IPCC SR1.5 – Notes on Calculations and Assumptions

Given that my previous post was about failing to reconcile the emissions estimates for 1.5°C and 2.0°C of warming in the IPCC fifth assessment report (AR5), I was intrigued to see how the new IPCC “special report on the impacts of global warming of 1.5 °C above pre-industrial levels” would fare. However, that will have to wait for another post, as first there are some “refinements” from AR5 in how results are obtained. From my analysis they would appear that key figures on temperatures and climate sensitivities are highly contrived.

Isn’t 1.5°C of warming already built in? 

Chapter 1 Page 24

Expert judgement based on the available evidence (including model simulations, radiative forcing and climate sensitivity) suggests that if all anthropogenic emissions were reduced to zero immediately, any further warming beyond the 1°C already experienced would likely be less than 0.5°C over the next two to three decades, and also likely less than 0.5°C on a century timescale.

This basically states that if all emissions were stopped now there is more than a 50% chance that warming would not exceed 1.5°C. But using previous assumptions 1.5°C should be already be built in. 

If ECS = 3.0 (as in AR5) then that implies the net effect of all GHGs and all aerosols is less than 396 ppm, despite CO2 on its own in September 2018 being 405.5 ppm (1.6°C of eventual warming). Further, in 2011 the impact of all GHGs combined was equivalent to 430 ppm, or an extra 40 ppm more than CO2 on its own. On that basis we are at the around 445 ppm or fractionally about the 2.0°C warming level. However, in AR5 it was assumed (based on vague estimates) that the negative human impacts of aerosols exactly offset the addition of other GHGs (e.g. methane) so that only CO2 is considered. Even then based on ECS = 3.0 without further emissions 1.5°C will be eventually reached.

But ECS has been lowered.

From Chapter 1 Annex Page 11

…Equilibrium Climate Sensitivity (ECS) of 2.7°C and Transient Climate Response (TCR) of 1.6°C and other parameters as given in Millar et al. (2017).

This raises the CO2-eq level to achieve 1.5°C of warming by 15-16 ppm from 396ppm and the CO2-eq level to achieve 2.0°C by 23-24 ppm from 444 ppm. Mauna Loa CO2 levels in September averaged 405.5 ppm. With ECS = 2.7 this is equivalent to just 1.44°C of eventual warming compared to 1.60°C  when ECS = 3.0. What is more significant is that if ECS were 2.8 eventual warming of 1.50°C would be in the pipeline sometime before the end of the year. ECS = 2.7 is the highest ECS that us currently compatible with the statement made above if CO2 alone is taken into account. Consider this in the light of 2013 AR5 WG1 SPM, which stated on page 16

Equilibrium climate sensitivity is likely in the range 1.5°C to 4.5°C

And in a footnote on the same page.

No best estimate for equilibrium climate sensitivity can now be given because of a lack of agreement on values across assessed lines of evidence and studies.

 In AR5 they chose ECS = 3.0 as it was in the middle of the range. A range unchanged since the Charney Report of 1979. I am not aware of any that establishes ECS is a range that would justify ECS = 2.7 that is not contradicted by other research. For instance Lewis and Curry 2018 gives a median estimate for ECS of 1.66.

Transient Climate Response (TCR)

But how does the Transient Climate Response (TCR) of 1.6°C fit into this? Some context can be had from the very start of the Summary for Policy-Makers SPM-4

A1.1. Reflecting the long-term warming trend since pre-industrial times, observed global mean surface temperature (GMST) for the decade 2006–2015 was 0.87°C (likely between 0.75°C and 0.99°C)

With TCR = 1.6°C for a doubling of CO2 levels what is the warming generated from a rise in CO2 levels from 280 to 400.83 ppm? That is a rise in CO2 levels from pre-industrial times to the average level in 2015. I calculate it to be 0.83°C. Make TCR = 1.7°C and that increases to 0.88°C. It is effectively assuming that both 100% of the rise in average temperatures in over 150 years is due to CO2 alone (consistent with AR5), and there has been no movement whatsoever from the short-term Transient Climate Response to the long-term Equilibrium Climate Sensitivity. However, if TCR is a variable figure derived from a calculation from revealed warming and CO2 rise, it becomes meaningless nonsense unless you can clearly demonstrate the other assumptions are robust. That is (1) 100% of past warming was due to human emissions (2) the impact of GHGs other than CO2 are effectively cancelled out by aerosols etc. (3) natural factors are net zero (4) the average temperature data anomaly is without any systematic biases. For instance, when measured CO2 levels were about 390ppm, the AR5 WG3 SPM stated in the last sentence on page 8

For comparison, the CO2-eq concentration in 2011 is estimated to be 430 ppm (uncertainty range 340 to 520 ppm)

It seems a pretty shaky foundation to the assumption that negative impact of aerosols (with uncertainties) will offset the combined impact of other GHG increases.

Summary and Concluding Comments

On the estimates of climate sensitivity, it appears to be set so that the IPCC can still claim that if emissions stopped tomorrow then there would be a greater than 50% chance of 1.5°C warming never been exceeded. The ECS value of 2.7°C is set at the maximum value, given the assumptions. But ceteris paribus, this will not hold if

  • One waits 3 years and CO2 levels continue increasing at a rate of the last few years.
  • ECS is slightly higher but still well within the accepted range of estimates. Indeed if ECS = 3.0, as in AR5 and AR4 in 2007, then 1.5C of warming was exceeded 5 years ago.
  • The impact of all GHGs together is slightly more than the offsetting impacts of other aerosols.
  • 0.06°C, or more, of the observed rise on temperature since 1850 is not due to GHG emissions.

Then there is the Transient Climate Response (TCR) which appears to be little more than taking the historical temperature change, assuming all of is down to human GHG emissions, and calculating a figure. Including rises in CO2 a century or more ago is hardly transient.

Based on my calculations, the results are highly contrived. They appear as a very fine balance between getting the maximum values for human-caused warming possible and not admitting that 1.5°C or even 2°C is already passed. There is a huge combination of empirical assumptions that are as equally valid as the ones used in the SR1.5 that go one way or the other. Rather than being a robust case, empirically it is highly improbable one.

Finally there is a conundrum here. I have calculated that if ECS = 2.7 and the starting level of CO2 is 280 ppm, then in round numbers, 1.5°C of warming results from CO2 levels of 412 ppm and 2.0°C of warming results from CO2 levels of 468 ppm. With CO2 levels in September 2018 at 406 ppm for 2.0°C of warming requires a rise in CO2 ten times greater than for 1.5°C of warming. So how can the IPCC claim that it is only about twice the amount of emissions? In my previous post I could not find an explanation, even though the emissions numbers reconciled with both past data and future emissions to generate 2.0°C of warming given certain assumptions. In the next I hope to provide an answer, which fits the figures quite closely, but looks pretty embarrassing.

Kevin Marshall