Why are there such different estimates of the global warming implied by the Paris pledges?

Why are there such different estimates of the global warming implied by the Paris pledges?

Share this:
Story detail:
Date: 20th November 2015
Author: CDKN Global
Type: Feature
Tags: temperature, UNFCCC

Do the Intended Nationally Determined Contributions (INDCS) - the pledges made by countries before the Paris climate summit - actually add up to a two degree deal, a three degree deal, or worse? Bert Metz of the European Climate Foundation explains how the scenarios are calculated.

More than 160 countries have submitted Intended Nationally Determined Contributions (INDCs) to the UN Framework Convention on Climate Change ahead of the Paris Conference of Parties. Research groups have published what these INDCs collectively mean in terms of the 2025 and 2030 global emission levels and the corresponding implications for long-term global mean temperature. Two high-profile UN reports have summarized this research: the UNFCCC Synthesis Report[1] and the UNEP Emissions Gap Report[2].

The common message of these publications is that INDCs are bending the curve of global emissions, lead to a lower 2100 global temperature than with current policies and are not enough to be on track to keeping global mean temperature increase below 2oC. However, estimates of what 2100 global mean temperature increase is implied in the collective INDCs vary considerably. The most widely quoted numbers are as follows.

The Climate Action Tracker[3] estimated an increase of 2.7oC (range 2.2-3.4) by the end of the century. The Climate Interactive team[4] estimated 3.5oC, and IEA[5] came up with an estimate of 2.7oC as well, all for unconditional INDCs only. The UNEP Emissions Gap Report, which assesses all available research findings, estimated 3.5oC (range 3-4) if looking at only unconditional INDCs and 3oC if including conditional ones as well (no range is given for this number, but it can be assumed it is comparable to the range for the unconditional INDCs, i.e 2.5-3.5). A more comprehensive overview and analysis has been published by WRI [9].

Why are there such differences?

There are a couple of reasons:

Individual estimates of global emission levels from implementing INDCs differ. Many of the INDCs submitted by countries do not specify 2025 or 2030 emission levels as a result of the INDC, so these emission levels need to be calculated, requiring certain assumptions to be made. The UNEP Emissions Gap report shows that between research groups there can be a difference of more than10%, with the CAT estimates being on the low end of the range.

Some research groups include all INDCs, unconditional and conditional ones, but others only include unconditional ones. The difference, according to the UNEP Emissions Gap Report is about 2 GtCO2e.

The calculation of long-term temperatures for a given global emission level differs. The 2100 temperatures depends not only on the 2025 and 2030 emission levels, but more on what happens with emissions thereafter. Different methodologies are used[6],[7],[8]. The Climate Interactive assumptions for post 2030 action are much more conservative than what others have done.

The definition of the statistical chance of staying below a certain temperature level differs. Due to the uncertainties in translating an emission level into a global mean temperature increase, it makes a difference if the threshold is defined as a 50% chance of staying below a certain temperature or as 66% or 90% chance. The UNEP Emissions Gap Report uses a 66% chance as the definition of the threshold. Their finding that with 66% chance INDCs will keep temperatures below 3.5oC (for unconditional INDCs).  The IEA and CAT findings that there is a 50% chance that temperatures will be kept below 2.7oC is equivalent to a 66% chance below 3oC.

So, when looking at the factors discussed above, it can be concluded that there are three main reasons for the differences: (1) the definition of the statistical threshold:  (2) the difference in assumptions about  post-2030 action; and (3) the estimate of the emission levels in 2025/2030 implied in the INDCs. The approach followed by the UNEP Emissions Gap Report (66% probability, looking at selected scenarios from the IPCC database for the post-2030 assumptions, and averaging the emission levels found by different research groups) gives a good basis for assessing the long-term temperature implications of the INDCs: 3.5oC for only unconditional and 3oC for conditional and unconditional pledges.


[1] UNFCCC Synthesis Report on the aggregate effect of intended nationally determined contributions, http://unfccc.int/resource/docs/2015/cop21/eng/07.pdf   (does not include temperature estimates, but has its own calculations of the emissions gap; press release mentions 2.7oC, see http://newsroom.unfccc.int/unfccc-newsroom/indc-synthesis-report-press-release )

[2] UNEP Emissions Gap Report 2015, Executive Summary, http://www.unep.org/publications

[3] Climate Action Tracker, http://climateactiontracker.org/assets/publications/CAT_global_temperature_update_October_2015.pdf

[4] Climate Interactive, https://www.climateinteractive.org/tools/scoreboard

[5] IEA, World Energy Outlook 2015 presentation, http://www.iea.org/media/weowebsite/2015/151110_WEO2015_presentation.pdf

[6] See C-Roads Reference Guide, https://www.climateinteractive.org/wp-content/uploads/2014/01/C-ROADS_Reference_Guide_v68.pdf and methodology used for post 2030 emission scenarios, https://www.climateinteractive.org/tools/scoreboard/scoreboard-science-and-data

[7] http://climateactiontracker.org/methodology/18/Global-pathways.html

[8] UNEP Emissions Gap Report 2015, in press, http://www.unep.org/publications

[9] http://www.wri.org/blog/2015/11/insider-why-are-indc-studies-reaching-different-temperature-estimates




Add new comment

Plain text

  • No HTML tags allowed.
  • Lines and paragraphs break automatically.
  • Web page addresses and email addresses turn into links automatically.
CAPTCHA This question is for testing whether or not you are a human visitor and to prevent automated spam submissions.