The following is discussion of what can be considered as serious systemic errors in the IPCC’s ‘SRREN’ – a report that examined ‘renewable’ energy usage scenarios into the future. It is divided into two serial parts (I & II) – one leading into another.
Look at the compared 4 scenarios, as summarized in this table (Table 10.3, page 1187, chapter 10 IPCC SRREN). Table 10.3 is key – it summarizes the chapter’s comparison of different ‘scenarios’.
As you can see, these are IEA-WEO2009 Baseline, ReMind-RECIPE, MiniCAM-EMF22 and ER-2010. ER-2010 is the Greenpeace scenario.
Consider the IEA scenario (marked with blue dot) in the table. The IEA scenario is ‘baseline’ in this comparison and it is a forward-looking (prospective) model. In their own words, the IEA WEO 2009 model “calculates the possible energy pathway without any substantial change in government policy and under the assumption of a minimal to moderate fossil fuel cost increase” and no “specific GHG emissions constraints.”.
But the IEA which releases scenarios every year calculated outcomes for only up to the year 2030 in 2009. At this point, the authors of the IPCC chapter did something curious. What they did was to carry out an extrapolation of the IEA scenario that uses the ‘key macroeconomic and energy indicators of IEA (2009)’, and brought it up to the year 2050. This extrapolation was ‘provided’ by the German Aerospace Center (DLR). This is how they describe it:
As the IEA (2009) projection only covers a time horizon up to 2030 for this scenario exercise, an extrapolation of the scenario has been used that was provided by the German Aerospace Center (DLR) that uses the key macroeconomic and energy indicators of IEA (2009) and brings them forward to 2050 (Teske et al., 2010).
Note that Teske et al 2010 is cited. The German Aerospace Center DLR devised the methodology and performed the analytic component in Teske et al 2010. Wolfram Krewitt of the DLR devised the methodology and it is applied in the earlier versions of the Greenpeace scenario, published in 2009 and 2007.
Therefore, the same Teske et al 2010 methodology – a paper which is a target of the assessment – was used to model the complete baseline IEA scenario as well.
This brings a serious question-mark over the inter-comparability and independence of the two scenarios as presented by the IPCC.
There is however, another more subtle, but distressing problem due to this approach.
The ER-2010 (Greenpeace) scenario – marked by red dot – is a backward-looking scenario. That is, key parameters are fixed at the start of the model analysis and only the ‘renewable’ energy source proportion is allowed to vary freely. At the same time ‘socioeconomic indicators’ are adopted from the IEA 2009 scenario as targets (see page 1189, box). As Krewitt et al 2007 (the precursor paper for the Teske et al 2010) says,
A target-oriented scenario of future energy demand and supply is developed in a backcasting process. The main target is to reduce global CO2 emissions to around 10 Gt/a in 2050, thus limiting global average temperature increase to 2C and preventing dangerous anthropogenic interference with the climate system.
The 10 Gt/annum figure for carbon emissions shrunk further to 3.7 Gt/annum in Teske et al 2010.
The main target is to reduce global CO2 emissions to 3.7 Gt/a in 2050, thus limiting global average temperature increase to below 2°C and preventing dangerous anthropogenic interference with the climate system.
These a priori constraints can be considered the reason for their method yielding a high percentage of ‘renewable’ energy source usage by 2050 in ER-2010 .
Then, turn to the fact that the same GDP/capita projection is employed as input, to generate the energy source proportions for the baseline scenario (i.e., IEA-2009 2050). The IEA-WEO-2009 however in addition to being a baseline scenario for the above comparison is forward-looking – only socioeconomic targets are input while the relative proportions of all energy sources (including ‘renewable’ sources) is allowed to vary freely.
The above adversely affects meaningful inter-comparability between the two scenarios. As a result, the Greenpeace scenario’s projections of ‘renewable’ energy adoption and the baseline IEA scenario projections for carbon emissions are both over-inflated in relation to each other, respectively. Attempting the same level of economic progress as found in Teske et al ER-2010 would necessarily drive up carbon emissions in the IEA scenario, and attainment of the same level of economic progress is possible in the Teske et al – Greenpeace scenario only because only ‘renewable’ energy sources are allowed to expand freely with no constraint.
As an example, the situation is analogous to a biological experiment where some of the control cases also appear in the study group or the use of a new experimental methodology where a fraction of the target subjects are themselves used in the training set. An even more accurate analogy would be like attempting to compare two clinical studies: the first one, where a small number of patients with good outcomes are selected and the study examines patient records retrospectively to see that such patients received quality healthcare at substantial healthcare costs, and a second prospective study which aims to attain the same for all included patients, with the exact same good outcomes operating as constraints. Healthcare costs (akin to CO2) in the second study would necessarily be high! The result is circularity of inference, and the root of the problem lies in the cross-inheritance of input parameters between scenarios of fundamentally opposite types.
With loosely based assumptions, substantial bias and circular argument, the very plausibility of such an eventuality as Greenpeace’s, in global energy production perhaps ought to be questioned first. It must be remembered that the IPCC report characterizes scenarios as “plausible description(s) of how the future may develop based on a coherent and internally consistent set of assumptions about key driving forces” (p 1166). In its definition of a ‘scenario’, the IPCC again stresses plausibility. A scenario is a “plausible description of how the future may develop based on a coherent and internally consistent set of assumptions about key relationships and driving forces (e.g., rate of technological change, prices) on social and economic development, energy use, etc” (p 1418). However, when Steve McIntyre argued that the IPCC has not clearly demonstrated or ‘showed’ actual plausibility of these scenarios (‘the realism of these scenarios needs to be closely examined’), the IPCC response, as recounted by a prominent news weblog, was that the “authors of the IPCC chapter involved declined to evaluate the scenarios they looked at in terms of whether they thought they were plausible, let alone likely”.
Nonetheless, the outcomes arrived at in the Greenpeace-EREC scenario was reported by the IPCC, as though they were a distinct and real possibility, as representative of the entire IPCC SRREN report.
At the minimum, it is evident that the baseline and on the scenarios are not truly independent, in terms of methodology and assumptions and it affects their inter-comparability. Clinical trials and meta-analyses rigorously control for such comparison biases which occur yet, on occasion. It is perplexing that important economic analyses would proceed with biases built in to its very structure, to reach wide-ranging conclusions as the IPCC’s report did.
1) IPCC Special Report on Renewable Energy Sources and Climate Change Mitigation [O. Edenhofer, R. Pichs‐Madruga, Y. Sokona, K. Seyboth, P. Matschoss, S. Kadner, T. Zwickel, P. Eickemeier, G. Hansen, S. Schlömer, C. von Stechow (eds)], Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA.
2) Teske et al Energy Efficiency (2011) 4:409–433
3) Krewitt et al Energy Policy 34 (2009) 5764–5775
4) Krewitt et al Energy Policy 35 (2007) 4969–4980
5) IEA 2009 World Energy Outlook (www.worldenergyoutlook.com)
6) Babbage (http://www.economist.com/blogs/babbage/2011/06/ipcc-and-greenpeace)