The Denver Brown Cloud studies from the perspective of model assessment needs and the role of meteorology

The 1987-1988 Denver Brown Cloud Study addressed Denver's severe winter brown cloud episodes in a three-month field study followed by the application of a receptor-oriented chemical mass balance modeling method to apportion light extinction to primary emissions. However, this modeling method did not apportion light extinction from secondary particulates to individual sources. Recently, an independent meteorological and chemical modeling study, driven by the need to apportion secondary particles, reported on the initial numerical simulation of visibility for two days during the winter of 1987-1988. The MM4 meteorological model used in this recent study was operated in a four-dimensional, data-assimilation mode using standard, low-resolution synoptic observations and analyses. The chemical and aerosol numerical modeling system, which uses the MM4 meteorological model numerical output, was derived from the family of models developed for the National Acid Precipitation Assessment Program. This paper summarizes the meteorological features underlying the Denver brown cloud and assesses the consistency of the reported numerical simulations with the detailed meteorological observations available for the same periods. We found that the meteorological model, in its current form and range of application, does not address some of the critical mesoscale structures responsible for the initiation and maintenance of the brown cloud episodes studied nor, under some circumstances, does it properly predict boundary layer properties compared with observations. In addition, cloud systems and surface moisture fluxes are not addressed by the model, although many chemical processes depend on the presence of additional moisture. Comparison between model results and observations suggest that moist processes, particularly during the initiation phase of episodes, should be a critical issue for future experimental and modeling exercises. In addition, the limited model results available, when evaluated against the detailed boundary layer observations of the Brown Cloud Study, suggest that model uncertainties in the treatment of local terrain influence, boundary layer depth, and humidity may have obscured model overestimates of optical extinction. Finally, we suggest that although models will remain limited in many aspects, their use in combination with detailed observations of meteorology and chemistry will prove a valuable tool in the assessment of urban visibility issues.