Mesoscale modeling of smoke transport from equatorial Southeast Asian Maritime Continent to the Philippines: First comparison of ensemble analysis with in situ observations

Abstract

Atmospheric transport of smoke from equatorial Southeast Asian Maritime Continent (Indonesia, Singapore, and Malaysia) to the Philippines was recently verified by the first‐ever measurement of aerosol composition in the region of the Sulu Sea from a research vessel named Vasco. However, numerical modeling of such transport can have large uncertainties due to the lack of observations for parameterization schemes and for describing fire emission and meteorology in this region. These uncertainties are analyzed here, for the first time, with an ensemble of 24 Weather Research and Forecasting model with Chemistry (WRF‐Chem) simulations. The ensemble reproduces the time series of observed surface nonsea‐salt PM2.5 concentrations observed from the Vasco vessel during 17–30 September 2011 and overall agrees with satellite (Cloud‐Aerosol Lidar and Infrared Pathfinder Satellite Observation (CALIPSO) and Moderate Resolution Imaging Spectroradiometer (MODIS)) and Aerosol Robotic Network (AERONET) data. The difference of meteorology between National Centers for Environmental Prediction (NCEP's) Final (FNL) and European Center for Medium range Weather Forecasting (ECMWF's) ERA renders the biggest spread in the ensemble (up to 20 μg m−3 or 200% in surface PM2.5), with FNL showing systematically superior results. The second biggest uncertainty is from fire emissions; the 2 day maximum Fire Locating and Modelling of Burning Emissions (FLAMBE) emission is superior than the instantaneous one. While Grell‐Devenyi (G3) and Betts‐Miller‐Janjić cumulus schemes only produce a difference of 3 μg m−3 of surface PM2.5 over the Sulu Sea, the ensemble mean agrees best with Climate Prediction Center (CPC) MORPHing (CMORPH)'s spatial distribution of precipitation. Simulation with FNL‐G3, 2 day maximum FLAMBE, and 800 m injection height outperforms other ensemble members. Finally, the global transport model (Navy Aerosol Analysis and Prediction System (NAAPS)) outperforms all WRF‐Chem simulations in describing smoke transport on 20 September 2011, suggesting the challenges to model tropical meteorology at mesoscale and finer scale.

The Uncharacteristic Occurrence of the June 2013 Biomass-Burning Haze Event in Southeast Asia: Effects of the Madden-Julian Oscillation and Tropical Cyclone Activity

“…Sensitivity WRF simulations (plots not shown here) were previously run with three cumulus parameterization schemes, namely, Betts-Miller-Janjic (BMJ) [54,55], Kain-Fritsch (KF) [56] and Grell 3D (G3D) cumulus scheme [57] over the model domain for the same study period. In summary, the BMJ scheme showed the best agreement to CMORPH data, and hence, all WRF-Chem simulations for this episode are run with BMJ, which is consistent with a recent study by Ge et al [58] for the case of September 2011 in the Maritime Continent. Pennelly et al [59] also report that BMJ simulates precipitation with better accuracy than KF and G3D at 30 km grid resolution.…”

Biomass Burning Plumes in the Vicinity of the California Coast: Airborne Characterization of Physicochemical Properties, Heating Rates, and Spatiotemporal Features

“…As transport and vertical distributions of BB smoke are shown to be sensitive to smoke injection height in many aerosol modeling studies (Colarco et al, ; Ge et al, ; J. Wang et al, ), another NAAPS simulation, aiming for possible improvement of the vertical distribution of smoke for the NiCE and FASE cases, is tested using the observed average BB layer height as the smoke injection height (approximately 500 m based on the soundings). The default model smoke injection method is to inject smoke mass into the lowest four model layers, with the fourth layer centered around 300 m above ground.…”

Uncertainties Caused by Major Meteorological Analysis Data Sets in Simulating Air Quality Over India

“…P. Kumar et al () indicated that the model performance on meteorological fields initialized from the ECMWF data set was superior to that initialized from the NCEP data sets over the Indian region. The effect of different meteorological analysis data sets on volcanic ash dispersion (Mulena et al, ) and biomass burning smoke transport (Ge et al, ) has also been investigated. However, overall effect of the two data sets on simulated ambient pollutant concentrations in addition to meteorological fields over India for a whole year has not yet been studied.…”