Climate Model Projections
Dr. Roy Spencer, a climate scientist from the University of Alabama, Hunstville examined the claims of accuracy of climate models by comparing them to actual measured temperatures.
The plot below (Fig. 1) shows the monthly global (60N-60S) average ocean surface temperature variations since 1979 for 68 model simulations from 13 different climate models. The 42 years of observations we now have since 1979 (bold black line) shows that warming is occurring much more slowly than the average climate model says it should have.
In terms of the linear temperature trends since 1979, Fig. 2 shows that 2 of the top-cited ocean temperature datasets have warming trends near the bottom of the range of climate model simulations.
Deep Ocean Warming Could Be Mostly Natural
A related issue is how much the deep oceans are warming. As I have mentioned before, the (inarguable) energy imbalance associated with deep-ocean warming in recent decades is only about 1 part (less than 1 Watt per sq. m) in 300 of the natural energy flows in the climate system.
This is a very tiny energy imbalance in the climate system. We know NONE of the natural energy flows to that level of accuracy.
What that means is that global warming could be mostly natural, and we would not even know it.
I’m not claiming that is the case. I am merely pointing out the level of faith that is involved in the adjustments made to climate models, which necessarily produce warming due to increasing CO2 because those models simply assume that there is no other source of warming.
Yes, more CO2 must produce some warming. But the amount of warming makes all the difference to global energy policies.
Seldom is the public ever informed of these glaring discrepancies between basic science and what politicians and pop-scientists tell us.
Why does it matter?
It matters because there is no Climate Crisis. There is no Climate Emergency.
Yes, irregular warming is occurring. Yes, it is at least partly due to human greenhouse gas emissions. But seldom are the benefits of a somewhat warmer climate system mentioned, or the benefits of more CO2 in the atmosphere (which is required for life on Earth to exist).
But if we waste trillions of dollars (that’s just here in the U.S. — meanwhile, China will always do what is in the best interests of China) then that is trillions of dollars not available for the real necessities of life.
Prosperity will suffer, and for no good reason.
When the actual study comes out, there are a number of areas to scrutinize. The biggest is the criteria for selecting only 17 models (out of potentially hundreds). Additionally, what were the forcing scenarios used in each? The propagandists use RCP8.5 as BAU, whereas this study may input model forcings of much smaller concentrations. The offensive unit of the CliSci Team uses exaggerated scenarios, while the defensive unit of the CliSci Team points to the reasonable information hidden in the bureaucratic verbiage of massive UN IPCC documents where the summaries contradict the buried science.
As asked earlier, why end the evaluation at 2017? Additionally, why not compare model hindcasts to early 20th Century actuals? We know models don’t pick up significant warming during that period.
In addition to estimated (surface air?) GISTEMP data, what were the other “multiple” sources used to evaluate the models? Averaging many different sorts of data to arrive at a single global average temperature is fraught with many opportunities for noble cause bias.
Is this study meant to “prove” something rather than “explain” something? If the study makes it into the UN IPCC AR6, will rebuttals/comments be included also?
This seems like an oranges-to-lemons comparison. We have two different suites of models compared to two different sets of “actual” observations, one in Celsius and one in Fahrenheit. At a minimum, plot GISTEMP and ERSSTv5 onto both sets of model results.
I have a concern with the title conceding the position to the alarmist cause then having to work back. Instead of the article beginning “Pro: Climate Models Have Been Demonstrated to be Accurate.” I’d have the bold headline “How accurate are climate models” followed by some con arguments, then the Alarmist argument, then reiterate our argument, or something similar.
I understand trying to be completely balanced but many will read the headline and think the alarmist models are correct.
Another way might be to just flip the arguments. But it’s our document don’t hamstring truth. Don’t give away the initiative before the discussion even starts.
Waooh …
The GISS clowns even made disappear the 1998 El Niño … that’s a great achievement in fakery and charlatanism.
The Temperature Data Manipulators had to make 1998 insignficant to enable them to be able to claim that year after year in the 21st Century was the “hottest year evah!”.
If the scam artists had left 1998 were it was originally, then they couldn’t have said “hottest year evah!” until 2016.
The Demoting of 1998, allowed the alarmists to continue their scary tale of unprecedented warming for years.
These guys are accomplished liars.
Fortunately, the UAH satellite chart keeps 1998 in the proper perspective with regard to subsequent years.
The UAH chart:
On the UAH satellite chart, the year 1998 is statistically tied with 2016, for the warmest year since the Early Twentieth Century (the margin of error of the measuring instrument is 0.1C).
NASA and NOAA couldn’t say “hottest year evah!” year after year, if they went by the UAH chart. They would have to keep their mouths shut until they got to 2016.
So NASA and NOAA modified their charts to allow them to lie to the American people and the world about the state of the Earth’s climate.
Is it an error of the ‘instrument’ or of the conversion process from radiances to temperature, which itself requires computer modelling?
Why would NASA and NOAA go by the UAH chart when they are measuring two different things? Also, if the used the RSS chart instead of the UAH chart then there would be good agreement.
Why such faith here that UAH is better than RSS, given the history? At least once RSS forced UAH to amend its record – towards warmer.
“Why such faith here that UAH is better than RSS, given the history?”
Well, the weather balloon data agreeing with the UAH chart data would be one reason I would favor UAH over others. We have two different data sets that agree on temperature.
I woud be suspect of any data set that did not agree with the weather balloon data, which would include all the data sets except UAH.
IIRC: 1) One of the satellite AMSU units (14?) was faulty and UAH removed its data and 2) UAH uses empirical data to adjust for satellite drift and RSS uses models. Finally, I think RSS’s Mears is a nasty twit to keep harping on errors corrected by UAH decades ago.
“But if we waste trillions of dollars (that’s just here in the U.S. — meanwhile, China will always do what is in the best interests of China) then that is trillions of dollars not available for the real necessities of life.”
It’s far worse than this. The West’s “net zero” carbon policy will simply be to no effect in term of stated global CO2 emission reductions needed to meet intentions of the Paris Accord with the entire rest of the world (not just China) already building hundreds of new coal fired plants.
We are talking about ~5 billion people who are already on the road to prosperity (Bangladesh GDP is growing at 15% a year, Pakistan 9-10%, India has large coal reserves and they are going to tender on their massive development, Africa will be building dozens of power stations over the next few years). They have new coal gen stations. Both Al Gore and Kerry have spoken to some of these countries – all have said they are not going to compromise their growth.
Nevermind just the electrification of these countries. With prosperity they will be more than doubling the worlds automobile and truck fleet. They will be needing thousands of cement and lime plants, iron and steel plants,
mining operations for there construction boom over 30-40years. These folk will want air conditioning, travel, nice homes…. I think a rewrite with this very real, imminent foresight is required. CO2 emissions will accelerate and drown out the puny reductions the West’s talking about. The big CO2 experiment is underway and unstoppable.
“The big CO2 experiment is underway and unstoppable.”
That’s the bottom line. When wil the Western alarmists face reality and stop trying to put off the inevitable?
It’s not so bad: There’s no evidence that additional CO2 will cause us any harm. The only harm from CO2 will come from delusional Western leaders spending tax dollars trying to fix something that doesn’t need fixing.
Which RCP is the basis for the model runs?
Figures don’t lie, but liars figure [model].
Truthfulness would demand that the model forecast vs actual include a second y-axis with the model vs actual CO2.
“Moreover, after accounting for differences between modeled and actual changes in atmospheric carbon dioxide” – meaning the models passed after adjustment; we just don’t know all the adjustments.
I am not sure what mr. Rotter wants from us. Pro and con led me to believe I am about to see two points of view on an issue, and maybe a hypothesis/ synthesis at the end. What I see is two sets of data not provably related, and no hypothesis, conclusion or question to bind the two very separate chapters together. Besides, discussing climate on a fifty-year spread of data ignores all those perplexing cycles and subcycles and emergent spikes and geologic upheavals that visitors to this site love to endlessly discuss. In other words, there ain’t no nothing to talk about here.
Or am I being stupid again?
Since no-one knows (including the IPCC) how much of the observed warming is due to the increasing CO2 concentration while the models assume all the observed warming is due to CO2, there is no way the models can ever be validated by the observations.
Is there space for a discussion on the futility of determining an average global anomaly when the concept of a global average temperature is suspect?
I just cut up a sheep and packed it into my freezer. I set the thermostat to -4 degrees Celsius, but the meat is not frozen. The thermometer reads 14 degrees between packets, but -6 at the funny riffled plate at the side of the frieezer.
I am also baking a birthday cake for my cat, the oven is set at 180degrees Fahrenheit.
The thermometer in my smartphone says the lounge table is at 23 Celcius, but in my sunny bathroom, it is 31.
What is the average temperature of my house?
Only a climastrologist could say…
If I were a climastrologist I would respond, “What answer do you want to support your agenda?”.
A climastrologist would use the oven temperature, since it maximizes the warming.
The way the average temperature is calculated depends on the question that is to be answered. Finding the mid-range temperature, as is done for daily temperatures, is probably the least useful approach. However, the point is that the average temperature of the house probably isn’t germane to any practical problem. That is why I have long advocated for showing the range of temperature as two separate lines, and to do it for all the climate regions instead of the whole Earth. You have demonstrated that there is more useful information in the locations of the individual temperature measured in your house.
no doubt about the model for last 40 years , BUT it will change next years due to AMO new cycle … wait and see … It’s the same if you look on a small portion of a sinusoid , you can just see a straight line ….
I’ve been making the argument for years that we need Crowd-Sourced, Open Source Climate Models. Michael Mann’s Hockeystick would never have passed public scrutiny. Just look what Steve McIntyre did to it. That kind of scrutiny needs to be applied to all of Climate Change.
Also, WUWT should go to Dr. Spencer’s Blog and download the data:
https://www.nsstc.uah.edu/data/msu/v6.0/tlt/uahncdc_lt_6.0.txt
Sort the data by Month and then Year.
Create Monthly Charts for the South Pole Data.
You will see that if you remove Water Vapor and the UHI Effect, which is what the S Pole does, you will see that increasing CO2 by 30% has absolutely 0.00% impact on temperatures.
In theory, you are arguing for peer-reviewed publication. Unfortunately, what was once peer review has been corrupted to be pal review.
You cannot argue with folks who consistently put their fingers on the scale and “adjust” the raw data to fit their hypotheses. See here:
And furthermore, the “adjustments” plotted against CO2 is most revealing:
https://realclimatescience.com/2021/03/noaa-temperature-adjustments-are-doing-exactly-what-theyre-supposed-to/
This is outright criminal behaviour. Who are the people who perpetrated this, and why are they not in prison?
With each successive year, the temperature adjustments compared to CO2 concentrations are getting larger and larger exponentially. This would normally imply that year over year measuring devices and/or observational techniques are becoming much worse as time goes along or that the relationship between the two is screwy. Am I not seeing something here?
On the top graph, where did GISS get an “observed” warming of about 0.8 to 0.9 C from 2000 to 2019? Haven’t we been in a “pause” since about 1998? Welcome to the GISS pseudo-scientific method: if the data don’t match the models, fudge the data.
“Obversations” is in the title of Fig 1…oops…
If “Carbon Pollution” is a real thing, we should reject all forms of Carbon, right? I mean Carbon is just bad! How about that Diamond thing, every woman seems to want? It’s pure Carbon, should not be allowed!
“There’s an old saying that ‘the proof is in the pudding’,”
This is absolutely one of my pet peeves. The old saying IS NOT “the proof is in the pudding” the saying IS, “The proof of the pudding [is in the eating]!” Why can’t people get this right?
1 – Nearly two decades ago I reviewed the source for one of the then major models and found: first, that most of the code related to using parallelism to increase apparent resolution; and, second, that the model itself largely consisted of a 1960s fortran card deck with thousands of grad-student style adhoc adjustments and improvements since – many of them effectively hard wiring in stuff to yield better hind casting results.
As a result I have believed ever since that the right way to model climate is to model energy flows into, and out of, cubic regions – i.e. diviy earth + Atmos into blocks of some number of cubic meters and start by trying first to measure and then to model energy flows in and out of those blocks. There are (?) no models like this, although the best practical forecasting tools incorporate some similar ideas.
2 – the pro/con model results cannot be interpreted because both rely on doubtful data, the assumption that “temperature anomalies” exist and are measurable, and “physics models” of the kind that start with a spherical chicken in a vacuum. If I assume that the creators of these charts are honest in their own contexts I’d guess both were completely correct and the apparent contradiction a result of the fact that they reflect different data, different conceptualizations of “anomalies” and different interpretations of past forecasts. Bottom line: neither tells us anything of value.
you’re missing the most jarring discrepancy between model and output: the tropical troposphere anomaly, which is supposed to be the driver of global warming
http://www.drroyspencer.com/wp-content/uploads/CMIP5-73-models-vs-obs-20N-20S-MT-5-yr-means1.png
fortunately extending the graph to present presents no effort at all as the whole update fits neatly under the legend
Anybody, other than Dr. Spencer, know how to do it? If not, does anybody know Dr. Spencer?
Could it be posted to this Thread? Other archive?
do what? the graph is right there already
I may be a ‘short’ Dave, but I can reason with the best of them. The graph is not the data; new data must be developed to insert onto the graph. All of the inputs to the new data must be assembled and manipulated to be consistent with the procedures used to develop the existing graphed data. My mistake was assuming that all readers would know that.
Just downloaded the latest UAH [ uahncdc_mt_6.0.txt, “TRPCS 20S-20N” column ] and RSS [ RSS_Monthly_MSU_AMSU_Channel_TMT_Anomalies_Land_and_Ocean_v04_0.txt, “-20.0/20.0” column ] data files (to April 2021) for something else.
Following the “Squares – Avg 2 Satellite datasets” and “5-Yr Running Avgs (Trend line intercept = 0 at 1979 for all)” indicators I get the following :
Year … (UAH+RSS)/2 … Averages … Trend_at_1979=0
1979 … -0.135
1980 … -0.017
1981 … -0.174 … -0.048 … 0.168
1982 … -0.151 … -0.107 … 0.109
1983 … 0.238 … -0.195 … 0.021
1984 … -0.433 … -0.221 … -0.005
1985 … -0.455 … -0.139 … 0.077
1986 … -0.305 … -0.195 … 0.022
1987 … 0.261 … -0.196 … 0.020
1988 … -0.041 … -0.115 … 0.102
1989 … -0.439 … -0.048 … 0.169
1990 … -0.049 … -0.118 … 0.098
1991 … 0.029 … -0.133 … 0.083
1992 … -0.093 … -0.065 … 0.151
1993 … -0.116 … -0.039 … 0.177
1994 … -0.097 … -0.071 … 0.146
1995 … 0.079 … -0.038 … 0.178
1996 … -0.126 … 0.113 … 0.329
1997 … 0.069 … 0.086 … 0.303
1998 … 0.639 … 0.032 … 0.248
1999 … -0.229 … 0.057 … 0.273
2000 … -0.195 … 0.076 … 0.292
2001 … 0.002 … -0.011 … 0.205
2002 … 0.163 … 0.052 … 0.269
2003 … 0.205 … 0.134 … 0.350
2004 … 0.086 … 0.152 … 0.368
2005 … 0.214 … 0.132 … 0.348
2006 … 0.091 … 0.040 … 0.256
2007 … 0.062 … 0.046 … 0.262
2008 … -0.251 … 0.075 … 0.292
2009 … 0.116 … 0.037 … 0.253
2010 … 0.360 … 0.008 … 0.224
2011 … -0.101 … 0.084 … 0.300
2012 … -0.085 … 0.090 … 0.306
2013 … 0.127 … 0.097 … 0.313
2014 … 0.147 … 0.233 … 0.450
2015 … 0.397 … 0.310 … 0.526
2016 … 0.580 … 0.310 … 0.527
2017 … 0.298 … 0.364 … 0.580
2018 … 0.130 … 0.384 … 0.600
2019 … 0.416
2020 … 0.494
By my reckoning (2018, 0.6) is in the “gap” between the “bottom 2, sky-blue and green, lines” and the “sky-blue + orange + sky-blue” ones at the middle-right of the graph, but YMMV …
Yeah, Mark, my Mk1 eyeballs tell the same story. We have temperatures driven high by a Super El Nino still below the “cloud” of model predictions. Additionally, temperatures over the last few years have been falling, further invalidating the UN IPCC CliSciFi models.
Thanks alot for helping out Mark.
By the way, what does the “Averages’ column signify?
Mark, it just struck me that your result is functionally the same as Dr. Roy Spencer’s depiction of the observed SST vs 68 CMIP6 model runs, above; the actuals plot along the bottom of the graph along with the “coolest” model result!
“By the way, what does the “Averages’ column signify?”
They’re the “(5-year, centered, rolling) Averages” of the previous column [ (UAH + RSS) / 2, annual anomalies ], to provide as close a match as possible to the “5-Yr Running Avgs …” label on Dr. Spencer’s graph.
The last, “Trend line at 1979 = 0”, column should just be that “Averages” column with the “appropriate” fixed vertical shift.
– – – – – –
PS : I forgot that the “tmtglhmam_6.0.txt” UAH file contains monthly [ “GLOBAL”, “NH”, “SH” and ] “TRPC” area averages to three decimal places.
The “uahncdc_*” files have more “zones” (e.g. “USA48”), but the data is only provided to 2dp.
“There is a computer disease that anybody who works with computers knows about. It’s a very serious disease and it interferes completely with the work. The trouble with computers is that you ‘play’ with them.” — Richard P. Feynman.
Looks like I’ll be spending this Sunday “not working” on my computer …
Thanks, again, for your “not working” so hard.
Can we stop calling ERSST ‘observations’?
ERSST is a reconstruction involving various mathematical processes, interpolations, averages etc, largely based on the actual observations supplied by ICOADS (https://icoads.noaa.gov/). One doesn’t need to look very deep into that to see how variable (both in technoolgy and density) that data actually is, and indeed how actually sparse it is in many areas.
Because it is a reconstruction it necessarily has some assumptions baked in.
These two recent WUWT articles document two back-to-back “pauses” in global warming (a step shift between them occurring around CY2014), neither of which were predicted by any of the 68 simulations from the 13 computers models used in CMIP6 (see Figure 1 above):
https://wattsupwiththat.com/2014/07/03/rss-shows-no-global-warming-for-17-years-10-months/
and
https://wattsupwiththat.com/2021/05/03/the-new-pause-lengthens-by-two-months-to-6-years/
And one wonders why CMIP6 did not conduct model comparisons against more accurate and/or difference data sets, such as the ARGO float network for sea-surface—and near sea-surface—temperature trending, and NOAA’s U.S. Climate Reference Network for lower atmospheric air temperature-over-land trending.
I don’t get it. Schmidt & Company’s model shows a +0.75C anomaly from the average of the 1951 to 1980 period baseline to the average of the 2017 to 2021 period. My understanding is that the total change from the end of the LIA is only about +0.8C. WTF?
Dr. Spencer points out that the official estimated change in ocean heat content (OHC) is orders of magnitude less than our ability to measure. Do you want to bet your future based on scary stories?
I am not convinced that long-term climate simulation is possible (as opposed to simulating the weather). The NCAR-based Community Earth System Model (CESM) ran the same simulation 40 times for the period 1920 and 2100, changing the starting temperature by less than one trillionth of a degree each time. The results show staggering variability, to the point where you cannot even tell if winters in North America will get hotter or colder.
40 Earths: NCAR’s Large Ensemble reveals staggering climate variability | NCAR & UCAR News
The problem is the famous Butterfly Effect which seems to me to make the models fairly useless.