Time filter

Source Type

Boulogne-Billancourt, France

Guttikunda S.K.,Desert Research Institute | Calori G.,ARIA Technologies
Atmospheric Environment | Year: 2013

In Delhi, between 2008 and 2011, at seven monitoring stations, the daily average of particulates with diameter <2.5 μm (PM2.5) was 123 ± 87 μg m-3 and particulates with diameter <10 μm (PM10) was 208 ± 137 μg m-3. The bulk of the pollution is due to motorization, power generation, and construction activities. In this paper, we present a multi-pollutant emissions inventory for the National Capital Territory of Delhi, covering the main district and its satellite cities - Gurgaon, Noida, Faridabad, and Ghaziabad. For the base year 2010, we estimate emissions (to the nearest 000's) of 63,000 tons of PM2.5, 114,000 tons of PM10, 37,000 tons of sulfur dioxide, 376,000 tons of nitrogen oxides, 1.42 million tons of carbon monoxide, and 261,000 tons of volatile organic compounds. The inventory is further spatially disaggregated into 80 × 80 grids at 0.01° resolution for each of the contributing sectors, which include vehicle exhaust, road dust re-suspension, domestic cooking and heating, power plants, industries (including brick kilns), diesel generator sets and waste burning. The GIS based spatial inventory coupled with temporal resolution of 1 h, was utilized for chemical transport modeling using the ATMoS dispersion model. The modeled annual average PM2.5 concentrations were 122 ± 10 μg m-3 for South Delhi; 90 ± 20 μg m-3 for Gurgaon and Dwarka; 93 ± 26 μg m-3 for North-West Delhi; 93 ± 23 μg m-3 for North-East Delhi; 42 ± 10 μg m-3 for Greater Noida; 77 ± 11 μg m-3 for Faridabad industrial area. The results have been compared to measured ambient PM pollution to validate the emissions inventory. © 2012 Elsevier Ltd.

Liao G.-L.,Guangxi Meteorological Service Center | Zeng P.,Guangxi Meteorological Service Center | Zhang Q.-J.,ARIA Technologies | Cheng P.,Guangxi Weather Modification Office | Mo Y.-C.,Guangxi Meteorological Service Center
HARMO 2014 - 16th International Conference on Harmonisation within Atmospheric Dispersion Modelling for Regulatory Purposes, Proceedings | Year: 2014

Based on the data provided by surface automatic weather stations over Guangxi, Nanning conventional weather station, and the air pollution daily average concentrations in Nanning, a serious air pollution process in Nanning has been analyzed by using Empirical Orthogonal Function (EOF), air trajectory and other analytical methods. The EOF analysis shows that the second mode of the serious air pollution process has obvious diurnal variation. Influenced by local meso-scale and micro-scale circulation systems, pollution is localized which leads to the important air pollution level. © Crown Copyright 2014 Dstl.

Oldrini O.,MOKILI | Nibart M.,ARIA Technologies | Armand P.,CEA DAM Ile-de-France | Olry C.,ARIA Technologies | And 2 more authors.
Proceedings of the 15th International Conference on Harmonisation within Atmospheric Dispersion Modelling for Regulatory Purposes, HARMO 2013 | Year: 2013

Parallel SWIFT is used in the AIRCITY project to model the air quality at a few meters resolution over the whole Paris area. Hence, it is crucial to provide the capability to correctly handle the downscaling from meso scale modelling to local scale modelling over build-up areas. Build-up areas have directional specificities due to the orientation of streets, or also the river Seine for the Paris case, that lead to directional bulk effects. To overcome this, we have developed an approach based on directional drag coefficients. Drag coefficients of buildings within their neighbourhood are derived once and for all for a set of wind directions. A methodology has been implemented in Micro-SWIFT to derive this, but data obtained from other CFD models can be used. This static database of drag coefficients is then used to create directional matrices of Cionco type canopy densities, at the wished resolution. The capability to handle and create canopy laws according to the flow direction has been implemented in SWIFT. Results are presented on test cases.

Kaplan H.,IIBR | Olry C.,ARIA Technologies | Moussafir J.,ARIA Technologies | Oldrini O.,MOKILI | And 2 more authors.
International Journal of Environment and Pollution | Year: 2014

In the context of the FEDER AIRCITY project, Paris City is modelled with a 3 m resolution with the purpose of computing pollutant concentration relevant for human exposition. This project involves AIRPARIF (Ile-de-France Air Quality Monitoring Network), for emission and immission data, CEA (French Atomic and Alternative Energy Agency) for HPC availability, National Geographic Institute (IGN) for city 3D data. For pollutant dispersion modelling, PMSS software was selected including a parallelised fast 3D wind field and turbulence code and a parallelised stochastic LPDM. A key point in urban air quality is NO2 concentration, so the NO/NO2 transformation at the street scale is a key issue. In this paper, a preliminary independent study will focus on fundamental questions. After theoretical consideration, a practical case is carried on over the Paris Opera district. The results are compared with standard continuous measurements at two air quality stations of the AIRPARIF monitoring network. Copyright © 2014 Inderscience Enterprises Ltd.

Duchenne C.,CEA DAM Ile-de-France | Armand P.,CEA DAM Ile-de-France | Oldrini O.,MOKILI | Olry C.,ARIA Technologies | Moussafir J.,ARIA Technologies
HARMO 2011 - Proceedings of the 14th International Conference on Harmonisation within Atmospheric Dispersion Modelling for Regulatory Purposes | Year: 2011

The development of PMSS, the parallel version of the Micro-SWIFT-SPRAY suite, allows now to process 3D flow calculation, atmospheric dispersion of hazardous species and health impact assessment at the microscale, on extended Urban areas, in a reduced computing time. Test cases on Paris show the feasibility of such calculations and the ability to obtain a very detailed representation of output fields, even far away from the release source. Evaluations of speedup prove that the parallelization of PSWIFT is very efficient both for tile's parallelization and timeframe's parallelization. Performances obtained with PSPRAY are quite good too but, as communications between cores are more important, parallelization may be less efficient for huge numbers of core.

Discover hidden collaborations