Overview of the delhi Air Quality forecasting system
- 1. Delhi 2010 and beyond Demonstration of the Air Pollution Modeling System Dr. Sarath Guttikunda, Dr. Giuseppe Calori & Dr. Benjamin Guinot New Delhi, India May 2010 ARIA Technologies SA LEOSPHERE 8/10, Rue de la Ferme, 92100 Boulogne Billancourt - FRANCE Bâtiment 503, Centre scientifique d'Orsay Phone : +33 (0)1 46 08 68 60, –Fax : +33 (0)1 41 41 93 17 Plateau du Moulon, 91400 Orsay – FRANCE e-mail : info@aria.fr, web: http:/ /www.aria.fr Phone: +33 (0)1 69 35 88 17 - Fax: + 33 (0)1 69 35 87 11 e-mail: info@leosphere.fr, web: http:/ /www.leosphere.com
- 2. Program Objectives An Air Quality Forecasting System for Delhi To be operational for Commonwealth Games, October, 2010 And beyond A strategy to dessiminate information to national and local agencies, media, and public • Products customized for web (AQ indexes, Google maps & forecasts) • Time series for hot spots • Customized spatial maps 2
- 3. Program Partners Project is supported and financed by the French Ministry of Economy, Industry & Employment of the Government of France under Indo-French Technical Assistance Programme (FASEP) Indian Counterpart French consortium 3
- 4. Program Components • A 3D-nested air pollution modeling system • With support from the local monitoring and a unique aerosol LIDAR network • Multi-pollutant analysis including PM, SO2, NOx, CO, HCs, & Ozone • Dissemination of results • Training 4
- 5. Air Quality Modeling System Nests – National to Urban (Delhi)
- 6. Air Quality Modeling System Components • A 3D chemical transport eulerian air pollution modeling system • 4 Nests from National Scale @ 64km to Urban Scale @ 1 km (covering ~3000km and 50km respectively) • Multi-pollutant analysis including PM, SO2, NOx, CO, HCs, & Ozone • Boundary conditions for NEST 1 are from MOZART 6
- 7. Nesting Configuration 1. National @ 64 km 2 1 3000 km 7
- 8. Nesting Configuration 1. National @ 64 km 2. Regional @ 16 km 3 2 800 km 8
- 9. Nesting Configuration 1. National @ 64 km 2. Regional @ 16 km 3. Sub-regional @ 4 km 4 3 200 km 9
- 10. Nesting Configuration 1. National @ 64 km 2. Regional @ 16 km 3. Sub-regional @ 4 km 4. Urban @ 1 km 4 50 km 10
- 11. Air Quality Modeling System Monitoring Support
- 12. Monitoring Support Stationary monitors across the country 340+ stations operated by CPCB and SPCBs LIDAR – 4 Aerosol and 1 Wind 12
- 13. LIDAR Configurations Nest 2 Nest 3 Nest 4 city 50 km Sample of aerosol multilayering obs. by Lidar in Kanpur 200 km Time 800 km
- 14. LIDAR Installations – Phase 1 Nest 2 Nest 3
- 15. LIDAR Configurations – Phase 2 Nest 4 During the Games Tentative locations DU CWG village Siri Fort SC ?? Seeking further cooperation with local institutions to support LIDAR 50 km operations
- 16. Air Quality Modeling System System Architecture
- 17. Modeling Architecture Synoptic meteorological forecast (NCEP) Gap WRF SurfPRO Geographic data Meteorological input 4670000 4660000 Emission input 4650000 4640000 4630000 4620000 760000 780000 800000 820000 EMMA CHIMERE/ Emission inventories FARM LIDAR Inputs for calibration pollution fields 17
- 18. Simulation Times WRF: 8 processors CHIMERE: 4 processors FARM: 6 processors 54 hours of simulation from 12 UT (day 0) to 18 UT (day+2) CHIMERE CHIMERE FARM FARM NCEP download NCEP download WRF WRF (2 domains) (2 domains) (2 domains) (2 domains) END 1hr 3hr 2hr 4hr 13:00 GMT 14:00 GMT 19:00 GMT 23:00 GMT 18:30hr local time 00h00 local time 04h30 local time
- 19. Input Data - Meteorology WRF Meteorological forecast for 48 hrs for 4 Nests ~ computational time 5 hrs 19
- 20. Air Quality Modeling System Emissions Information
- 21. 1. Multi-source & Multipollutant inventory Meteorological outputs Meteorological Data from WRF Processing Normal Meteorological Variables: Dust & Sea salt wind velocities, temperature, pressure, Emissions relative humidity, precipitation, cloud Biomass Emissions cover, etc Biogenic Emissions CHIMERE/FARM – Eulerian Large point Emissions Chemical Transport Model – Post sources Processing Analysis coupled for aerosols Volcanic Emissions Anthropogenic Area Emissions
- 22. Input Data - Emissions 1. Global and Regional Inventories Organization of Emissions Data anthropogenic RETRO Point National (23 countries) Regional (94 regions) Urban (22 cities) (355 sources) REAS EDGAR 1° x 1° down to Lat/ 1 km x 1 km long direct Historical: Current: Projections: 75-95 90-00 95-30 BB (5-yr) (1-yr) (5-yr) Gridded Dust emissions Species: SO 2 BC NH 3 NO x PM 10 OC SO 2, NO x CO PM 2.5 NMVOC
- 23. • • • Road Network Administrative Point Sources • • • GIS Systems Surveys
- 24. • http://www.jamstec.go.jp/frsgc/research/d4/emission.htm • Developped by the Frontier Research Center for Global Change • 0.5 degrees resolution • NOx, SO2, CO, CO2, N2O, NH3, BC, OC, CH4, NMVOC • Reference years 1995 - 2000 • Prospective inventories 2004-2009 • Past and future inventories (1980 - 2020)
- 25. Input Data - Emissions 2. New National Scale Inventories Transport Emissions T rucks Buses Buses 2010 45% 2010 16% 2010 17% 4- Wheelers Trucks 8% 41% 4- Trucks Wheelers 3- Buses 39% 8% Whee lers 23% 11% 3- Railw ays Wheele rs 12% 2-Wheelers 16% Railways Railways 4-Wheelers 3-Wheelers 0% 2- 2- 4% 17% 3% Wheelers 0% Wheelers 26% 14% PM10 SO2 CO 25
- 26. Input Data - Emissions 1. New National Scale Inventories 4000000 3500000 3000000 2500000 21 Urban 2000000 20 Water and Land Mixtures 19 Forest/Field Mosaic 18 Mixed Forest 17 Deciduous Shrubs 16 Evergreen Shrubs 15 Ocean 14 Inland Water 13 Bogs and Marshes 1500000 12 Ice Caps and Glaciers 11 Semidesert 10 Irrigated Crops 9 Tundra 8 Desert 7 Tall Grass 6 Evergreen Broadleaf Trees 1000000 5 Deciduous Broadleaf Trees 4 Deciduous Needleleaf Trees 3 Evergreen Needleleaf Trees 2 Short Grass 1 Crops, Mixed Farming 500000 -1000000 -500000 0 500000 1000000 1500000 Power plants Landuse 26
- 27. Input Data - Emissions 2. Emission Distribution Schemes Population Urban Points 27
- 28. Input Data - Emissions 2. Emission Distribution Schemes Non-urban Highways Extents 28
- 29. Input Data - Emissions 2. New National Scale Inventories Domain 1 Domain 2 29
- 30. Input Data - Emissions 3. Urban Scale Inventories Industrial Population layout by Wards 30
- 31. Input Data - Emissions 3. Urban Scale Inventories Wooded & Metro & rail Green areas lines 31
- 32. Input Data - Emissions 3. Urban Scale Inventories gridded NH 1 NH 10 NH 24 Ghaziabad CP Outer Inner Ring Ring road road India Gate Noida NH 8 NH 2 Gurgaon Faridabad Dominant transport corridors Traffic flow survey data 32
- 33. Input Data - Emissions 3. Urban Scale Inventories – Other sources Summer Winter Study results from PM2.5 source apportionment of measured samples (2005, Chowdhary and Russell, et al.,) Source apportionment & emissions inventory from CPCB, 2010 33
- 34. Air Quality Modeling System Output Formats
- 35. Preliminary Concentrations output NO2 and Ozone on 1st domain (morning)
- 36. Preliminary Concentrations output NO2 and Ozone on 2nd domain (morning)
- 37. Preliminary Concentrations output NO2 and Ozone on 4th domain (midday) NCR – Delhi region (52 x 52 @ 1km)
- 38. Preliminary Concentrations output New Delhi & Jaipur
- 39. Results – Pollution Maps Demo of User Interfaces 39
- 40. Next Steps Finalizing inventories Model configurations LIDAR operations Customizing outputs Dissemination Strategy Trainings Seeking further cooperation with institutions to support data collection 40
- 41. Questions? Send email to Dr. Sarath Guttikunda sguttikunda@aria.fr Dr. Giuseppe Calori gcalori@aria.fr