You can very effectively use MODIS/MISR/OMI/TOMS data for PM estimation using satellite datasets. There are lots of papers available on this aspect. If you need any specific paper for PM estimation using AOD, AAOD or AOT let me know.

Mr.Luis Inostroza, there are several methods for detecting airpollutants. For example if you want to find concentrations of the harmful trace gases in the atmosphere, DOAS - Differential Optical Absorption Spectroscopy is one such technique which is widely used in case of satellite measurements for retrieving the trace gas constituent profiles and concentrations. for example this paper gives an idea about the retrieval of ch4 and co2 with the above mentioned technique (from Envisat data).

http://onlinelibrary.wiley.com/doi/10.1029/2000JD900191/full

There are some methods by satellites, lidar and ground spectrotradiometers. Satellites mainly give columnar content, as ground spectroradiometers. Lidar gives the profile. While satellites have a repetition time depending on their swath, lidar and spectroradiometers make local measurements and operate continuously. You can have an idea looking the site http://www.temis.nl/. As far as the books describing the different techniques, as DOAS, looks also to the series Physics of Earth and Space Environment http://www.springer.com/series/5117. Satellites have been already cited and I want to add GOME(ERS 2) and Sciamachy (ENVISAT), GOME 2 (METOP) of which I was PI, with OMI. As far as spectroradiometers look at my ground spectroradiometer and french DOAS(SAOZ) by Pommerau or German DOAS by Platt.

LIDAR, MODIS, and insitu data (spectroradiometer field data) are suitable for such study. Multiregression between remote sensing and observed results may improve your study results.

Which air pollutant you are specifically intrested in are you looking at aerosols?

Hi everybody! Thanks for such interesting answers. As I said I am searching for method to identify Particulate Material, 2.5-10, as they have been highlighted as important air pollutants in Latin American cities, mainly due to burning practices and dusty roads. In the beginning I was thinking whether would be possible to use LANDSAT 8, as this new sensor has some aditional bands, designed in principle for such purposes... What do you think?

Yes you can use Landsat 8 new channels for aerosol columnar measurerements. There are some different methodologies as far as for TOMS or my approach as it is described in my papers, referred in www.temis.nl

For monitoring of PM2.5 and PM2.5, you should have high temporal satellite - You can use MODIS AOD for mapping of these particulates - here is few studies related to MODIS AOD:

http://www.atmos-chem-phys.net/10/10399/2010/acp-10-10399-2010.pdf

http://www.sciencedirect.com/science/article/pii/S0034425713001417