Loading Session...

Diversity, Uncertainty, and Action: Coping with a Plurality of Climate Models

Session Information

Two difficult challenges for climate science and policy making are (i) achieving a proper understanding of the sources and the extent of scientific uncertainties regarding future climate outcomes and (ii) adopting appropriate decision and policy frameworks in light of them. In part, uncertainties are associated with the existence of a diversity of models climate scientists use to predict climate outcomes: there is no single predictively best climate model. At the same time, appealing to a diverse set of models is often assumed to mitigate existing uncertainties and, where uncertainties cannot be reduced, provide the input needed for broadly precautionary decision procedures. This symposium brings together a group of philosophers and a renowned climate scientist to address a set of questions concerning model diversity, uncertainty, and climate policy. What is the nature of climate model diversity and what does this tell us about how confident we can be in predictions on which different climate models agree? What can we infer from the fact that as climate models ostensibly improve their range of predictions does not seem to get narrower? Finally, given that some uncertainty is likely to remain, what strategies are available for accommodating uncertainties in climate policy decisions?

03 Nov 2018 03:45 PM - 05:45 PM(America/Los_Angeles)
Venue : University (Fourth Floor Union Street Tower)
20181103T1545 20181103T1745 America/Los_Angeles Diversity, Uncertainty, and Action: Coping with a Plurality of Climate Models

Two difficult challenges for climate science and policy making are (i) achieving a proper understanding of the sources and the extent of scientific uncertainties regarding future climate outcomes and (ii) adopting appropriate decision and policy frameworks in light of them. In part, uncertainties are associated with the existence of a diversity of models climate scientists use to predict climate outcomes: there is no single predictively best climate model. At the same time, appealing to a diverse set of models is often assumed to mitigate existing uncertainties and, where uncertainties cannot be reduced, provide the input needed for broadly precautionary decision procedures. This symposium brings together a group of philosophers and a renowned climate scientist to address a set of questions concerning model diversity, uncertainty, and climate policy. What is the nature of climate model diversity and what does this tell us about how confident we can be in predictions on which different climate models agree? What can we infer from the fact that as climate models ostensibly improve their range of predictions does not seem to get narrower? Finally, given that some uncertainty is likely to remain, what strategies are available for accommodating uncertainties in climate policy decisions?

University (Fourth Floor Union Street Tower) PSA2018: The 26th Biennial Meeting of the Philosophy of Science Association office@philsci.org

Presentations

Uncertainties, Values, and Climate Targets

Philosophy of Science 03:45 PM - 04:15 PM (America/Los_Angeles) 2018/11/03 22:45:00 UTC - 2018/11/03 23:15:00 UTC
Mathias Frisch (Leibniz Universität Hannover)
I analyze a recent disagreement concerning the argument from inductive risk by asking how we should structure decision-making processes in the face of uncertainties in climate forecasts. I distinguish different hedging strategies (aimed at avoiding endorsing uncertain hypotheses) and different decision strategies with which they can be paired. If climate predictions are deeply uncertain, then the argument from inductive risk might be avoided by adopting a precautionary approach. Yet this approach does not provide the fine-grained predictions required for climate policy targets. Deciding on a climate policy requires that climate scientists 'stick their necks out' more than hedging strategies allow.
Presenters
MF
Mathias Frisch
Leibniz University Hannover

Climate Models: Still Uncertain, yet Improved

Philosophy of Science 04:15 PM - 04:45 PM (America/Los_Angeles) 2018/11/03 23:15:00 UTC - 2018/11/03 23:45:00 UTC
Julie Jebeile (Université Catholique de Louvain), Anouk Barberousse (Université Paris-Sorbonne)
Computer power and our understanding of climate processes continue to improve, and are expected to reduce the uncertainty affecting model projections. One way to assess such a reduction is to estimate the model spread, which measures the range of estimates from multiple models. A decrease in uncertainty is expected to result in a decrease in model spread. However the models considered in successive reports by the IPCC do not exhibit any reduction in model spread. Does this mean that there has been no model improvement resulting in a decrease in uncertainty? Our aim is to explain this seemingly paradoxical situation.
Presenters
JJ
Julie Jebeile
Université Catholique De Louvain
Co-Authors
AB
Anouk Barberousse
Sorbonne Université

Rebooting Robustness in Climate Science

Philosophy of Science 04:45 PM - 05:15 PM (America/Los_Angeles) 2018/11/03 23:45:00 UTC - 2018/11/04 00:15:00 UTC
Eric Winsberg (University of South Florida)
I explore a novel account of robustness according to which multiple lines of evidence for a hypothesis add cumulative support for the hypothesis whenever each one rules out a plausible alternative explanation. I show that we can make sense of a number of claims that multiple lines of evidence support climate hypotheses. First, I show how multiple lines of evidence, coming both from simulation outputs as well as from real world data, have worked to constrain our best estimates of equilibrium climate sensitivity. Second, I show that explanatory robustness provides the best account of the epistemic value of model agreement.
Presenters
EW
Eric Winsberg
University Of South Florida

Qualitative Differences in Attempts to Account for Climate Model Dependence

Philosophy of Science 05:15 PM - 05:45 PM (America/Los_Angeles) 2018/11/04 00:15:00 UTC - 2018/11/04 00:45:00 UTC
Gab Abramowitz (UNSW Sydney)
Most climate projection studies utilise simulations from a range of different climate models, nominally to obtain a collection of independent projection estimates. Yet the community has no agreed metrics for quantifying climate model independence and explicit attempts to address dependence are extremely rare. This presentation will give a qualitative overview of climate model dependence and contextualise existing approaches to addressing it within a single overarching framework.
Presenters
GA
Gab Abramowitz
UNSW Sydney
749 visits

Session Participants

Online
Session speakers, moderators & attendees
University of South Florida
UNSW Sydney
Université catholique de Louvain
Leibniz University Hannover
Michigan State University
No attendee has checked-in to this session!
39 attendees saved this session

Session Chat

Live Chat
Chat with participants attending this session

Questions & Answers

Answered
Submit questions for the presenters

Session Polls

Active
Participate in live polls

Need Help?

Technical Issues?

If you're experiencing playback problems, try adjusting the quality or refreshing the page.

Questions for Speakers?

Use the Q&A tab to submit questions that may be addressed in follow-up sessions.