Estimates of uncertainty in predictions of global mean surface temperature
Estimates of uncertainty in predictions of global mean surface temperature
复制标题
DOI：
10.1175/jcli4012.1
复制
发表时间：
2007-03-01
影响因子：
4.9
通讯作者：
Allen, M. R.
中科院分区：
文献类型：
--
作者：
Kettleborough, J. A.;Booth, B. B. B.;Allen, M. R.
详细讨论了一种估计未来气候变化不确定性的方法，并将其应用于全球平均气温变化的预测。该方法使用最佳指纹来估计20世纪变暖模型模拟的不确定性。然后，利用世纪变暖与21世纪世纪变暖之间的线性、紧凑关系，对这些估计值进行预测。通过改变能量平衡模式的参数，对20世纪全球平均温度预测中使用线性关系所产生的误差进行了估计。包括这个误差对预测的影响很小。对于排放情景特别报告（SRES）A1FI情景，1995年至2035年全球平均温度变化将大于1.5 K的可能性为50%。在SRES B2下，直到2055年才超过相同的阈值。对于给定的辐射强迫历史，这些结果对于模型的发展应该是相对稳健的。
A method for estimating uncertainty in future climate change is discussed in detail and applied to predictions of global mean temperature change. The method uses optimal fingerprinting to make estimates of uncertainty in model simulations of twentieth-century warming. These estimates are then projected forward in time using a linear, compact relationship between twentieth-century warming and twenty-first-century warming. This relationship is established from a large ensemble of energy balance models.By varying the energy balance model parameters an estimate is made of the error associated with using the linear relationship in forecasts of twentieth-century global mean temperature. Including this error has very little impact on the forecasts. There is a 50% chance that the global mean temperature change between 1995 and 2035 will be greater than 1.5 K for the Special Report on Emissions Scenarios (SRES) A1FI scenario. Under SRES B2 the same threshold is not exceeded until 2055. These results should be relatively robust to model developments for a given radiative forcing history.