Uncertainty representations of mean sea-level change: a telephone game?

For the long-term management of coastal flood risks, investment and policy strategies need to be developed in the light of the full range of uncertainties associated with mean sea-level rise (SLR). This, however, remains a challenge due to deep uncertainties involved in SLR assessments, many ways of representing uncertainties, and a lack of common terminology for referring to these. To contribute to addressing these limitations, this paper first develops a typology of representations of SLR uncertainty by categorising these at three levels: (i) SLR scenarios versus SLR predictions, (ii) the type of variable that is used to represent SLR uncertainty, and (iii) partial versus complete uncertainty representations. Next, it is analysed how mean SLR uncertainty is represented and how representations are converted within the following three strands of literature: SLR assessments, impact assessments, and decision analyses. We find that SLR assessments mostly produce partial or complete precise probabilistic scenarios. The likely ranges in the report of the Intergovernmental Panel on Climate Change are a noteworthy example of partial imprecise probabilistic scenarios. SLR impact assessments and decision analyses mostly use deterministic scenarios. In conversions of uncertainty representations, a range of arbitrary assumptions are made, for example on functional forms of probability distributions and relevant confidence levels. The loss of quality and the loss of information can be reduced by disregarding deterministic and complete precise probabilistic predictions for decisions with time horizons of several decades or centuries, and by constructing imprecise probabilistic predictions and using these in approaches for robust decision-making.

The online version of this article is accesible here

 

An agent-based modeling for housing prices with bounded rationality

This study proposes an agent-based model to investigate major stakeholders behaviors in the housing market. The proposed model mimics the heterogeneous behaviors of individual buyers and sellers in a housing market considering bounded rationality. The simulation results of case study in Shanghai are robust and reproduce stylized facts including as volatility clustering, absence of autocorrelations, heavy tail, loss asymmetry, and aggregational gaussianity on the absolute return.

Access the full text here.

GCF’s Decision Theater in the media

 

Tagesspiegel author Susanne Ehlerding gives a thoughtful insight on what climate models can and cannot do, and how the Decision Theater of GCF contributes to developing models interactively between scientists and stakeholders so that models can serve as practically useful guidance aids.

Read the full article here