Skip to main content
CERES Home Page CERES Home Page


Aerosols for radiative transfer calculations are obtained by a hierarchical selection: high resolution MODIS retrievals are preferred, lower resolution daily averaged database of MODIS retrievals come next, and then the MATCH aerosol assimilation. Aerosols are obtained only from high resolution MODIS retrievals or MATCH. David Fillmore in Boulder supplies MATCH files for CERES.

Height profiles from MATCH which are typically different for each aerosol type, each day, and each location. The height profile effects the LW forcing, which can be significant regionally for larger aerosol particles like dust and sea salt, at both surface and TOA. The vertical placement of aerosols relative to clouds can have a strong effect on the absorption of SW by dust, black carbon, and insoluble organic carbon. The 7 aerosol types are treated as external mixtures, as are aerosols and clouds.

Image: Bias of SW due to changes in Aersols

Figure 4: Biases of CRS calculations with SSF observations. CRS and SSF Edition2F (Edition2B) use MODIS Collection 5 (4). CRS Edition2F vs. SSF Edition2F and CRS Edition2B vs SSF Edition2B for months of May though December in respective years.

The upper left panel in Figure 4 shows the relative bias of the untuned calculation for reflected SW at TOA with respect to CERES observations in all sky conditions. The relative bias was evaluated from a raw monthly mean of the footprint-by-footprint bias divided by the correpsonding raw monthly mean of the footprint-by-footprint observed flux. It is not a grid-box statistic. Because higher latitudes are observed more frequently than lower latitudes, a footprint-by-footprint statistic places less weight on the tropics than does a grid-box (area-weighted) statistic. The large scale comparison in Figure 4 spans eight-month blocks (May to December), one (2006) for CRS Edition2F and six (2000-2005) for CRS Edition2B. For relative biases of untuned all-sky global SW (upper left) and the correponding tuned all-sky global SW (lower left), CRS Edition2F performance is quite comparable to that of CRS Edition2B. The upper right panel of Figure 4 indicates the untuned, all-sky global simulation of OLR in Edition2F is very much like that in Edition2B. For a description of the Terra CRS software and the more extensive comparison of Edition2F with data see the CRS Terra Edition2F Data Quality Summary (DQS) link to external site.

CERES Data Product Information

  Image: NASA Logo NASA Official: Dr. Norman Loeb
Page Curator: Edward Kizer
Page Last Modified: 11/14/2018 11:28:48 EST
Site Last Modified: 11/14/2018 11:28:59 EST