07.7.2021

Patrick J. Michaels: Analysis of Wolf Administration’s Flawed Use of Climate and CO2-Emission Models

By Dr. Patrick J. Michaels, Senior Fellow, CO2 Coalition and Competitive Enterprise Institute

Governor Wolf’s Executive Order 2019-7 on addressing climate change in Pennsylvania is largely based upon the 2018 Pennsylvania Climate Action Plan. Therefore, my analysis is about that Plan.  It suffers from serious problems that call into question the basis for the Governor’s Order.

The 2018 Pennsylvania Climate Action Plan [PCAP] is a document that theoretically is based on the best available science and is being used as the basis for sweeping and extensive regulation of the emissions of greenhouse gases—mainly carbon dioxide and methane—from the Commonwealth of Pennsylvania.

My comments on it are largely along three lines;

  1.  The validity of the climate models that serve as its basis, and
  2.  The validity of the emission assumptions that underpin the plan, and
  3.  The amount of warming and sea level rise mitigated if Pennsylvania emissions from electricity generation were completely curtailed in the beginning of this century.

It will be demonstrated that, in general, these models simply do not work when simulating climate changes in recent decades, and that the emission assumption, known as “RCP 8.5” is now recognized to be a gross exaggeration of changes in the atmosphere of the future. In combination, use of the only working climate model (see below) and a more likely emissions scenario negates the credibility of the Pennsylvania Climate Action Plan report. (RCP means “representative concentration pathway” and the 8.5 is the change of downwelling energy, in this case 8.5 watts per square meter, that pathway would produce.)

I’ll begin with the observed behavior of the atmosphere versus model predictions in recent decades.  The underlying data and the enclosed illustration were readily available at the time PCAP was published.  The fact that they were ignored casts substantial doubt on the scientific credibility of the PCAP document.

Figure 12 shows the predicted and observed average tropospheric temperature over the tropics.  Predictions are the US Department of Energy’s fifth Coupled Model Intercomparison Project (CMIP-5) model suite data that was readily available at the KNMI Climate Explorer site.24

A careful look at Figure 12 reveals that only one of the 102 model runs correctly simulates what has been observed.  This is the Russian climate model INM-CM4, which also has the least prospective warming of all of them, with an equilibrium climate sensitivity (ECS) of 2.05⁰C, compared to the CMIP-5 average of 3.2⁰C. 

If PCAP followed best scientific practice, it would use this model along with a more realistic projection of future emissions than it used (see below).  This is similar to what operational meteorologists do every day.  They generally dont take all the available daily forecast models and average them up, as some perform better or worse depending upon the daily weather situation.  Instead, they rely on the one(s) that perform the best.

Had PCAP followed this best practice, its projected 2000-2050 statewide average warming would drop from 5.4⁰F to 3.3⁰, a reduction of 40%.

Christy and McNider (2017) 25, in a paper readily available to the PCAP writing team, further demonstrated that the models are predicting several times more warming at altitude in the tropics than is being observed.

The implications of this error are manifold, as a substantial amount of moisture that falls as precipitation over Pennsylvania originates in the tropics.  The amount that enters the air is determined by the temperature contrast between the surface and the upper reaches of the lower atmosphere.  Getting this wrong (with too little contrast) means that precipitation forecasts for Pennsylvania are systematically underestimated in the PCAP.

In summary, on this point, it is clear that had PCAP used the working climate model and a realistic emissions scenario (see below) that it would have forecast less than half the warming that it did, likely rendering the issue much less frightening, with adaptation (which is already occurring) the most likely response, rather than a wholesale re-ordering of the lives of the Commonwealth’s citizens.

Governor Wolf’s Executive Order 2019-7 on addressing climate change in Pennsylvania is largely based upon the 2018 Pennsylvania Climate Action Plan. Therefore, my analysis is about that Plan.  It suffers from serious problems that may call into question the basis for the Governor’s Order. The 2018 Pennsylvania Climate Action Plan [PCAP] is a document that theoretically is based on the best available science and is being used as the basis for sweeping and extensive regulation of the emissions of greenhouse gases—mainly carbon dioxide and methane—from the Commonwealth of Pennsylvania.

My comments on it are largely along three lines;

  1.  The validity of the climate models that serve as its basis, and
  2.  The validity of the emission assumptions that underpin the plan, and
  3.  The amount of warming and sea level rise mitigated if Pennsylvania emissions from electricity generation were completely curtailed in the beginning of this century.

It will be demonstrated that, in general, these models simply do not work when simulating climate changes in recent decades, and that the emission assumption, known as “RCP 8.5” is now recognized to be a gross exaggeration of changes in the atmosphere of the future. In combination, use of the only working climate model (see below) and a more likely emissions scenario negates the credibility of the Pennsylvania Climate Action Plan report. (RCP means “representative concentration pathway” and the 8.5 is the change of downwelling energy, in this case 8.5 watts per square meter, that pathway would produce.)

I’ll begin with the observed behavior of the atmosphere versus model predictions in recent decades.  The underlying data and the enclosed illustration were readily available at the time PCAP was published.  The fact that they were ignored casts substantial doubt on the scientific credibility of the PCAP document.

Figure 12 shows the predicted and observed average tropospheric temperature over the tropics.  Predictions are the US Department of Energy’s fifth Coupled Model Intercomparison Project (CMIP-5) model suite data that was readily available at the KNMI Climate Explorer site.24

A careful look at Figure 12 reveals that only one of the 102 model runs correctly simulates what has been observed.  This is the Russian climate model INM-CM4, which also has the least prospective warming of all of them, with an equilibrium climate sensitivity (ECS) of 2.05⁰C, compared to the CMIP-5 average of 3.4⁰C. 

If PCAP followed best scientific practice, it would use this model along with a more realistic projection of future emissions than it used (see below).  This is similar to what operational meteorologists do every day.  They generally dont take all the available daily forecast models and average them up, as some perform better or worse depending upon the daily weather situation.  Instead, they rely on the one(s) that perform the best.

Had PCAP followed this best practice, its projected 2000-2050 statewide average warming would drop from 5.4⁰F to 3.3⁰, a reduction of 40%.

Christy and McNider (2017) 25, in a paper readily available to the PCAP writing team, further demonstrated that the models are predicting several times more warming at altitude in the tropics than is being observed.

The implications of this error are manifold, as a substantial amount of moisture that falls as precipitation over Pennsylvania originates in the tropics.  The amount that enters the air is determined by the temperature contrast between the surface and the upper reaches of the lower atmosphere.  Getting this wrong (with too little contrast) means that precipitation forecasts for Pennsylvania are systematically underestimated in the PCAP.

In summary, on this point, it is clear that had PCAP used the working climate model and a realistic emissions scenario (see below) that it would have forecast less than half the warming that it did, likely rendering the issue much less frightening, with adaptation (which is already occurring) the most likely response, rather than a wholesale re-ordering of the lives of the Commonwealth’s citizens.


Figure 12. Solid red line—average of all the CMIP-5 climate models available at the time PCAP was written; Thin colored lines—individual CMIP-5 models; solid figures—weather balloon, satellite, and reanalysis data for the tropical troposphere.  Source:  Christy, J.R.:2017, [in “State of the Climate 2016”], Bull. Amer. Meteor. Soc. 98, (8), S16-S17. DOI:1-.1175/2017BAMSStateoftheClimate.I.

As important as it is to follow best scientific practice in modeling and forecasting tomorrow’s weather, it is even more vitally important to use the best assumed future emissions pathway when basing important policy decisions on that model.

-On page 5, PCAP states:

“This report adopts the Representative Concentration Pathway 8.5 (RCP 8.5). This pathway is the one that the world is currently on and is one of two emissions pathways adopted by a large number of climate modeling groups.”

RCP 8.5 is the most extreme emissions scenario employed in the most recent (2013) comprehensive report of the United Nations’ Intergovernmental Panel on Climate Change (IPCC). Since the publication of PCAP in 2018, the use of RCP 8.5 has been roundly criticized.

In 2020, Zeke Hausfather and Glen Peters published an article26 in the prestigious journal Nature which began with the blunt statement, “Stop using the worst-case scenario for climate warming as the most likely outcome — more realistic baselines make for better policy.”

According to Hausfather and Peters, RCP 8.5 “paints a dystopian future that is fossil-fuel intensive and excludes any climate mitigation policies, leading to nearly 5°C of warming by the end of the century.”

Sole reliance on RCP 8.5 invites strong criticism because of its unreality.  Hausfather and Peters note that “Emission pathways to get to RCP 8.5 generally require an unprecedented fivefold increase in coal use by the end of the century, an amount larger than some estimates of recoverable coal reserves.” 

They conclude that “We must all — from physical scientists and climate-impact modelers to communicators and policymakers — stop presenting the worst-case scenario as the most likely one.” This includes the authors of the PCAP.

A sidebar, shown below, in the Hausfather and Peters article intercompares the various emissions scenarios, finding that the latest version of RCP 8.5 (called here “SSP5-8.5) in the upcoming (2022) IPCC Sixth Assessment report to be Highly unlikely and often wrongly used as ‘business as usual’ [emphasis in original].

As important as it is to follow best scientific practice in modeling and forecasting tomorrow’s weather, it is even more vitally important to use the best assumed future emissions pathway when basing important policy decisions on that model.

-On page 5, PCAP states:

“This report adopts the Representative Concentration Pathway 8.5 (RCP 8.5). This pathway is the one that the world is currently on and is one of two emissions pathways adopted by a large number of climate modeling groups.”

RCP 8.5 is the most extreme emissions scenario employed in the most recent (2013) comprehensive report of the United Nations’ Intergovernmental Panel on Climate Change (IPCC). Since the publication of PCAP in 2018, the use of RCP 8.5 has been roundly criticized.

In 2020, Zeke Hausfather and Glen Peters published an article26 in the prestigious journal Nature which began with the blunt statement, “Stop using the worst-case scenario for climate warming as the most likely outcome — more realistic baselines make for better policy.”

According to Hausfather and Peters, RCP 8.5 “paints a dystopian future that is fossil-fuel intensive and excludes any climate mitigation policies, leading to nearly 5°C of warming by the end of the century.”

Sole reliance on RCP 8.5 invites strong criticism because of its unreality.  Hausfather and Peters note that “Emission pathways to get to RCP 8.5 generally require an unprecedented fivefold increase in coal use by the end of the century, an amount larger than some estimates of recoverable coal reserves.” 

They conclude that “We must all — from physical scientists and climate-impact modelers to communicators and policymakers — stop presenting the worst-case scenario as the most likely one.” This includes the authors of the PCAP.

A sidebar, shown below, in the Hausfather and Peters article intercompares the various emissions scenarios, finding that the latest version of RCP 8.5 (called here “SSP5-8.5) in the upcoming (2022) IPCC Sixth Assessment report to be Highly unlikely and often wrongly used as ‘business as usual’ [emphasis in original].

Figure 13.  Sidebar showing various emissions pathways that accompanied the 2020  Nature article by Hausfather and Peters.  Source:  Hausfather, Z., and M. Peters, Nature, Vol 577, 618-620.

This strong language applies to the PCAP.  As Hausfather told the BBC concerning his work, RCP 8.5 is “exceedingly unlikely.”

As shown in Figure 13, the likely scenario, given policies that already exist, is an RCP value of between 4 and 6, instead of 8.5. This would reduce prospective warming by between 30 and 50%, as warming rises linearly with the RCP value.  Using the most conservative version of the correct RCP value reduces warming another 1.6⁰F, leaving a remainder from 2000-2050 of 1.7⁰F, a far cry from the original PCAP value of 5.4⁰.

As a thought experiment, suppose that Pennsylvania ceased any carbon dioxide emissions from electricity generation a decade ago.  The EPA’s Model for the Assessment of Greenhouse-Gas Induced Climate Change27 (the acronym is MAGICC) is the standard tool used to assess the climate impacts of any emissions reductions.  Using its standard assumption of an equilibrium climate sensitivity of 3.0⁰C, the amount of warming that zero electricity generation emissions from Pennsylvania would mitigate by 2050 is 0.04⁰F and the amount of sea-level rise mitigated is 0.002 (two-thousandths) of an inch.  Both are well below our ability to even measure.

Summary of Comments

Using the model that works, the Russian INM-CM4, reduces prospective warming by 60%.  Using the likely emissions pathway reduces this further by 30%, ultimately reducing the 2000-2050 warming from 5.4⁰ to 1.7⁰F, or approximately one degree Celsius.  The amount of warming and sea-level rise mitigated by Pennsylvania under any circumstance could not be measured by 2050 (or 2100, for that matter). The Pennsylvania Climate Action Plan report, which serves as the basis for Governor Wolf’s Executive Order 2019-07, needs to be dramatically revised, and should no longer be used as the basis for any policy proposals in its present form.

This strong language applies to the PCAP.  As Hausfather told the BBC concerning his work, RCP 8.5 is “exceedingly unlikely.”

As shown in Figure 2, the likely scenario, given policies that already exist, is an RCP value of between 4 and 6, instead of 8.5. This would reduce prospective warming by between 30 and 50%, as warming rises linearly with the RCP value.  Using the most conservative version of the correct RCP value reduces warming another 1.6⁰F, leaving a remainder from 2000-2050 of 1.7⁰F, a far cry from the original PCAP value of 5.4⁰.

As a thought experiment, suppose that Pennsylvania ceased any carbon dioxide emissions from electricity generation a decade ago.  The EPA’s Model for the Assessment of Greenhouse-Gas Induced Climate Change27 (the acronym is MAGICC) is the standard tool used to assess the climate impacts of any emissions reductions.  Using its standard assumption of an equilibrium climate sensitivity of 3.0⁰C, the amount of warming that zero electricity generation emissions from Pennsylvania would mitigate by 2050 is 0.04⁰F and the amount of sea-level rise mitigated is 0.002 (two-thousandths) of an inch.  Both are well below our ability to even measure.

Summary of Comments

Using the model that works, the Russian INM-CM4, reduces prospective warming by 60%.  Using the likely emissions pathway reduces this further by 30%, ultimately reducing the 2000-2050 warming from 5.4⁰ to 1.7⁰F, or approximately one degree Celsius.  The amount of warming and sea-level rise mitigated by Pennsylvania under any circumstance could not be measured by 2050 (or 2100, for that matter). The Pennsylvania Climate Action Plan report, which serves as the basis for Governor Wolf’s Executive Order 2019-07, needs to be dramatically revised, and should no longer be used as the basis for any policy proposals in its present form.

]]>

Join our effort to build a greener future.