Youngseob Kim

Evaluation of the Weather Research and Forecast/Urban Model Over Greater Paris

Meteorological modelling in the planetary boundary layer (PBL) over Greater Paris is performed using the Weather Research and Forecast (WRF) numerical model. The simulated meteorological fields are evaluated by comparison with mean diurnal observational data or mean vertical profiles of temperature, wind speed, humidity and boundary-layer height from 6 to 27 May 2005. Different PBL schemes, which parametrize the atmospheric turbulence in the PBL using different turbulence closure schemes, may be used in the WRF model. The sensitivity of the results to four PBL schemes (two non-local closure schemes and two local closure schemes) is estimated. Uncertainties in the PBL schemes are compared to the influence of the urban canopy model (UCM) and the updated Coordination of Information on the Environment (CORINE) land-use data. Using the UCM and the CORINE land-use data produces more realistic modelled meteorological fields. The wind speed, which is overestimated in the simulations without the UCM, is improved below 1,000 m height. Furthermore, the modelled PBL heights during nighttime are strongly modified, with an increase that may be as high as 200 %. At night, the impact of changing the PBL scheme is lower than the impact of using the UCM and the CORINE land-use data.

Comparison of different gas-phase mechanisms and aerosol modules for simulating particulate matter formation.

ABSTRACT The effects of two gas-phase chemical kinetic mechanisms, Regional Atmospheric Chemistry Mechanism version 2 (RACM2) and Carbon-Bond 05 (CB05), and two secondary organic aerosol (SOA) modules, the Secondary Organic Aerosoi Model (SORGAM) and AER/EPRI/Caltech model (AEC), on fine (aerodynamic diameter ≤2.5 μm) particulate matter (PM2.5) formation is studied. The major sources of uncertainty in the chemistry of SOA formation are investigated. The use of all major SOA precursors and the treatment of SOA oligomerization are found to be the most important factors for SOA formation, leading to 66% and 60% more SOA, respectively. The explicit representation of high-NOx and low-NOx gas-phase chemical regimes is also important with increases in SOA of 30–120% depending on the approach used to implement the distinct SOA yields within the gas-phase chemical kinetic mechanism; further work is needed to develop gas-phase mechanisms that are fully compatible with SOA formation algorithms. The treatment of isoprene SOA as hydrophobic or hydrophilic leads to a significant difference, with more SOA being formed in the latter case. The activity coefficients may also be a major source of uncertainty, as they may differ significantly between atmospheric particles, which contain a myriad of SOA, primary organic aerosol (POA), and inorganic aerosol species, and particles formed in a smog chamber from a single precursor under dry conditions. Significant interactions exist between the uncertainties of the gas-phase chemistry and those of the SOA module. IMPLICATIONS The current state of the science is more advanced for the gas-phase chemistry of ozone formation than for the chemistry and gas/particle partitioning of particulate matter (PM) formation. As a result, there are larger uncertainties associated with aerosol modules than with gas-phase chemical kinetic mechanisms. Nevertheless, the uncertainties associated with those modules are not additive in an air quality model and there are close interactions between the gas-phase chemical mechanism and the secondary aerosol formation. In particular, the effect of the NOx regime on SOA formation should be explicitly treated in air quality models.

Precursors and formation of secondary organic aerosols fromwildfires in the Euro-Mediterranean region

This work aims at quantifying the relative contribution of secondary organic aerosol (SOA) precursors emitted by wildfires to organic aerosol (OA) formation, during summer 2007 over the Euro-Mediterranean region, where intense wildfires occurred. A new SOA formation mechanism, HOaro, including recently identified aromatic volatile organic compounds (VOCs) emitted from wildfires is developed based on smog chamber experiment measurements, under low and high-NOx regimes. The aromatic VOCs included in the mechanism are toluene, xylene, benzene, phenol, cresol, catechol, furan, naph5 thalene, methylnaphthalene, syringol, guaiacol and structurally assigned and unassigned compounds with at least 6 carbon atoms per molecule (USC>6). This mechanism HOaro is an extension of the HO (Hydrophilic/Hydrophobic organic) aerosol mechanism: the oxidation of the precursor forms surrogate species with specific thermodynamic properties (volatility, oxidation degree, affinity to water). The SOA concentrations over the Euro-Mediterranean region in summer 2007 are simulated using the chemistry transport model (CTM) Polair3D of the air-quality plateform Polyphemus, where the mechanism HOaro 10 was implemented. To estimate the relative contribution of the aromatic VOCs, intermediate, semi and low volatile organic compounds (I/S/L-VOCs) to wildfires OA concentrations, different estimations of the gaseous I/S/L-VOC emissions (from primary organic aerosol (POA) using a factor of 1.5 or from non-methanic organic gas (NMOG) using a factor of 0.36) and their ageing (one-step oxidation vs multi-generational oxidation), are also tested in the CTM. Most of the particle organic aerosol (OA) concentrations are formed from I/S/L-VOCs. In average during the summer 2007 15 and over the Euro-Mediterranean domain, they are about 10 times higher than the OA concentrations formed from VOCs. However, locally, the OA concentrations formed from VOCs can represent up to 30% of the OA concentrations from biomass burning. Amongst the VOCs, the main contributors to SOA formation are phenol, benzene and catechol (47%), USC>6 compounds (23%), and toluene and xylene (12%). Sensitivity studies of the influence of the VOCs and the I/S/L-VOCs emissions and chemical ageing mechanisms on PM2.5 concentrations show that surface PM2.5 concentrations are more sensitive to the 20 parameterization used for gaseous I/S/L-VOCs emissions than for ageing. Estimating the gaseous I/S/L-VOCs emissions from POA or from NMOG has a high impact on local surface PM2.5 concentrations (reaching -30% in Balkans, -8 to -16% in the fire plume and +8 to +16% in Greece). Considering the VOC emissions results in a moderate increase of PM2.5 concentrations 1 Atmos. Chem. Phys. Discuss., https://doi.org/10.5194/acp-2018-1065 Manuscript under review for journal Atmos. Chem. Phys. Discussion started: 22 October 2018 c

Impact of wildfires on particulate matter in the Euro-Mediterranean in 2007: sensitivity to the parameterization of emissions in air quality models

This study examines the uncertainties on air quality modeling associated with the integration of wildfire emissions in chemistry-transport models (CTMs). To do so, aerosol concentrations during the summer of 2007, which was marked by severe fire episodes in the EuroMediterranean region especially in the Balkans (20–31 July, 24–30 August 2007) and Greece (24–30 August 2007), are analyzed. Through comparisons to observations from surface networks and satellite remote sensing, we evaluate the abilities of two CTMs, Polyphemus/Polair3D and CHIMERE, to simulate the impact of fires on the regional particulate matter (PM) concentrations and optical properties. During the two main fire events, fire emissions may contribute up to 90 % of surface PM2.5 concentrations in the fire regions (Balkans and Greece), with a significant regional impact associated with long-range transport. Good general performances of the models and a clear improvement of PM2.5 and aerosol optical depth (AOD) are shown when fires are taken into account in the models with high correlation coefficients. Two sources of uncertainties are specifically analyzed in terms of surface PM2.5 concentrations and AOD using sensitivity simulations: secondary organic aerosol (SOA) formation from intermediate and semi-volatile organic compounds (I/S-VOCs) and emissions’ injection heights. The analysis highlights that surface PM2.5 concentrations are highly sensitive to injection heights (with a sensitivity that can be as high as 50 % compared to the sensitivity to I/S-VOC emissions which is lower than 30 %). However, AOD which is vertically integrated is less sensitive to the injection heights (mostly below 20 %) but highly sensitive to I/S-VOC emissions (with sensitivity that can be as high as 40 %). The maximum statistical dispersion, which quantifies uncertainties related to fire emission modeling, is up to 75 % for PM2.5 in the Balkans and Greece, and varies between 36 % and 45 % for AOD above fire regions. The simulated number of daily exceedance of World Health Organization (WHO) recommendations for PM2.5 over the considered region reaches 30 days in regions affected by fires and ∼ 10 days in fire plumes, which is slightly underestimated compared to available observations. The maximum statistical dispersion (σ ) on this indicator is also large (with σ reaching 15 days), showing the need for better understanding of the transport and evolution of fire plumes in addition to fire emissions.

Sensitivity Analysis of Ambient Particulate Matter to Industrial Emissions Using a Plume-in-Grid Approach: Application in the Greater Paris Region

The Polyphemus Plume-in-Grid (PinG) model, which is based on a 3D Eulerian model and an imbedded puff model, was developed to represent the dispersion and transformation of air pollutants in industrial plumes. It was later improved to take into account particulate matter (PM) formation and transport in order to evaluate secondary PM formation in refinery plumes. The performance of the PinG model, applied to a refinery in the Greater Paris region, was previously evaluated at the regional scale for July 2009, showing satisfactory results for O3 and PM. The PinG model is applied here to the same refinery for a different period, April 2013, when local measurements were available. The refinery is located close to a large NH3 source, which is also treated here using the puff model in order to evaluate the interactions of the plumes of these two industrial sites. Modeled PM is compared here to local measurements in terms of mass concentrations and chemical composition. The measurement sites are located around the refinery and are impacted by the plumes of the two industrial sites. The results show good agreement between measured and modeled PM chemical composition. The sensitivity of the local concentrations to the refinery emissions is evaluated. It is mostly due to primary and secondary inorganic aerosols, emitted and formed in the plumes, and to secondary organic aerosols (SOA) formed from the refinery VOC fugitive emissions.