MAJOR CHANGES TO BELT INDEX CALCULATIONS (10/06/2004) **** Note: All older files can still be found in the old_bi directory **** **** The older files should be used with caution. Please read all the **** **** README files in the old_bi directory before using those files. **** We have recalculated belt indices for all NOAA-15 data since 2000 and for all NOAA-16 data since 2002. Several changes have been made in the procedure for doing these calculations that users should be aware of. The file named bi_YYYY.txt is for the current year, and uses data from the current primary POES satellite. All other files are named bi_NSS_YYYY.txt, where 'SS' is the POES satellite number. In the past we have updated all median sensor responses (that serve as the reference for calculating belt indices) on a yearly basis and use that updated reference set for the next year's processing. This was done to compensate for the gradual degradation of the solid-state detectors in the proton telescope systems. It has been pointed out by a user that while this might be appropriate for the 30 keV to >6900 keV proton belt indices, changing the reference obscures the year-to-year changes in the energetic electron and solar proton radiation environment as tracked by the daily belt indices. The solid-state detectors in the electron telescopes and in the omni-directional detector systems are much less susceptible to radiation damage and there is little need to update the median reference fluxes for these detectors. Therefore, in the reprocessing we have used the NOAA-15, year 1999 electron and solar proton median flux values for all subsequent years. For NOAA-16 we have used the reference values from 2001 in a similar way. We continued to use yearly updated reference fluxes for the calculation of the 30 keV to >6900 keV proton belt indices. A second change implemented in the reprocessing was to use the current year's L-value data in sorting data into inner, slot, and outer zone regions. The L-value set that had been used dated to 1995 and the secular change in the geomagnetic field over time rendered that L-value set out of date. A third change was to compute belt indices for all 22 energetic particle sensor channels observed by the SEM-2 instrument. In the past belt indices had been calculated only for observations from the horizon viewing (90-degree) detectors and the 4 omni-directional detectors. At geographic latitudes above about 30 degrees, the zenith viewing detector systems sense particles in the atmospheric loss cone. Because of this the count rates from the zenith viewing telescopes are ordinarily low, although during periods of high activity those count rates may become comparable with those from the horizon viewing detectors. At low geographic latitudes, however, the role between horizon viewing and vertical viewing telescopes is reversed. At these locations, the vertical viewing detectors sense the trapped particle population and provide the better measure of the radiation belt intensities. Primarily for this reason, but also because the zenith viewing detectors track the radiation belt particle population at higher latitudes, we have now computed daily belt indices for the zenith viewing detectors as well. Finally, we have implemented better data quality procedures in the calculations. These improvements have resulted in minor changes to the daily belt indices calculated from the horizon viewing (90 degree) proton telescope data and from the omni-directional detector data. The changes in the daily indices calculated from the horizon viewing electron telescope data can be more substantial as the reference fluxes are no longer updated. Also, belt indices from all 22 particle species and energy now appear in the listing.