The Fuzziness of the Paris Climate Targets

The Fuzziness of the Paris Climate Targets

By John McLean, Melbourne, Australia

Governments that signed the Paris Climate Agreement and committed to making payments to the Green Climate Fund have done so at their own peril. The absence of key data, methods of calculation and key thresholds regarding future temperature means that the potential for further financial liabilities is very significant.

On Tuesday, November 7, Syria announced its intention to join the Paris Climate Agreement leaving the USA as the only country to reject it, albeit through signing and then announcing its withdrawal. A total of 169 countries have now signed the agreement, 42 of those making financial commitments to the UNFCCC’s Green Climate Fund.

I suggest that few, if any, countries undertook due diligence before committing to this agreement. I don’t mean about the science, which to me is still contentious, but about the aims, details and the terms of the agreement. Not to do so has likely exposed them to future problems.

Article 2.1(a) of the agreement sets out the primary aim – “holding the increase in the global average temperature to below 2°C above pre-industrial levels and pursuing efforts to limit the temperature increase to 1.5°C above pre-industrial levels …”

Nowhere in the agreement is the pre-industrial global average temperature defined nor are the means by which it was or can be calculated and which year or years “pre-industrial” refers to.

Nor for that matter does it describe how average global temperatures should be calculated today, but at least some datasets are available (e.g. HadCRUT4, NASA GISS) to suggest a methodology.

We cannot calculate a pre-industrial global average if we don’t know the period that the expression refers to, so that is the first thing to establish.

The glossary to the 2013 IPCC climate assessment report says under its definition of “industrial Age”, like the two reports that preceded it, “In this report the terms pre-industrial and industrial refer, somewhat arbitrarily, to the periods before and after 1750, respectively”.

In contrast, chapter 7 of the same IPCC report says “For consistency with AR4 and Chapter 8 of this Report, … we evaluate the forcings between 1750 and approximately 2010. The reference year of 1750 is chosen to represent pre-industrial times, … “

Hawkins et al (2017), which we’ll return to shortly, argues instead for a time span from 1720 to 1800. In contrast at various times authors of papers have used data from 1850, the start of modern temperature records, as the “pre-industrial” benchmark, whereas a UNFCCC document, apparently sourced from Oxford University, refers to an “1861-1880 pre-industrial baseline” [1].

Let’s start with 1750. According to data from the UK’s Climatic Research Unit (CRU) at the University of East Anglia [2], only three observation stations reported temperature data in that year – Berlin (Germany), De Bilt (Netherlands) and Uppsala (Sweden). St Petersburg (Russia) reported data in the general period but data for 1750 is missing. The three reporting stations are all in Europe, which covers less than 10% of the Northern Hemisphere and at that time was in the grip of the Little Ice Age.

The ICOADS database shows that only 136 temperature observations were made at sea in 1750. All appear to be from a single ship travelling from near Kristiansand, Norway, to Britain and then to Mumbai, India, during the period from April 10 to September 18. Observations were made on most days but only one observation was made each day, seemingly at midday Greenwich (UK) time, which on the voyage meant on the hour somewhere between 7am and 1pm local time.

Year 1800 is not much better for calculating a pre-industrial average. The CRU data reveals that 37 observation stations were in operation. Two were on the east coast of the USA, one in Greenland and the remaining 34 all in Europe, which was still under the influence of the Little Ice Age. Temperature measurements at sea totalled 5438 for the year, but almost all were along trade routes from Europe, particularly around Africa to European territories or trading ports in southern or south-east Asia.

Finally let’s turn to the modern temperature record. The IPCC’s principal source of temperature data, the HadCRUT4 dataset, indicates global coverage of approximately 23% in 1850, 20% in 1860, 23% in 1870 and 34% by 1880. During the first three of the decades noted here it was usually the case that Europe or the trade routes mentioned above accounted for more than 60% of the data coverage and sometimes more than 70%.

No observational data at any time from 1750 to 1880 can be used to calculate the global average temperature because of the heavy focus on a small portion of the Earth’s surface, not to mention that the conditions in that region were unusually cold by twentieth and twenty-first century norms.

Perhaps the UNFCCC has relied on climate modelling to provide the temperature in question but that would be questionable too.

Hawkins et al (2017) [3], which was published more than 18 months after the Paris Climate Agreement was made available for signing, discusses the UNFCCC’s use of the term “pre-industrial temperature” and attempts to clarify the matter. It cites several historical temperature studies, each time including terms like “estimated”, “interpolated” or “modelled”, all of which are of uncertain accuracy, before comparing the HadCRUT4 global average temperature anomalies averaged over the period 1850 to 1990 to the output of three CIMP5 climate models.

The 1850-1900 averages are seriously compromised for reasons given above and each of the climate models was used in IPCC 5AR (2013)[4], which stated “… an analysis of the full suite of CMIP5 historical simulations … reveals that 111 out of 114 realisations show a GMST [global mean surface temperature] trend over 1998–2012 that is higher than the entire HadCRUT4 trend ensemble …”[5] (i.e. higher than the trend derived from temperature observations ).

The IPCC report offered some possible reasons for the apparent failure of CIMP5 models, saying among other things “There may also be a contribution from forcing inadequacies and, in some models, an overestimate of the response to increasing greenhouse gas …” [6], which is very relevant given the UNFCC’s emphasis on greenhouse gases.

Based on the IPCC’s statements climate models have little credibility, a point supported by other researchers.

This leaves the Paris Climate Agreement with an uncertain reference temperature, calculated or derived over an uncertain period, using methods that are not described.

Worse again, the agreement fails to state whether the unspecified temperature needs only be exceeded by 1.5°C (or 2.0°C) for one month, one year (i.e. an annual average) or as an average over some extended period such as five years. It also pays no regard to the influence of other climate forces that might cause warming, such as El Nino events.

If the UNFCCC declared tomorrow that temperatures were now 2°C, or even 1.5°C, above pre-industrial levels and that more money was required from developed countries then not one of those countries could dispute the point.

To put the likelihood of that in context, the first IPCC climate assessment report [7] says that temperatures were at that time already about 1°C above pre-industrial levels.

More recently the World Meteorological Organisation (WMO), a close ally of the UNFCCC, issued a media release in January 2017 saying that the 2016 global average temperature was “about 1.1°C higher than the pre-industrial period” [8]. Like the UNFCC it failed to define the reference temperature, the years in question or the method of calculation.

Governments that signed the Paris Climate Agreement and committed to making payments to the Green Climate Fund have done so at their own peril. The absence of key data, methods of calculation and key thresholds regarding future temperature means that the potential for further financial liabilities is very significant.

ENDS

© Scoop Media