Atmospheric Chemistry and Physics Discussions www.atmos-chem-phys-discuss.net 1680-7367 1680-7375 6 4 2006 10.5194/acpd-6-5933-2006 http://www.atmos-chem-phys-discuss.net/6/5933/2006/ http://www.atmos-chem-phys-discuss.net/6/5933/2006/acpd-6-5933-2006.html http://www.atmos-chem-phys-discuss.net/6/5933/2006/acpd-6-5933-2006.pdf 5933 5998 2006-07-10 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 F. M. San Martini E. J. Dunlea M. Grutter T. B. Onasch J. T. Jayne M. R. Canagaratna D. R. Worsnop C. E. Kolb J. H. Shorter S. C. Herndon M. S. Zahniser J. M. Ortega G. J. McRae L. T. Molina M. J. Molina Department of Earth, Atmospheric, and Planetary Sciences, Massachusetts Institute of Technology, Cambridge, MA, USA Cooperative Institute for Research in the Environmental Sciences (CIRES), University of Colorado at Boulder, Boulder, CO, USA Centro de Ciencias de la Atmósfera, Universidad Nacional Autónoma de México, Mexico City, Mexico Aerodyne Research Inc., Billerica, MA, USA Department of Chemical Engineering, Massachusetts Institute of Technology, Cambridge, MA, USA Molina Center on Energy and the Environment, La Jolla, CA, USA now at: the Board on Chemical Sciences and Technology, National Academies, Washington D.C., USA now at: Sandia National Laboratory, Livermore, CA, USA now at: University of California San Diego, La Jolla, CA, USA The equilibrium inorganic aerosol model ISORROPIA was embedded in a Markov Chain Monte Carlo algorithm to produce a powerful tool to analyze aerosol data and predict gas phase concentrations where these are unavailable. The method directly incorporates measurement uncertainty, prior knowledge, and provides for a formal framework to combine measurements of different quality. The method was applied to aerosol- and gas-phase precursor observations taken at La Merced during the Mexico City Metropolitan Area (MCMA) 2003 Field Campaign and served to discriminate between diverging gas-phase observations of ammonia. The model reproduced observations of particle-phase ammonium, nitrate, and sulfate well. The most likely concentrations of ammonia were found to vary between 4 and 26 ppbv, while the range for nitric acid was 0.1 to 55 ppbv. During periods where the aerosol chloride observations were consistently above the detection limit, the model was able to reproduce the aerosol chloride observations well and predicted the most likely gas-phase hydrochloric acid concentration varied between 0.4 and 5 ppbv. Despite the high ammonia concentrations observed and predicted by the model, when the aerosols were assumed to be in the efflorescence branch they are predicted to be acidic (pH~3).