Inverse modelling of cloud-aerosol interactions – Part 2: Sensitivity tests on liquid phase clouds using a Markov chain Monte Carlo based simulation approach

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.

[1]  M. Andreae,et al.  Size Matters More Than Chemistry for Cloud-Nucleating Ability of Aerosol Particles , 2006, Science.

[2]  S. Sorooshian,et al.  Effective and efficient global optimization for conceptual rainfall‐runoff models , 1992 .

[3]  R. Dingenen,et al.  Size distribution and chemical composition of marine aerosols: a compilation and review , 2000 .

[4]  J. Seinfeld,et al.  Can chemical effects on cloud droplet number rival the first indirect effect? , 2002 .

[5]  P. Quinn,et al.  Atmospheric Chemistry and Physics Discussions Interactive comment on “ Influence of particle size and chemistry on the cloud nucleating properties of aerosols ” , 2007 .

[6]  V. Kerminen,et al.  On the contribution of Aitken mode particles to cloud droplet populations at continental background areas – a parametric sensitivity study , 2007 .

[7]  J. Vrugt,et al.  UvA-DARE ( Digital Academic Repository ) Inverse modeling of cloud-aerosol interactions-- Part 1 : Detailed response surface analysis , 2011 .

[8]  Mrinal K. Sen,et al.  Computational methods for parameter estimation in climate models , 2008 .

[9]  Pasi Aalto,et al.  Aerosol characteristics of air masses in northern Europe: Influences of location, transport, sinks, and sources , 2005 .

[10]  H. Hakola,et al.  Using Aerosol Number to Volu me Ratio in Predicting Cloud Droplet Number Concentration , 2007 .

[11]  Bayesian solution for nonlinear and non‐Gaussian inverse problems by Markov chain Monte Carlo method , 2001 .

[12]  J. W. Fitzgerald.  Effect of Aerosol Composition on Cloud Droplet Size Distribution: A Numerical Study , 1974 .

[13]  Masato Koda,et al.  Estimation of urban air pollution , 1978, Autom..

[14]  M. Zahniser,et al.  Implementation of a Markov Chain Monte Carlo method to inorganic aerosol modeling of observations from the MCMA-2003 campaign – Part I: Model description and application to the La Merced site , 2006 .

[15]  N. Metropolis,et al.  Equation of State Calculations by Fast Computing Machines , 1953, Resonance.

[16]  B. Stevens,et al.  Untangling aerosol effects on clouds and precipitation in a buffered system , 2009, Nature.

[17]  C. Twohy,et al.  Droplet nuclei in non-precipitating clouds: composition and size matter , 2008 .

[18]  Jasper A. Vrugt,et al.  Significant variation in vegetation characteristics and dynamics from ecohydrological optimality of net carbon profit , 2009 .

[19]  Arto Voutilainen,et al.  Sequential Monte Carlo estimation of aerosol size distributions , 2004, Comput. Stat. Data Anal..

[20]  D. Higdon,et al.  Accelerating Markov Chain Monte Carlo Simulation by Differential Evolution with Self-Adaptive Randomized Subspace Sampling , 2009 .

[21]  Adrian F. M. Smith,et al.  Sampling-Based Approaches to Calculating Marginal Densities , 1990 .

[22]  K. Lehtinen,et al.  MECCO: A method to estimate concentrations of condensing organics—Description and evaluation of a Markov chain Monte Carlo application , 2010 .

[23]  Carlos J. Perez,et al.  Sensitivity estimations for Bayesian inference models solved by MCMC methods , 2006, Reliab. Eng. Syst. Saf..

[24]  R. Wood,et al.  Observational Strategies from the Micro- to Mesoscale , 2009 .

[25]  J. Vrugt,et al.  Inverse Modeling of Subsurface Flow and Transport Properties: A Review with New Developments , 2008 .

[26]  R. Charlson,et al.  Modification of the Köhler Equation to Include Soluble Trace Gases and Slightly Soluble Substances , 1998 .

[27]  Erkki Oja,et al.  Estimation of ECHAM5 climate model closure parameters with adaptive MCMC , 2010 .

[28]  J. A. Vrugt,et al.  Towards inverse modeling of cloud-aerosol interactions - Part 1: A detailed response surface analysis , 2011 .

[29]  Maria Cristina Facchini,et al.  The effect of physical and chemical aerosol properties on warm cloud droplet activation , 2005 .

[30]  P. Rich,et al.  Soil water dynamics under low‐ versus high‐ponderosa pine tree density: ecohydrological functioning and restoration implications , 2008 .

[31]  John H. Seinfeld,et al.  Constraining the contribution of organic acids and AMS m/z 44 to the organic aerosol budget: On the importance of meteorology, aerosol hygroscopicity, and region , 2010 .

[32]  Meinrat O. Andreae,et al.  Aerosol cloud precipitation interactions. Part 1. The nature and sources of cloud-active aerosols , 2008 .

[33]  S. Twomey,et al.  Determining the Susceptibility of Cloud Albedo to Changes in Droplet Concentration with the Advanced Very High Resolution Radiometer , 1994 .

[34]  S. Sorooshian,et al.  A Shuffled Complex Evolution Metropolis algorithm for optimization and uncertainty assessment of hydrologic model parameters , 2002 .

[35]  G. Feingold.  Modeling of the first indirect effect: Analysis of measurement requirements , 2003 .

[36]  J. Seinfeld,et al.  Parameterization of cloud droplet size distributions: comparison with parcel models and observations , 2009 .

[37]  Cajo J. F. ter Braak,et al.  Treatment of input uncertainty in hydrologic modeling: Doing hydrology backward with Markov chain Monte Carlo simulation , 2008 .

[38]  J van der Kruk,et al.  Inversion and sensitivity analysis of GPR data with waveguide dispersion using Markov Chain Monte Carlo simulation , 2010, Proceedings of the XIII Internarional Conference on Ground Penetrating Radar.

[39]  Erik Swietlicki,et al.  Organic aerosol and global climate modelling: a review , 2004 .

[40]  K. Mengersen,et al.  Bayesian mixture model estimation of aerosol particle size distributions , 2011 .

[41]  J. Vrugt,et al.  A formal likelihood function for parameter and predictive inference of hydrologic models with correlated, heteroscedastic, and non‐Gaussian errors , 2010 .

[42]  M. J. Best,et al.  Trade‐offs and responsiveness of the single‐layer urban canopy parametrization in WRF: An offline evaluation using the MOSCEM optimization algorithm and field observations , 2010 .

[43]  A. Nenes,et al.  Chemical and dynamical effects on cloud droplet number: Implications for estimates of the aerosol indirect effect: CHEMISTRY, DYNAMICS, AND DROPLET NUMBER , 2004 .

[44]  G. Roelofs,et al.  A model study of the influence of aerosol size and chemical properties on precipitation formation in warm clouds , 2004 .

[45]  P. Chuang.  Sensitivity of cloud condensation nuclei activation processes to kinetic parameters , 2006 .

[46]  M. Facchini,et al.  Cloud albedo enhancement by surface-active organic solutes in growing droplets , 1999, Nature.

[47]  Marko Laine,et al.  Aerosol model selection and uncertainty modelling by adaptive MCMC technique , 2008 .

[48]  T. Larson,et al.  The effects of microphysical parameterization on model predictions of sulfate production in clouds , 1990 .

[49]  J. Hudson.  Variability of the relationship between particle size and cloud‐nucleating ability , 2007 .

[50]  Martyn P. Clark,et al.  Multi‐objective calibration of forecast ensembles using Bayesian model averaging , 2006 .

[51]  Mark E. Borsuk,et al.  Robust Bayesian uncertainty analysis of climate system properties using Markov chain Monte Carlo methods , 2007 .

[52]  Sonia M. Kreidenweis,et al.  Influence of water‐soluble organic carbon on cloud drop number concentration , 2005 .

[53]  J. Brenguier,et al.  Cloud condensation nuclei and cloud droplet measurements during ACE-2 , 2000 .

[54]  Mrinal K. Sen,et al.  An Efficient Stochastic Bayesian Approach to Optimal Parameter and Uncertainty Estimation for Climate Model Predictions , 2004 .

[55]  Alfred Wiedensohler,et al.  Atmospheric particle number size distribution in central Europe: Statistical relations to air masses and meteorology , 2001 .

[56]  J. Seinfeld,et al.  Aerosol–cloud drop concentration closure in warm cumulus , 2004 .

[57]  G. Feingold,et al.  A modeling study of the effect of nitric acid on cloud properties , 2004 .

[58]  Bruce A. Robinson,et al.  Self-Adaptive Multimethod Search for Global Optimization in Real-Parameter Spaces , 2009, IEEE Transactions on Evolutionary Computation.

[59]  J. Seinfeld,et al.  Chemical Amplification (or Dampening) of the Twomey Effect: Conditions Derived from Droplet Activation Theory , 2004 .

[60]  V. Tyagi,et al.  Microbial Surfactants: A Review , 2006 .

[61]  D. Rubin,et al.  Inference from Iterative Simulation Using Multiple Sequences , 1992 .

[62]  T. Larson,et al.  Numerical study of droplet size dependent chemistry in oceanic, wintertime stratus cloud at southern mid-latitudes , 1990 .

[63]  Kurt K. Benke,et al.  Parameter uncertainty, sensitivity analysis and prediction error in a water-balance hydrological model , 2008, Math. Comput. Model..

[64]  Ghassan Chebbo,et al.  Application of MCMC-GSA model calibration method to urban runoff quality modeling , 2006, Reliab. Eng. Syst. Saf..