Exploiting MOPITT's 'Solar Channels'
The Measurements of Pollution in the Troposphere (MOPITT) satellite instrument has been routinely monitoring tropospheric concentrations of carbon monoxide (CO) since March 2000. MOPITT is equipped with gas correlation radiometers incorporating both length modulation and pressure modulation cells operating in two distinct spectral bands: the near-infrared (NIR) CO overtone band near 2.3 µm and the thermal-infrared (TIR) fundamental band near 4.7 µm. Conceptually, the NIR observations sense the attenuation (by CO molecules) of solar radiation reflected from the Earth's surface to the MOPITT instrument, whereas the TIR observations detect CO signatures in thermally emitted radiation from the Earth's surface and atmosphere. In principle, the NIR radiances provide information with respect to the CO total column with very weak sensitivity to the vertical distribution of CO whereas the TIR radiances are sensitive to differences in CO concentrations over broad layers in the troposphere. Over land, where the surface reflectance and hence NIR radiances are relatively high, the NIR radiances should provide more consistent sensitivity to CO in the lower troposphere than the TIR radiances, which are sensitive to lower tropospheric CO only in favorable thermal contrast conditions. Thus, the TIR and NIR radiances provide complementary information with respect to the CO vertical distribution and will ultimately be exploited in conjunction in a single retrieval product.
As a prerequisite to exploiting the NIR radiances for this 'merged' CO product, recent work has focused on demonstrating the value of purely NIR-based retrievals. The figure below compares MOPITT NIR-based retrieval results with standard TIR-based retrievals for a ten-day period over South America. As detailed in Deeter et al. (2009), NIR-based results were obtained by artificially amplifying the instrumental noise for the NIR radiances to account for 'geophysical noise.' Both the TIR and NIR products clearly reveal a large area of biomass burning emissions in the southern Amazon Basin. Quantitative validation results for the NIR product based on in-situ profiles at a variety of sites confirm the value of NIR-based retrievals and reveal no significant retrieval bias. The development of methods to optimally merge the TIR and NIR observations in a single product is ongoing.
Figure 1:
Comparison of MOPITT TIR- and NIR-based CO total column retrievals for the
Amazon Basin for September 1-10, 2006.
References
Deeter, M. N., D. P. Edwards, J. C. Gille, and J. R. Drummond (2009), CO retrievals based on MOPITT near-infrared observations, J. Geophys. Res., 114, D04303, doi:10.1029/2008JD010872.