Comparing CMIP5 & observations

[Last updated: 11th May 2022]

This page is an ongoing effort to compare observations of global temperature with CMIP5 simulations assessed by the IPCC 5th Assessment Report. The first two figures below are based on Figure 11.25a,b from IPCC AR5 which were originally produced in mid-2013.

The first panel shows the raw ‘spaghetti’ projections, with different observational datasets in black and the different emission scenarios (RCPs) shown in colours. The simulation data uses spatially complete coverage of surface air temperature whereas the observations use a spatially incomplete mix of air temperatures over land and sea surface temperatures over the ocean. It is expected that this factor alone would cause the observations to show smaller trends than the simulations.

Based on IPCC AR5 Figure 11.25a, showing observations and the CMIP5 model projections relative to 1986-2005. The black lines represent observational datasets (HadCRUT5.0, ERA5, NASA GISTEMP, NOAA GlobalTemp, BEST)

The second panel shows the AR5 assessment for global temperatures in the 2016-2035 period. The HadCRUT5.0 observations are shown in black with their 5-95% uncertainty. Other observational datasets are shown in blue. The light grey shading shows the CMIP5 5-95% range for historical (pre-2005) & all future forcing pathways (RCPs, post-2005) relative to 1986-2005; the grey lines show the min-max range. The dark grey shading shows the projections using a 2006-2012 reference period. The red hatching shows the IPCC AR5 indicative likely (>66%) range for the 2016-2035 period.

The observations for 2016-2021 fall in the warmer half of the ‘likely’ range. 2016 was warmed slightly by the El Nino event in the Pacific. The years 2015-2021 were all more than 1°C above an 1850-1900 (early-industrial) baseline. In AR5 it was assessed that 1986-2005 was 0.61K warmer than 1850-1900 using HadCRUT4. Subsequent updates to the global temperature datasets have increased this value meaning the green axis has been moved a little lower.

Based on IPCC AR5 Figure 11.25b with the HadCRUT5.0 global temperature time-series and uncertainty (black). The CMIP5 model projections are shown relative to 1986-2005 (light grey) and 2006-2012 (dark grey). The red hatching is the IPCC AR5 indicative likely range for global temperatures in the 2016-2035 period, with the black bar being the assessed 2016-2035 average. The blue lines represent other observational datasets (Cowtan & Way, NASA GISTEMP, NOAA GlobalTemp, BEST). The green axis shows temperatures relative to 1850-1900 (early-industrial period).

One interesting question to consider is: given post-2012 temperature data and scientific studies which have appeared after AR5, would the overall assessment (black bar & red hatching) be changed? In AR5 the assessment was that temperatures would be 0.3-0.7K above 1986-2005 for the 2016-2035 period average (as shown by the black bar). My personal view is that the upper limit may remain the same (or increase slightly), but there would be an argument for raising the lower boundary to 0.4K, mainly because of improved understanding of the effect of missing temperature data in the rapidly-warming Arctic.

There are several possible explanations for why the earlier observations are at the lower end of the CMIP5 range. First, there is internal climate variability, which can cause temperatures to temporarily rise faster or slower than expected. Second, the radiative forcings used after 2005 are from the RCPs, rather than as observed. Given that there have been some small volcanic eruptions and a dip in solar activity, this has likely caused some of the apparent discrepancy. Third, the real world may have a climate sensitivity towards the lower end of the CMIP5 range. Next, the exact position of the observations within the CMIP5 range depends slightly on the reference period chosen. Lastly, this is not an apples-with-apples comparison because it is comparing air temperatures everywhere (simulations) with blended and sparse observations of air temperature and sea temperatures. A combination of some of these factors is likely responsible.

11th May 2022: Graphics updated for 2021 global temperature data
4th May 2021: Text tweaked to comply with IPCC legal requests to say that these updates are ‘based on’ rather than ‘updated from’ IPCC figures. These graphics are (obviously) not official updates by the IPCC itself. (sigh)
25th January 2021: Graphics updated for 2020 global temperature data from HadCRUT5
9th September 2020: Page updated for 2019 global temperatures.
9th July 2019: Entire page updated for 2018 global temperatures.
21st February 2018: Entire page updated for 2017 global temperatures.
23rd January 2017: Entire page updated for 2016 global temperatures.
22nd February 2016: Update to Fig. 1.4 from IPCC AR5 added.
26th January 2016: Entire page updated for 2015 global temperatures.
2nd September 2015: Updated figure to use HadCRUT4.4 up to July 2015 and added link to Cowtan et al. (2015).
5th June 2015: Updated using data from HadCRUT4.3 up to April 2015, and the new NOAA dataset.
2nd February 2015: Cowtan & Way 2014 data added.
26th January 2015: Entire page updated for 2014 temperatures.
27th January 2014: Page created.

94 thoughts on “Comparing CMIP5 & observations

  1. Interesting that empirical global temperature rose when the entire world economies were shut down due to COVID. It would seem that cutting CO2 use has virtually no effect on global temperature.

Leave a Reply

Your email address will not be published. Required fields are marked *