Risk Analysis

Research Focus

The effects of various adaptation and mitigation measures may be uncertain, both with respect to effectiveness of their intended outcome as well as to potential side effects. In this working group we derive strategies to hedge against both types of uncertainties. Decision-makers typically ask not only for the efficiency of a policy under uncertainty (in terms of expected outcome), but also for robustness under uncertainty.

We expect that redirecting investment streams within the energy sector has a high potential to mitigate further global warming. However the effectiveness of such measures crucially depends on many factors such as the learning rate of renewables, the fossil resource base, leakage rate of CO2 stored in geological formations, climate sensitivity and climate response time scales. These are just a few of the crucial, yet uncertain characteristics of the coupled human-climate-system. We ask how investment streams should be allocated in face of these uncertainties, without and with anticipating future learning on those system properties. Thereby we derive the option value of various technologies, in particular in their role as insurance against potential failure of seemingly first-best technologies. Among other items we investigate how much of GWP should be invested in renewable sources and in CCS in parallel over the next 25 years.

Investment under anticipated learning

Climate system properties may be revealed under further investigation of historic excursions in the temperature/greenhouse gas record as conserved in paleo archives. We determine the economic value of information of further constraining climate response parameters.
Somewhat different measures are needed to deal with uncertainties that may reduce only through implementing certain mitigation measures: we will ultimately learn on the leakage rate of CO2 only after a certain amount has been sequestered underground. Here it will be important to operationalise a bond scheme that provides high incentives for future sequestration in secure formations. We will distil how the cycle of iterated investment, injection, observation and evaluation can be optimally supported by such an incentive scheme. Hereby it is crucial that the CO2 observation system is powerful enough to inform the capital market for rational further investments in specific geological formations. Together with geo-scientific institutions we aim to derive the value of information of a powerful observation network. Along those lines, also uncertainty on the learning rates of new energy sources will only reduce after investments in those sources has been undertaken to a certain extent. We derive cost effective investment streams that anticipate future investment-induced learning in the energy sector.

Management of deep uncertainty

While many aspects of uncertainty can be managed in terms of insurance, in situations of deep uncertainty, the precautionary principle or competing principles may be operationalised. In view of such deep uncertainty we aim at ranking adaptation and mitigation measures w.r.t. non-monetary metrics. Among other items, such a metric must be able to rank CO2 sequestration in geological formations against deep ocean injection of CO2. ‘Insurability’ of side-effects will be part of that metric. Most uncertainties related to the climate problem are of somewhat mixed nature between the poles ‘insurable’ vs. ‘deeply uncertain’. Accordingly we adopt concepts from the theory of imprecise measures to start from stylised, and then more and more complex decisions under such type of uncertainty.

Team

	Hermann Held		Elmar Kriegler
	Alexander Lorenz		Matthias Schmidt

Publications

• Schmidt, MGW, Lorenz A, Held H, Kriegler E:
  Climate Targets under Uncertainty: Challenges and Remedies
  Climatic Change Letters 104 (3-4), 783-791, JAN 2011

• E. Kriegler. J. Hall, H. Held, R. Dawson, H.-J. Schellnhuber (2009)
  Imprecise probability assessment of tipping points in the climate system.
  Proceedings of the National Academy of Sciences of the United States of America (PNAS), 106(13): 5041-5046. doi: 10.1073/pnas.0809117106

• E. Kriegler (2009)
  Updating under unknown unknowns: An extension of Bayes' rule.
  International Journal of Approximate Reasoning 50(4): 583-596. doi: 10.1016/j.ijar.2008.09.005

• H. Held, E. Kriegler, K. Lessmann, O. Edenhofer (2009)
  Efficient climate policies under technology and climate uncertainty.
  Energy Economics 31(1): S50-S61. doi: 10.1016/j.eneco.2008.12.012

• D. Patino-Echeverri, P. Fischbeck, E. Kriegler (2009)
  Economic and environmental costs of regulatory uncertainty for coal-fired power plants.
  Environmental Science & Technology 43(3): 578-584. doi: 10.1021/es800094h

• H. Held, T. Augustin, E. Kriegler (2008)
  Bayesian learning for a class of priors with prescribed marginals.
  International Journal of Approximate Reasoning 49(1) 2008: 212-233.