Seeing further: the story of science & the Royal Society — 19 STEPHEN H. SCHNEIDER

1. 1. Climate models can project decades ahead because large-scale climate is not chaotic on timescales longer than a decade, unlike weather.
2. 2. Delaying emissions reductions is a mistake because thermal inertia in oceans means temperature response lags emissions by decades, and late-century outcomes diverge sharply based on near-term choices.
3. 3. The IPCC estimates climate sensitivity is likely between 2 and 4.5°C, with a best guess of 3°C, but there is a 10% chance it exceeds 6.8°C.
4. 4. Scientists face a double ethical bind: they must communicate effectively in sound-bites while also providing full caveats in longer formats.
5. 5. Climate prediction is inherently Bayesian because we cannot experiment on the future; we update prior probabilities with new data and theory.
6. 6. Sea level rise of several metres in coming centuries has about a 50% chance based on Bayesian reasoning from paleoclimate data, though confidence is not high.
7. 7. Climate science must shift from overcoming uncertainty to managing it, treating the problem as risk management with value judgments.
8. 8. The IPCC developed a quantitative confidence scale: low confidence (<33%), medium (33-66%), high (>66%), very high (>95%), and very low (<5%).
9. 9. Industry-funded groups recruited non-climate scientists to spread doubt, similar to tobacco industry tactics, framing climate as 'unproved' and 'too uncertain for policy'.
10. 10. The science that warming is occurring and human-caused is settled; the uncertainty is about severity, which varies by a factor of six by 2100.
11. 11. Schneider argues it is already decades too late to start mitigation; strong action is overdue even if there is a small chance climate sensitivity is low.