I am not familiar with the ISCCP-FH-MPF product, but pulled down the file for January 2017 and viewed it in Panoply and compared it with our EBAF product for the same date. I compared the sxd1fl, surface downward shortwave flux all sky in ISCCP-FH-MPF, with the EBAF Surface Shortwave Flux Down All Sky (https://ceres-tool.larc.nasa.gov/ord-to ... ection.jsp
). Even though the ISCCP-FH-MPF product indicates a scaling factor of 10, but it appears to be a scaling factor of 100.
The global information from Panoply for the ISCCP-FH-MPF is, Data Min = 0.0; Max = 43370.0; and Mean = 19095.1
The same information for EBAF is Mean: 194.6490 ; Std Dev: 98.3445 ; min=0.000 , max=422.024.
If you apply a scaling factor of 100 for ISCCP-FH-MPF, the means would be 190.95 vs 194.65.
The solar incoming for CERES is daily varying with mean of 1361, and ISCCP is using 1367 with daily varying according to doi:10.1029/2003JD004457. Which likely accounts for most of the difference between the two values.
This does not seem to correspond with a factor of two difference. There are several downward shortwave flux in the ISCCP files, all-sky, clear, diffuse, and direct that might account for the difference.
If you want to provide the CERES product you are using and the exact fields in the two products, I can investigate further.