Using Multiple Satellites and Models to Estimate Atmospheric Composition and Source Magnitude Changes of Aerosols Over the Past 2 Decades: Regions of Increase, Decrease, and High Variability

Since 2000 we have had access to multiple daily (or higher) frequency measurements of aerosols and related trace gasses on a global scale. Combining the data from these different remotely sensed platforms in space and on the ground, has allowed us to understand and begin to quantify how the aerosol loadings over some regions have increased, while others have decreased, and others still have swung in multiple directions. We have observed changes which have occurred slowly and steadily from year to year, while we have observed others that have recurring patterns with intense inter-annual and intra-annual variation.

Combining these multiple measurements with the CESM, we have developed a new emissions dataset. The magnitudes, spatial, and temporal variation have been improved using an ensemble Kalman Filter as well as weekly frequency measurements of NO₂from OMI, CO from MOPITT, AOD from MODIS, and height from MISR. Our end product is a global aerosol source map, at weekly resolution.

These new a priori emissions are run in the CESM GCM in online mode, to better understand and match the observed regional and temporal increases, decreases, and highly variable events. Differences in the chemical, physical, and in-situ processing of the short lived, intermediate lived, and long-lived species allows for improved understanding of where emissions have been changing, as compared to where long-range transport or in-situ processing are more responsible for the observed atmospheric changes.

We observe clearly decreases in sources from North America and Western Europe. We observe regions over East Asia which have increased and later decreased, while other regions are still continuing to increase. Furthermore, we notice that in Southeast Asia, Africa, and regions of South America, that sources have generally increased. On top of this, we work to separate the impacts of large-scale decreases in sources on average, as compared with possible increases in short and intense sources, due to biomass burning and/or rapid changes in land cover associated with expanding urbanization.

Finally, weaknesses in the model assumptions associated with vertical transport, mis-characterized removal and in-situ processing, and missed intense but variable events are further discussed. In some cases, uncertainties in emissions can be expanded to cover these observations. However, in other cases, missing processes are highlighted for future work. Finally, limitations of the current measurements are also highlighted for high latitudes and for regions of rapid urban development.