Constraining Climate Sensitivity from the Seasonal Cycle in Surface Temperature

摘要: Abstract The estimated range of climate sensitivity has remained unchanged for decades, resulting in large uncertainties long-term projections future under increased greenhouse gas concentrations. Here the multi-thousand-member ensemble model simulations from climateprediction.net project and a neural network are used to establish relation between amplitude seasonal cycle regional temperature. Most models with high sensitivities found overestimate compared observations. A probability density function is then calculated present-day reanalysis instrumental datasets. Subject number assumptions on datasets used, it that very unlikely (5% probability) be either below 1.5–2 K or above about 5–6.5 K, best agreement 3 3.5 K. This narrower than most prob...