Recently, our DMDU members Judy Lawrence, Rob Lempert, and Marjolijn Haasnoot wrote a correspondence to Nature. Here we publish an extended version of this correspondence:
Nature Correspondence
From: Judy Lawrence[1], Marjolijn Haasnoot[2], Robert Lempert[3]
The Nature Comment 29 January 2020 by Hausfather and Peters1 argues that the RCP8.5 emissions scenario often used by the IPCC is a highly unlikely “worst case” and that the IPCC should focus instead on more plausible emissions scenarios consistent with about 3°C warming by 2100. They also recommend assigning a single set of best-estimate probabilities to all emissions scenarios, making a decades old argument that if the experts don’t provide a precise set of probabilities the users will themselves.
We agree that the IPCC should also consider emissions scenarios lower than RCP8.5 and indicate that RCP8.5 is not among the most likely. But placing a single set of probabilities on emissions scenarios is neither a necessary nor a robust basis for a climate change risk assessment. Instead decision makers and the IPCC AR6 have available much more appropriate means to manage risk using Decision Making under Deep Uncertainty (DMDU) methods2. Rather than demanding consensus on a particular probability distribution, DMDU methods focus attention on the implications of alternative scenarios and the extent to which response strategies exist in common across a wide range of scenarios. DMDU methods enable short-term and necessary preparatory actions to be made that leave flexibility to change pathway in the future, depending upon whether emissions follow a higher or lower scenario. By lessening the need to assign probabilities, decision makers can better understand the combinations of uncertainties that most affect their choices and thus unlock decision inertia and delay that arise when using only one scenario1.
Decision makers’ policy preferences often depend on their risk tolerance3, the impact of the scenarios and the timescales of adaptation including the lead time required for planning and implementation, and the expected lifetime of a decision4. Informing such decisions requires considering a wide range of likely and unlikely scenarios. For example, sea-level rise has large uncertainties beyond 2050 that have potentially large impacts on people and assets and for which coastal adaptation decisions such as flood defences have a long lead-time and lifetime (e.g. > 50 years)4. DMDU approaches2 such as Dynamic Adaptive Policy Pathways and Robust Decision Making can help manage a range of uncertainties through robust and adaptive decision making by identifying, short-term actions and long-term options5. Such strategies include monitoring of signals (warnings) and trigger points (decision points) to inform implementation or adjustments of adaptive plans. Using only the likely range and a single set of best-estimate probabilities can give decisions makers a false sense of certainty leading to costly adjustments if the world evolves along unanticipated paths.
1. Hausfather, Z. & Peters, G. P. Emissions – the ‘business as usual’ story is misleading. Nature 577, 618–620 (2020).
2. Marchau, V.A.W.J., Walker, W. E., Bloemen, P. J. T. M. & Popper, S. W. Decision making under Deep Uncertainty – From Theory to Practice. (Springer {US}, 2018).
3. Hinkel, J. et al. Meeting User Needs for Sea Level Rise Information: A Decision Analysis Perspective. Earth’s Futures. 7, 320–337 (2019).
4. Haasnoot, M. et al. Adaptation to uncertain sea-level rise; how uncertainty in Antarctic mass-loss impacts the coastal adaptation strategy of the Netherlands. Environ. Res. Lett. (2019) doi:10.1088/1748-9326/ab666c.
5. Lempert,
R.J. Addressing Uncertainty in Developing Long-Term Greenhouse Gas Emission
Reduction Strategies, World Research Institute Perspectives (2019).
[1] New Zealand Climate Change Research Institute, Victoria University of Wellington, New Zealand ORCHID # 0000-0001-6798-3636 Corresponding author: email judy.lawrence@vuw.ac.nz
[2] Deltares and the University of Utrecht The Netherlands. ORCHID # 0000-0002-9062-4698
[3] RAND Corporation, USA ORCHID # 0000-0003-0537-3159