Air pollution has emerged as a major environmental challenge in Indian cities and across the states. Complexity of the problem is due to multiplicity of emission sources, such as, industries, automobiles, generator sets, domestic fuel burning, road side dusts, construction activities, etc and a vibrant meteorology across climatic zones. Our India air quality forecasting system interjects contributions of pollution for national, regional and urban areas, to better support a long-term air quality management plan and a short-term health alert system. Various products from the forecasting system include
PM2.5
24-hr average for Particulate Matter (less than 2.5 micron-m) (PM2.5)
 |
 |
 |
 |
PM10
24-hr average for Particulate Matter (less than 10 micron-m) (PM10)
 |
 |
 |
 |
SO2
24-hr average for Sulfur Dioxide (SO2)
 |
 |
 |
 |
NO2
24-hr average for Nitrogen Dioxide (NO2)
 |
 |
 |
 |
CO
24-hr average for Carbon Monoxide (CO)
 |
 |
 |
 |
Hourly Anime
Animations of hourly concentration maps for the next 72 hours
 |
 |
 |
 |
 |
 |
 |
Fire Counts
24-hr Open fires map for the Indian Subcontinent (updated with VIIRS feed every 3 hours)
 |
 |
 |
 |
Dust Stroms
Simulated Dust Storm Zones for the Next Three Days
 |
 |
 |
 |
Our forecasting system also includes a high-resolution medium-range meteorological system for the next 72-hours, processed through 3D-WRF meteorological model and the GFS meteorological fields . The modeling domain covers the Indian subcontinent at a temporal resolution of 1 hour. All the simulation fields are updated everyday at ~17:00 local time.
Rain
Total Precipitation – Daily
 |
 |
 |
 |
Temp (day)
Average daytime near ground (2m) temperature
 |
 |
 |
 |
Temp (night)
Average nighttime near ground (2m) temperature
 |
 |
 |
 |
Winds (day)
Average daytime near surface (10m) winds
 |
 |
 |
 |
Winds (night)
Average nighttime near surface (10m) winds
 |
 |
 |
 |
Met Anime
72-hr animations for surface wind speeds and direction, precipitation, surface temperature, and mixing layer height
 hourly snaps |
 hourly snaps |
 hourly snaps |
 |
WRF model simulated summary of these meteorological parameters by district for the year 2015, is available for browsing and downloading here.
Data above is from stations operated and maintained by the Central Pollution Control Board (CPCB) and the State Pollution Control Boards. All the data is available on CPCB website. A secondary source to download archived data is @ openaq.org. The color code represents the Air Quality Index (AQI) scales of India. More on the air quality monitoring in India is here.
PM2.5 forecasts by State Forecasts (browse here)
 |
Modeled PM2.5 source contributions by State (browse here)
 |
 |
 |
 |
Reconstruction of meteorological fields dedicated to urban and regional scale dispersion modeling- Comprehensive emission inventory of anthropogenic and natural sources, as urban and regional scales. The regional emissions inventory is maintained at 0.25° resolution covering the Indian sub-continent. The emission fields also include natural sources, such as dust storms, fires, sea salt, lighting activities, updated every day
- Simulation of air pollutants dispersion using the WRF-CAMx chemical transport model. The simulated results are available in multiple tiers
- Forecasts – daily (24 hour) average concentrations (maps)
- Forecasts – daily (24 hour) summaries by state – average PM2.5 concentrations (maps) and average source contributions (donuts)
- Forecasts – animations of hourly average concentrations (maps)
- Forecasts – hourly average concentrations for all the criteria pollutants by district (line graphs)
- Forecasts – hourly average modeled source contributions to PM2.5 concentrations by district (bar graphs)
An effective air quality management requires reliable information on ambient air pollution levels with substantial spatial and temporal coverage. Air quality data are generally available through air quality measurement campaigns; which is also an ideal database to statistically forecast pollution levels in the next 3-4 days, based on past experiences. However, this database is not enough to explain the highs and the lows, source contributions, and spatial patterns in the observed trends. As part of the forecasting system, we are able to cover this gap via emissions inventory development, dispersion modeling, and source apportionment.