Is it generally understood that we (globally) have been living in an RCP8.5 scenario over the past 7-8 years?

Such that a predictive climate model using RCP8.5 assumptions that was run circa 2011 can be considered indicative of the time between then and now?

I am working on a calibration of modeled dry-bulb temp data versus real observations over the intervening years but I realize that any correction I come up with would necessarily have any deviation from reality and RCP8.5 baked into it.


In short, I would say that it is too soon to say. But, if I were forced to guess: I would indeed incline to say that we live in a RCP 8.5-ish scenario.

I'm not an expert at all on climate models or RCP scenarios, but I found the question interesting, so this is my attempt to answer it based on the most reliable data sources I found. I hope to motivate expert to jump in to comment, improve, challenge or support this answer.

RCP scenarios

The scenarios involve future pathways in the emissions of a large range of different chemicals (Greenhouse gases and aerosols), so if we look at how actual emissions of $CO_2$ have matched the scenario we might think we are in one scenario, but is we instead look at the emissions of $NH_3$ we might find ourselves in another scenario. The best way to summarise all those different parameters is to consider the total radiative forcing, that is the combined effect of all those compounds in Earth's energy balance. In fact the numbers that label the different RCP's are the expected total radiative forcing by year 2100 in $W/m^2$.

The following figure shows the radiative forcing (this time as $CO_2$ equivalent) of each scenario. And as you can see, by 2018, the differences are so small that I think it is too soon to say with any reasonable certainty that we are flowing one or another scenario. enter image description here (Figure from Wikipedia commons originally here)

Scenarios versus measurements

Nevertheless, we can attempt to plot the actual measurements against what was predicted by the scenarios. The best variable to contrast would arguably be the total radiative forcing. However, that is a quantity difficult to calculate and I haven't found published values like those in the IPCC Assesment report 5 (AR5) Annex II table AII.1.2, beyond 2011. Up to that year, the forcings were beautifully summarized in figure 8.18 of AR5 chapter 8: enter image description here

Given the lack of total radiative forcing data, we can instead make the comparison with some other variable. Such analysis will be liable to depend on the chosen variable as pointed out before. Therefore, to minimize that problem let's pick a variable with a large influence in the difference between the four scenarios. I think that would be $CO_2$ emissions from fossil fuels and industry.

The the emissions predicted by each scenario can be obtained at the RCP Database (version 2.0). That data looks like this: enter image description here

Now, for the actual estimated emissions, perhaps a reliable source is the Global Carbon Project, that provide the emissions data up to 2016. Putting all that together we get the following figure: enter image description here

So, it looks RCP8.5-ish, but the differences between the scenarios are still quite small, so it is difficult to say.

Another good variable to compare would be $CO_2$ concentration, that can be directly measured instead of just estimated. In that case using NOAA's Global Greenhouse Gas Reference Network data for global $CO_2$ concentration, against the ones prescribed for each scenario (taken as well from RCP Database (version 2.0)) the figure looks like this enter image description here

That looks more like RCP4.5-ish. But again, probably too soon to say. Also, it might be that even given more time, the path followed by the climate system would be like one RCP in some aspects and like other RCP in other aspect. For example, $CO_2$ emissions might follow roughly RCP8.4 but methane emission might follow RCP4.5. Or somewhere in between.

Another approach would be to use use an actual climate model forced by RCP8.5 (not just the RCP), so you can compare mean surface temperature, but that would be tricky for many reasons and also model dependent.

If anybody knows of published estimated total radiative forcing from 2012 onward, that would be nice to add and compare. It might make the answer a bit more clear but I think that in any case, it is too soon to give a definite answer.

  • $\begingroup$ Comparing the situation of between now and the release of AR5 is what is asked, interesting and what you have not done. AR5 working groups were set 2009 and the physical basis part released 2013, so the climate models have been run between these dates and model results for years before this as you can see from your plot don't match reality (property of the models). Since all your data is from AR5 there is no way you could compare how the predictions of AR5 have matched the reality better than AR5 itself. That being said I too don't know where to get estimated forcings for 2012-2017. $\endgroup$ – Communisty Mar 10 '18 at 22:33
  • 1
    $\begingroup$ @Communisty Ok, fair enough. I've changed the answer to a different approach. What do you think now? $\endgroup$ – Camilo Rada Mar 11 '18 at 2:32
  • $\begingroup$ Good, the discrepancy between CO2 emissions and measurements could be perhaps because bio- and hydrosphere are better buffers than the models estimate. $\endgroup$ – Communisty Mar 12 '18 at 8:54
  • $\begingroup$ @Communisty I agree. Perhaps I should emphasize the difference between basically prescribed variables as emissions and modeled variables like concentration. Some time the RCP are also referred as “benchmark emissions scenarios”, which put more emphasis in the emissions. Arguably, makes more sense to answer this question looking at emissions instead of concentrations. $\endgroup$ – Camilo Rada Mar 12 '18 at 17:00

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.