Inverse modelling of cloud-aerosol interactions – Part 2: Sensitivity tests on liquid phase clouds using a Markov chain Monte Carlo based simulation approach
Inverse modelling of cloud-aerosol interactions – Part 2: Sensitivity tests on liquid phase clouds using a Markov chain Monte Carlo based simulation approach
复制标题
DOI：
10.5194/acp-12-2823-2012
复制
发表时间：
2011-07
影响因子：
6.3
通讯作者：
D. Partridge;J. Vrugt;P. Tunved;A. Ekman;H. Struthers;A. Sorooshian
中科院分区：
文献类型：
--
作者：
D. Partridge;J. Vrugt;P. Tunved;A. Ekman;H. Struthers;A. Sorooshian
本文提出了一种将马尔可夫链蒙特卡罗（MCMC）算法与绝热云块模式相结合来研究云-气溶胶相互作用的新方法。尽管以前进行了大量的数值云-气溶胶敏感性研究，但很少有人使用统计分析工具来研究云模式对输入气溶胶物理化学参数的全球敏感性。使用数值生成的云滴数浓度（CDNC）分布（即合成数据）作为云观测，这种逆建模框架被证明可以成功地估计正确的校准参数，以及它们的后验概率分布。所采用的分析方法提供了一个新的，综合的框架来评估的全球敏感性的推导出的CDNC分布的输入参数描述的对数正态特性的积累模式的气溶胶和粒子化学。在很大程度上，先前研究的结果得到了证实，但本研究也提供了一些额外的见解。相对敏感性从非常清洁的海洋北极条件到非常污染的大陆环境有一个过渡，在这种条件下，对数正态气溶胶参数代表累积模式气溶胶数浓度和平均半径，对确定CDNC分布最为重要（聚集模式中的气溶胶浓度>1000 cm 3），其中颗粒化学比聚集模式的数量浓度和尺寸更重要。云模式输入参数之间的竞争和补偿表明，如果可溶质量分数减小，则气溶胶数浓度、几何标准差和累积模态的平均半径必须增大，才能达到相同的CDNC分布。这项研究表明，逆模拟提供了一个灵活的，透明的和综合的方法，有效地探索云-气溶胶相互作用的参数敏感性和相关性。
This paper presents a novel approach to investi- gate cloud-aerosol interactions by coupling a Markov chain Monte Carlo (MCMC) algorithm to an adiabatic cloud parcel model. Despite the number of numerical cloud-aerosol sen- sitivity studies previously conducted few have used statistical analysis tools to investigate the global sensitivity of a cloud model to input aerosol physiochemical parameters. Using numerically generated cloud droplet number concentration (CDNC) distributions (i.e. synthetic data) as cloud observa- tions, this inverse modelling framework is shown to success- fully estimate the correct calibration parameters, and their underlying posterior probability distribution. The employed analysis method provides a new, integrative framework to evaluate the global sensitivity of the derived CDNC distribution to the input parameters describing the lognormal properties of the accumulation mode aerosol and the particle chemistry. To a large extent, results from prior studies are confirmed, but the present study also provides some additional insights. There is a transition in relative sen- sitivity from very clean marine Arctic conditions where the lognormal aerosol parameters representing the accumulation mode aerosol number concentration and mean radius and are found to be most important for determining the CDNC dis- tribution to very polluted continental environments (aerosol concentration in the accumulation mode>1000 cm 3 ) where particle chemistry is more important than both number con- centration and size of the accumulation mode. The competition and compensation between the cloud model input parameters illustrates that if the soluble mass fraction is reduced, the aerosol number concentration, geo- metric standard deviation and mean radius of the accumula- tion mode must increase in order to achieve the same CDNC distribution. This study demonstrates that inverse modelling provides a flexible, transparent and integrative method for efficiently exploring cloud-aerosol interactions with respect to parame- ter sensitivity and correlation.