Observation of CO from space over megacities

Increases of the global population combined with the economic growth in many of the developing countries are leading to an increase of urban surface area and thus globally the associated air quality issue. Human activities emit vast quantities of pollutants with carbon monoxide (CO) a prime example. Several nadir-viewing thermal infrared sounders monitor this gas from space but their limited sensitivity to the boundary layer is a well-known disadvantage of this technique. Our study investigates the performance of a new retrieval algorithm applied to Measurements of Pollution in the Troposphere (MOPITT) data (version 5) that combines the thermal infrared (TIR) with near-infrared (NIR) bands that are more sensitive to the boundary layer. This new data product is compared with the TIR-only product as well as measurements from the Infrared Atmospheric Sounding Interferometer (IASI), also a TIR sensor. The study focuses on eight megacities: Moscow, Paris, Mexico, Tehran, Baghdad, Los Angeles, Sao Paulo and Delhi. High-resolution maps of the CO distribution over these locations have been generated using a new pixel averaging technique that clearly demonstrates a CO hit-spot. Combining the satellite data with wind data from meteorological reanalysis a clear dependence of the CO distribution with near-surface wind speed direction is found. A clear reduction of CO emission over all sites between 2000-2005 and 2006-2011 is observed, reaching ~ 13% over Mexico and a megacity as Baghdad emitted the same amount of CO between 2006 and 2011 than Tehran or Mexico between 2000 and 2005