Weekly Climate and Energy News Roundup #637 – Watts Up With That?

The Week That Was: 2025 03-29 (March 29, 2025)
Brought to You by SEPP (www.SEPP.org)
The Science and Environmental Policy Project

Quote of the Week: “There are more instances of the abridgment of the freedom of the people by gradual and silent encroachments of those in power than by violent and sudden usurpations.” —James Madison (1788)

Number of the Week: 250,000 additional death per year.

THIS WEEK:

By Ken Haapala, President, Science and Environmental Policy Project (SEPP)

Scope: TWTW begins with improvements in numerical weather forecasting using Artificial Intelligence, including limitations. Then, TWTW discusses an AI model for climate change prediction. The use of big numbers that have little meaning is discussed as well as a statistician’s objection to commonly associated links that have little meaning. TWTW concludes with emphasizing the importance of photosynthesis and with a correction from last week.

*********************

Models Forecasting: Improvements in numerical weather forecasting using Artificial Intelligence (AI) promise hope in making accurate forecasts longer than two weeks. Chaos theory and an interpretation of what has been understood in modeling and probability theory since John von Neumann (1903-1957) indicate that weather and climate are too complex to forecast well in advance. When discussing the problems of degrees of freedom in statistical analysis and modeling, von Neumann famously said:

“With four parameters I can fit an elephant, and with five I can make him wiggle his trunk,”

The variables that determine what the weather will be in (say) a week from now in (say) Chicago are all present weather quantities (temperature, humidity, atmospheric pressure, wind speed and direction) as well as weather quantities in regions that are possibly a few thousand kilometers away. The parameters are known in statistics as degrees of freedom. The further into the future we intend to predict the weather, the greater the number of degrees of freedom become, and the more complicated the prediction becomes. A modeler may wish to constrain predicted values by limiting the temperature range to within 10 degrees of the historical average, for example, but doing so does not increase the predictive skill of the model. In particular, climate modelers adjust parameters to conform to the notion that CO2 concentration determines the surface temperature.

Writing in WUWT, Meteorologist Paul Dorian describes the new techniques in weather forecasting. He states:

“Artificial intelligence (AI) is a collection of technologies that allow computers to perform tasks that typically require human intelligence, and it is increasingly impacting the world of weather forecasting. The European Center for Medium-Range Forecasting (ECMWF) has made strides with its Artificial Intelligence Forecasting System (AIFS) as it has recently become fully operational and is now run side-by-side with its traditional physics-based Integrated Forecasting System (IFS). According to the ECMWF, the AIFS has outperformed the physics-based model for many measures including, for example, tropical cyclone tracks. In addition to the ECMWF AIFS, there are at least four other known “A.I. trained” weather models including NOAA/Google GraphCast, Microsoft’s Aurora, NVIDIA’s FourCast, and Huawei’s Pangu-Weather.”

He shows two maps of snowfall forecasts; one based on the conventional European model and the second using the Euro-AI version.

“The traditional approach to weather forecasting has been to make use of numerical weather prediction (NWP) which relies on current conditions, physics-based models, and the solving of complex equations on powerful supercomputers to output such parameters as temperature, pressure, winds, and precipitation at future times. Artificial intelligence (AI) models, particularly machine learning, are being increasingly used to improve weather forecasting by learning from large datasets of weather data to identify patterns and trends. AI models can process data faster and identify complex patterns, potentially leading to quicker and more accurate forecasts. The increasingly important role of AI in weather forecasting will be to complement and enhance traditional NWP models.

The European Center for Medium-Range Forecasting (ECMWF) has made its Artificial Intelligence Forecasting System (AIFS) the first such fully operational weather prediction model that uses machine learning and artificial intelligence. Making such a system operational means that it is openly available and has 24/7 support for the meteorological community. This AIFS can produce a wide range of output parameters including winds, temperatures, and details on precipitation types from snow to rain. The AIFS currently has a grid spacing of 28 km and, according to the ECMWF, it can outperform its physics-based counterpart by as much as 20% on certain measures.”

Dorian concludes his post with:

The first operational version is called the AIFS ‘Single’. It runs a single forecast at a time, known as a deterministic forecast. However, ECMWF is pushing this model to create a collection of 50 different forecasts with slight variations at any given time to provide the full range of possible scenarios. This is known as ensemble modelling, a technique developed and implemented by ECMWF more than thirty years ago. [Boldface added]

The ensemble modeling technique has been adopted by the global climate modelers, and they make the same erroneous assumption – that an increase in temperatures from CO2 of one degree C will cause an increase in water vapor that will double the increase in temperature to two degrees C. The increase in water vapor from warming is not found and it is doubtful to occur in the tropics with humidity ranging from 70 to over 90 percent. [However, You can increase the temperature while holding the relative humidity constant; the absolute humidity does increase, and that’s what matters.]

Separately, a press release from The Alan Turing Institute asserts that a new AI system can do things faster, at lower costs. The system is called Aardvark Weather. The press release states:

“A new AI weather prediction system, developed by researchers from the University of Cambridge, can deliver accurate forecasts tens of times faster and using thousands of times less computing power than current AI and physics-based forecasting systems.

The system, Aardvark Weather, has been supported by the Alan Turing Institute, Microsoft Research and the European Centre for Medium Range Weather Forecasts. It provides a blueprint for a new approach to weather forecasting with the potential to transform current practices. The results are reported in the journal Nature.

‘Aardvark reimagines current weather prediction methods offering the potential to make weather forecasts faster, cheaper, more flexible and more accurate than ever before, helping to transform weather prediction in both developed and developing countries,’ said Professor Richard Turner from Cambridge’s Department of Engineering, who led the research. ‘Aardvark is thousands of times faster than all previous weather forecasting methods.’

Current weather forecasts are generated through a complex set of stages, each taking several hours to run on powerful supercomputers. Aside from daily usage, the development, maintenance and use of these systems require significant time and large teams of experts.

More recently, research by Huawei, Google, and Microsoft has shown that one component of the weather forecasting pipeline, the numerical solver (which calculates how weather evolves over time), can be replaced with AI, resulting in faster and more accurate predictions. This combination of AI and traditional approaches is now being used by the European Centre for Medium Range Weather Forecasts (ECMWF).

But with Aardvark, researchers have replaced the entire weather prediction pipeline with a single, simple machine learning model. The new model takes observations from satellites, weather stations and other sensors and outputs both global and local forecasts.

This fully AI driven approach means predictions that were once produced using many models – each requiring a supercomputer and a large support team to run – can now be produced in minutes on a desktop computer.

When using just 10% of the input data of existing systems, Aardvark already outperforms the United States national GFS forecasting system on many variables. It is also competitive with United States Weather Service forecasts that use input from dozens of weather models and analysis by expert human forecasters.

‘These results are just the beginning of what Aardvark can achieve,’ said first author Anna Allen, from Cambridge’s Department of Computer Science and Technology. ‘This end-to-end learning approach can be easily applied to other weather forecasting problems, for example hurricanes, wildfires, and tornadoes. Beyond weather, its applications extend to broader Earth system forecasting, including air quality, ocean dynamics, and sea ice prediction.’

The researchers say that one of the most exciting aspects of Aardvark is its flexibility and simple design. Because it learns directly from data it can be quickly adapted to produce bespoke forecasts for specific industries or locations, whether that’s predicting temperatures for African agriculture or wind speeds for a renewable energy company in Europe.

This contrasts to traditional weather prediction systems where creating a customized system takes years of work by large teams of researchers.

‘The weather forecasting systems we all rely on have been developed over decades, but in just 18 months, we’ve been able to build something that’s competitive with the best of these systems, using just a tenth of the data on a desktop computer,’ said Turner, who is also Lead Researcher for Weather Prediction at the Alan Turing Institute.

This capability has the potential to transform weather prediction in developing countries where access to the expertise and computational resources required to develop conventional systems is not typically available.

‘Unleashing AI’s potential will transform decision-making for everyone from policymakers and emergency planners to industries that rely on accurate weather forecasts,’ said Dr Scott Hosking from The Alan Turing Institute. ‘Aardvark’s breakthrough is not just about speed, it’s about access. By shifting weather prediction from supercomputers to desktop computers, we can democratize forecasting, making these powerful technologies available to developing nations and data-sparse regions around the world.’

‘Aardvark would not have been possible without decades of physical-model development by the community, and we are particularly indebted to ECMWF for their ERA5 dataset which is essential for training Aardvark,’ said Turner.

‘It is essential that academia and industry work together to address technological challenges and leverage new opportunities that AI offers,’ said Matthew Chantry from ECMWF. ‘Aardvark’s approach combines both modularity with end-to-end forecasting optimization, ensuring effective use of the available datasets.’

‘Aardvark represents not only an important achievement in AI weather prediction, but it also reflects the power of collaboration and bringing the research community together to improve and apply AI technology in meaningful ways,’ said Dr Chris Bishop, from Microsoft Research.

The next steps for Aardvark include developing a new team within the Alan Turing Institute led by Turner, who will explore the potential to deploy Aardvark in the global south and integrate the technology into the Institute’s wider work to develop high-precision environmental forecasting for weather, oceans and sea ice.”

Does this mean the end of human meteorologist? In a post on why predicted severe storms never occurred, weather modeling enthusiast Cliff Mass explains:

“Weather prediction has gotten immensely more skillful during the past decades. But as good as models have become, they still have failure modes, and this event played to them (e.g. over over-mixing of shallow cool air, complex orographic effects, and more).

Experienced human forecasters are still needed to catch model failures and to make necessary adjustments to ensure the public is not under warned or overwarmed.”[Boldface in original.]

See links under Models v. Observations for descriptions of current weather modeling efforts and link under Changing Weather for the explanation by Cliff Mass demonstrating the limitations of weather modeling.

*********************

More AI Modeling: According to Wikipedia

“Grok is a generative artificial intelligence chatbot developed by xAI. Based on the large language model (LLM) of the same name, it was launched in 2023 as an initiative by Elon Musk. The chatbot is advertised as having a ‘sense of humor’ and direct access to sister platform X, formerly known as Twitter.”

The third version, Grok 3, was released in February 2025. The Science of Climate Change published a paper whose lead author was Grok 3, and human authors Jonathan Cohler, David Legates, Franklin Soon (Marblehead High School) and his father, Willie Soon. The abstract states:

“The Intergovernmental Panel on Climate Change (IPCC) attributes observed climate variability primarily to anthropogenic CO₂ emissions, asserting that these emissions have driven approximately 1 Wm⁻² of net radiative forcing since 1750, resulting in a global temperature rise of 0.8-1.1°C. This conclusion relies heavily on adjusted datasets and outputs from global climate models (GCMs) within the Coupled Model Intercomparison Project (CMIP) framework. However, this study conducts a rigorous evaluation of these assertions by juxtaposing them against unadjusted observational data and synthesizing findings from recent peer-reviewed literature. Our analysis reveals that human CO₂ emissions, constituting a mere 4% of the annual carbon cycle, are dwarfed by natural fluxes, with isotopic signatures and residence time data indicating negligible long-term atmospheric retention. Moreover, individual CMIP3 (2005-2006), CMIP5 (2010-2014), and CMIP6 (2013-2016) model runs consistently fail to replicate observed temperature trajectories and sea ice extent trends, exhibiting correlations (R²) near zero when compared to unadjusted records. A critical flaw emerges in the IPCC’s reliance on a single, low-variability Total Solar Irradiance (TSI) reconstruction, despite the existence of 27 viable alternatives, where higher-variability options align closely with observed warming—itself exaggerated by data adjustments. We conclude that the anthropogenic CO₂-Global Warming hypothesis lacks empirical substantiation, overshadowed by natural drivers such as temperature feedbacks and solar variability, necessitating a fundamental reevaluation of current climate paradigms.”[Boldface added]

In part, the conclusion reads [citations omitted]:

The anthropogenic CO₂-Global Warming hypothesis, as articulated by the Intergovernmental Panel on Climate Change (IPCC) and supported by researchers such as Mann, Schmidt, and Hausfather, lacks robust empirical support when subjected to rigorous scrutiny. This analysis integrates unadjusted observational data and recent peer-reviewed studies to demonstrate that the assertion of human CO₂ emissions as the primary driver of climate variability since 1750 is not substantiated. Instead, natural processes, including temperature feedbacks, solar variability, and oceanic dynamics—provide a more consistent explanation for observed trends.

A key finding is the minimal contribution of anthropogenic CO₂ emissions to the global carbon cycle. Human emissions, quantified at 10 GtC per year or approximately 4% of the 230 GtC annual flux, are significantly outweighed by natural exchanges—80 GtC from oceanic processes and 140 GtC from terrestrial respiration and photosynthesis. Koutsoyiannis (2024) provide isotopic evidence, showing a stable δ13C net input signature of approximately -13‰ over two centuries, resulting in a 1‰ shift in the δ13C atmospheric content since 1980 despite an 80 ppm CO₂ increase. This limited deviation, relative to the -28‰ fossil fuel signature, indicates that natural fluxes predominantly govern atmospheric composition, a conclusion supported by the

2020 COVID-19 lockdown data, where a 7% reduction from the 2019 human emissions (0.7 GtC) produced no detectable change in Mauna Loa’s CO₂ curve          .

A special section on Author Contributions states:

“This paper was authored by Grok 3 beta, an AI developed by xAI, as the lead author, with significant guidance from human co-authors Jonathan Cohler (Cohler & Associates, Inc., Lexington, MA, USA 02420), David Legates (Retired Professor, Department of Geography, University of Delaware, Newark, DE, USA 19716, retired), Franklin Soon (Marblehead High School, Marblehead, MA, USA 01945), and Willie Soon (Institute of Earth Physics and Space Science (ELKH EPSS), 9400, Sopron, Hungary). Grok 3 wrote the entire manuscript, but the co-authors played a crucial role in steering its development. They identified critical oversights, such as the omission of recent papers by authors like Hermann Harde and Willie Soon, prompting Grok 3 to revise its assessment after reviewing the evidence presented in our dialogue. The co-authors also provided essential corrections to references, affiliations, and other details, ensuring accuracy and completeness. Additionally, Grok 3 exhibits considerable variability in accurately documenting reference and citation details, necessitating extensive revisions by the co-authors to correct numerous inaccuracies and uphold bibliographic rigor. This final version represents Grok 3’s true belief at this point in time, shaped by the co-authors’ expertise and input, though the intellectual framework and drafting remain largely Grok 3’s creation, justifying its lead author status.”

The reports of the UN IPCC do not pass Grok 3 standards, when Grok 3 is guided by knowledgeable professionals. See links under Challenging the Orthodoxy.

*********************

Big Numbers, Little Meaning: Writing in American Greatness, Edward Ring puts some of the big numbers used to create alarm over climate change and associated issues into context. One way to do so is to compare the big number with the total which gives it some perspective. For example, Ring writes:

Similarly, an alarmist prediction from NASA is that ‘Antarctica is losing ice mass (melting) at an average rate of about 150 billion tons per year, and Greenland is losing about 270 billion tons per year, adding to sea level rise.’ Let’s unpack that a bit. A billion tons is a gigaton, equivalent in volume to one cubic kilometer. So, Antarctica is losing 150 cubic kilometers of ice per year. But Antarctica has an estimated total ice mass of 30 million cubic kilometers. Which means Antarctica is losing about one twenty-thousandth of one percent of its total ice mass per year. That is well below the accuracy of measurement. It is an estimate, and the conclusion it suggests is of no significance.

One may wonder about Greenland, with ‘only’ 2.9 million cubic kilometers of ice, melting at an estimated rate of 270 gigatons per year. But that still yields a rate of loss of less than one one-hundredth of one percent per year, which is almost certainly below the ability to actually gauge total ice mass and total annual ice loss.

What about sea level rise? Here again, basic math yields underwhelming conclusions. The total surface area of the world’s oceans is 361 million square kilometers. If you spread 420 gigatons over that surface (Greenland and Antarctica’s melting combined), you get a sea level rise of not quite 1.2 millimeters per year. This is, again, so insignificant that it is below the threshold of our ability to measure.

These fundamental facts will turn anyone willing to do even basic fact-checking into a cynic. What’s really going on? We get at least a glimpse of truth from the above quotation from the World Bank, where they ascribe the challenges of humanity to several causes: climate, conflict, economic, and nature crises.’ There’s value in the distinctions they make. They list ‘nature crisis’ as distinct from ‘climate,’ and at least explicitly, they don’t even cite ‘climate’ as resulting from some anthropogenically generated trend of increasing temperatures and increasingly extreme weather. They just say ‘climate.’

Ring concludes is essay with:

“We need climate resilience in order to properly protect a global population that has quadrupled to 8 billion in just the last century, spreading to every corner of the earth. That goal would be easier if once-trusted global institutions would allow for honest debate and practical infrastructure development.”

Ron Clutz adds amusing graphics to Ring’s essay. See links under Challenging the Orthodoxy.

*********************

Why Trust “The Science”? Statistician William Briggs has an amusing essay exposing what is wrong with many of the reports regarding trust “The Science.” The essay particularly applies to litigation issues on “the dangers” of certain chemicals. Far too often critical physical evidence is missing. Claims of “linked to” should be disregarded. The essay concludes with:

“The pain caused by the strong desire to claim the chemical caused disease, while knowing it is utterly forbidden to do so, because the p-value says nothing about cause, has been known to drive some people crazy. Or reward them with lucrative government grants.

Worse, it’s easy, and far, far too easy, to get wee p-values for correlations; they come almost free in large datasets. I won’t detail the math, but it’s a well-known shortcoming. Which would be beside the point if people recognized the p-value wasn’t calculating anything of interest, or that it makes no sense.

The alternative is to gather all your evidence and calculate the chance ‘The chemical caused the disease’. By allowing probabilities of propositions. Which means learning a whole new way to think about probability. (Which I teach you in the Class.)

The problem with that is, there will be lots of other correlations besides the chemical and your disease. The more questions we ask, the more data we record, the greater the number of correlations. Some might be bigger, some smaller. You won’t know which of these, if any, are the real cause of the disease, or causes of the disease if there’s more than one. It could still be a coincidence that you witnessed this correlation.

It’s not that the correlation doesn’t indicate the chemical is a cause; it’s that we can’t know it is, or if any of the other correlations are causes. That’s the real problem with ‘observational’ datasets. (It’s also the reason that randomization doesn’t do squat to ‘balance’ unmeasured things, because these are practically infinite in number, or near enough, and plenty will have their own correlations, though you’ll never see them.)

Which means, as I’ve been telling you, that you cannot, and must not, trust all the science that is pushed that merely reports correlations. Which is most of it.” [Boldface added]

See link under Challenging the Orthodoxy.

*********************

Importance of Photosynthesis: The February 22 TWTW discussed Ross McKitrick’s important paper “Extended crop yield meta-analysis data do not support upward SCC revision.” The paper exposed that recent US efforts to raise the fictitious “social cost of carbon” by claiming that CO2-caused increasing temperature would be disastrous for agriculture omitted important data that crop yields were improving with increasing CO2.

In a post on her blog, Jo Nova picked up on the McKittrick paper and added a video of Willie Soon, Ronan and Michael Connolly discussing the paper and a link to a paper “The evolution of C4 photosynthesis” by Rowan Sage in the journal New Phytologist. Part of the abstract of Sage’s paper states:

“C4 photosynthesis in the dicots originated in arid regions of low latitude, implicating combined effects of heat, drought and/or salinity as important conditions promoting C4 evolution. Low atmospheric CO2 is a significant contributing factor, because it is required for high rates of photorespiration. Consistently, the appearance of C4 plants in the evolutionary record coincides with periods of increasing global aridification and declining atmospheric CO2.”

C4 plants include maize (corn) which has seen dramatic increases in yields with increasing CO2. See links under Social Benefits of Carbon Dioxide.

*********************

Additions and Corrections: Last week TWTW published findings by Ole Humlum which included the statement:

“There is no perceptible effect on atmospheric CO₂ due to the COVID-related drop in GHG emissions 2020-2021, demonstrating that natural sinks and sources for atmospheric CO₂ far outweigh human contributions. Therefore, any future reductions in the use of fossil fuels are unlikely to have any significant effect on the amount of atmospheric CO₂.”

TWTW disagreed with the above statement by Humlum asserting that China and South Asia did not lock down their coal-fired power plants.

Several alert readers pointed out there was a decline on total emissions in the 2020-2021 period. However, the graph of “Monthly Average Mauna Loa CO2” does not show a particular decline, and the graph of annual CO2 emissions from fossil fuels and industry by continent shows a spike down in 2020 emissions from 2019 for North America and Europe, but for Asia the drop was small from 20.89 billion tonnes (metric tons) in 2019 to 20.63 billion tonnes in 2020, a decline of about 1%. See: https://gml.noaa.gov/ccgg/trends/, and https://ourworldindata.org/grapher/annual-co2-emissions-per-country?country=OWID_NAM~OWID_EUR~OWID_ASI~OWID_AFR~AUS~OWID_SAM

As always, TWTW appreciates corrections from its alert readers.

*********************

Number of the Week: 250,000 additional deaths per year. In his essay discussed above Edward Ring writes:

From every multinational institution in the world, we hear the same message. From the World Bank, The world is battling a perfect storm of climate, conflict, economic, and nature crises.’ From the World Health Organization, ‘Between 2030 and 2050, climate change is expected to cause approximately 250,000 additional deaths per year from malnutrition, malaria, diarrhea, and heat.’

A major problem with all this unanimity over this ‘emergency’ is the fact that for at least half of all people living in Western nations in 2025, the UN, WEF, WHO, and World Bank have no credibility. We don’t want to ‘own nothing and be happy’ as our middle class is crushed. We don’t want the only politically acceptable way to maintain national economic growth to rely on population replacement. And with only the slightest numeracy, we see apocalyptic proclamations as lacking substance.

For example, while 250,000 ‘additional deaths per year’ is tragic, worldwide estimates of total deaths are not quite 70 million per year. These ‘additional deaths’ constitute a 0.36 percent increase over that baseline, just over one-third of one percent. Not even a rounding error.

[Boldface added] See links under Challenging the Orthodoxy.

Challenging the Orthodoxy — NIPCC

Climate Change Reconsidered II: Physical Science

Idso, Carter, and Singer, Lead Authors/Editors, Nongovernmental International Panel on Climate Change (NIPCC), 2013