Sources of aa data ================== 1. The NGDC CD-ROM (NGDC01). The 3-hour aa indices and the daily mean value from 1-1-1868 to 31-12-1987. 2. A magnetic tape (source unknown) - in the format advertised in Mayauds text on aa for 1868-1967 - but the tape had values to the end of 1977. The file includes the northern and southern K indices and the 3-hour aa indices (no daily mean). The aa indices are ALL in EXACT agreement with the values on the CD-ROM 1868-1977. Values computed here ==================== We decided that we would assemble and maintain a complete aa index file (set of yearly files) in the format on the Mayaud tape. We later decided that it would be useful to also maintain the file of 3-hour aa indices (one file with all years called AA3HR.DAT) in the CD-ROM format. 1978-1987 ========== The Hartland and Canberra (Toolangi 1978-79) K-indices were needed to add to the aa data from the CD-ROM for 1978-1987. This also enabled us to check the aa values on the CD-ROM by direct computation from the K indices. We had the 3-hour aa indices and the K indices from Hartland and the formula used to compute the aa index. We used these pieces of information to deduce the Toolangi K index up to the end of 1979 and the Canberra K index therafter. BGS were sent the Canberra K indices (from Canberra) for 1986-1988 and up to 30-06-1989 (on floppy disk). This allowed us to check that the procedure used to compute the Southern Hemisphere K index (for 1978-1985) worked, by checking for 1986 and 1987. A further check would have been to use the listed K indices (from Jan 1982) in the Australian monthly issues of the 'Geophysical Observatory Report'. The following is a list of the aa values on the cd-rom that we have decided must be wrong and have corrected in the BGS yearly files (Mayaud tape format) and the BGS AA3HR.DAT file (cd-rom format), which also includes a more accurate daily mean value than the original cd-rom data. A * below a northern or southern (or both) K index denotes the probable source of the error in the aa index. Correct Correct Date 3-Hour aa indices Northern Hemisphere Southern Hemisphere (top-ours,bottom-Paris) 3-Hour K indices 3-Hour K indices ==== ======================= =================== =================== 1978 810 8 8 8 9 31 16 8 12 1 1 1 0 3 2 1 2 1 1 1 2 3 2 1 1 1978 810 8 8 8 9 23 16 8 12 * 1978 814 20 16 8 16 8 20 9 2 3 2 1 2 1 3 0 0 1 2 1 2 1 1 2 0 1978 814 20 16 8 16 8 20 12 2 * 197810 5 33 5 2 12 8 16 16 12 4 1 0 1 1 2 2 2 1 0 0 2 1 2 2 1 197810 5 33 5 2 12 8 16 16 8 * 1979 425 189189167133187 65 79 79 7 7 6 5 5 3 5 5 5 5 6 6 7 5 4 4 1979 425 189134167133187 65 79 79 * 1980 2 5 5 5 9 12 9 5 2 5 0 0 0 1 0 0 0 0 1 1 2 2 2 1 0 1 1980 2 5 5 5 9 9 9 5 2 5 * 1980 325 5 5 5 5 12 12 12 37 1 0 0 1 2 2 2 4 0 1 1 0 1 1 1 2 1980 325 5 5 5 2 12 12 12 37 * 1980 415 45 24 46 12 20 32 45 45 4 3 3 2 3 3 4 4 3 2 4 1 1 3 3 3 1980 415 45 24 46 20 20 32 45 45 * 1980 425 24 16 8 46 32 12 2 5 3 2 1 3 3 2 0 1 2 2 1 4 3 1 0 0 1980 425 24 16 8 34 32 12 2 5 * 1980 430 12 12 8 12 32 16 12 12 1 2 1 2 3 2 2 2 2 1 1 1 3 2 1 1 1980 430 12 12 8 12 24 12 8 8 * * * * 1980 515 5 12 16 5 12 2 5 9 1 2 2 1 1 0 1 2 0 1 2 0 2 0 0 0 1980 515 5 12 16 9 12 2 5 9 * 1980 525 24 24 81137 59 32 5 12 3 3 4 5 4 3 1 2 2 2 5 6 4 3 0 1 1980 525 24 24 81116 59 32 5 12 * 1980 630 20 12 8 20 17 20 12 12 3 2 1 1 3 3 2 2 1 1 1 3 0 1 1 1 1980 630 20 12 8 20 16 12 20 17 * * * * * * 1980 930 16 8 9 5 2 5 12 2 2 1 0 0 0 1 2 0 2 1 2 1 0 0 1 0 1980 930 16 8 9 12 9 30 8 9 * * * * * 19801123 16 16 12 5 5 24 46 24 2 2 1 0 0 2 3 3 2 2 2 1 1 3 4 2 19801123 16 16 12 5 5 32 46 24 * 19801130 45 32 16 24 24 59 59 67 4 3 2 2 2 4 4 5 3 3 2 3 3 4 4 3 19801130 45 32 16 24 24 46 59 45 * * 1981 411 2 12 32 24 80 80 24 80 0 1 3 2 5 5 3 5 0 2 3 3 4 4 2 4 1981 411 2 12 32 24 80 59 24 80 * 1983 224 20 24 12 16 38 59 59 12 3 3 1 2 2 4 4 2 1 2 2 2 4 4 4 1 1983 224 20 58 12 16 38 59 59 12 * 1986 120 20 24 32 24 32 12 24 45 3 3 3 2 3 2 3 4 1 2 3 3 3 1 2 3 1986 120 20 24 32 24 32 8 24 45 * 1986 121 67 9 8 20 32 80 80 67 5 2 1 3 3 5 5 5 3 0 1 1 3 4 4 3 1986 121 67 5 5 8 32 80 80 67 * * * 1986 419 20 12 8 16 32 24 24 12 3 2 1 2 3 3 3 2 1 1 1 2 3 2 2 1 1986 419 20 12 8 24 32 24 24 12 * * 1986 813 16 24 32 38 20 24 8 8 2 3 3 2 1 3 1 1 2 2 3 4 3 2 1 1 1986 813 16 24 32 38 20 24 12 8 * * 1986 920 12 8 46 12 16 16 37 45 2 1 3 2 2 2 4 4 1 1 4 1 2 2 2 3 1986 920 12 8 46 12 16 16 37 67 * 19861128 2 12 8 8 9 8 16 8 0 2 1 1 0 1 2 1 0 1 1 1 2 1 2 1 19861128 5 12 8 12 9 8 16 8 * * * * 1987 526 5 12 12 24 12 32 58 12 1 2 2 2 2 3 5 2 0 1 1 3 1 3 2 1 1987 526 5 12 24 24 12 32 58 12 * 19871012 12 12 8 5 24 32 38 24 2 1 1 1 2 3 2 2 1 2 1 0 3 3 4 3 19871012 12 12 8 5 24 32 12 24 * 19871212 32 24 16 24 16 16 8 2 3 2 2 2 2 2 1 0 3 3 2 3 2 2 1 0 19871212 32 24 16 24 16 12 8 2 * * Examples -------- First one :- For Hartland K = 3 there is no possible Toolangi K which gives aa = 23. First we checked that the Hartland K index had been entered correctly, which it had. Then we asked Canberra for the Toolangi K, which turned out to be 3. The Hartland/Toolangi combination, which produces an aa index of 23, is Hartland K = 2 and Toolangi K = 3. Since we know that the Hartland K = 3, we deduce that the official compilers of aa either entered or were sent the wrong Hartland K index. The correct aa for the 3/3 combination is 31. Last one :- Having checked that for both Hartland and Canberra, K = 2 is correct, aa = 12 must be wrong, since this would mean a combination of either Hartland = 2 and Canberra = 1 or Hartland = 1 and Canberra = 2. Therefore the compilers of aa have an error in one of the K values, but it is impossible for us to tell which one. In addition =========== The following is a list of changes we have made to the BGS version of the files for the times when: either corrections have made to K indices, which we were sure were incorrect; or gaps have been filled to agree with the cd-rom data : Obs K 3-Hour K-Value changed correct error Date period(s) from to aa values ===== ==== ========= ======================= ========= Too 10 8 78 5 1 3 31 Too 4 4 79 1,2 2 1 4 3 113 66 Too 25 4 79 2 2 5 189 Cnb 25 3 80 4 2 0 5 Cnb 25 4 80 4 1 4 46 Cnb 30 4 80 5-8 2 1 0 0 3 2 1 1 32 16 12 12 Cnb 15 5 80 4 1 0 5 Cnb 25 5 80 4 3 6 137 Cnb 30 9 80 5-8 2 2 3 2 0 0 1 0 2 5 12 2 Cnb 23 11 80 6 1 3 24 Cnb 30 11 80 6,8 2 2 4 3 59 67 Cnb 11 4 81 6 1 4 80 Cnb 24 2 83 2 3 2 24 Cnb 20 1 86 6 2 1 12 Cnb 21 1 86 2-4 2 0 3 0 1 1 9 8 20 Had 28 4 86 1-8 missing 3 3 3 3 2 2 1 1 24 32 46 32 12 9 5 5 Had 29 4 86 1-8 missing 2 3 2 3 3 1 2 2 12 24 16 24 32 8 9 9 Had 30 4 86 1-8 missing 2 2 1 2 1 3 2 1 16 12 5 9 12 20 12 5 Had 1 5 86 1-8 missing 1 1 2 2 2 0 2 2 5 8 12 12 12 2 12 9 Had 8 6 86 8 missing 3 20 Cnb 20 9 86 8 1 3 45 Cnb 19 7 87 6-8 missing 2 2 2 24 24 24 Cnb 20 7 87 1-8 missing 1 1 1 1 1 1 1 1 8 20 12 12 12 20 20 20 Cnb 21 7 87 1-2 missing 1 1 8 12 Cnb 3 8 87 1 missing 1 8 Cnb 16 9 87 1-8 missing 4 2 3 3 3 3 3 3 59 12 32 32 32 32 32 32 1988 onwards ============ For 1-1-1988 onwards we have computed aa indices from the K indices from Hartland and Canberra. (The January 1988 K indices on the floppy disk are ALL different from those published in the 'Geophysical Observatory Report' Vol.36 No.1. - Canberra have informed us that the correct K indices for January 1988 are on the floppy disk, therefore the correct ones were used in the calculations of the aa indices.) The only sources of the aa-index available to check the computed aa indices, at this time, are the weekly telexed provisional aa indices from Paris (from 16 Jan 1989) or the monthly preliminary tables, issued by ISGI, which lists the am index, pm index, the northern index, the southern index and includes a monthly mean value. The following is a list of changes we have made to the BGS version of the files for the times when: either corrections have been made to K indices, which we were sure were incorrect; or gaps have been filled to agree with either of the above two sources: Obs K 3-Hour K-Values changed correct error Date period(s) from to aa values ===== ==== ========= ======================== ========= Cnb 5 9 88 1-3 missing 2 1 0 16 8 5 Cnb 22 1 89 5 4 5 102 Cnb 25 1 89 6 4 1 12 Had 8 4 89 6 3 4 45 Had 7 5 89 1 4 1 5 Cnb 6 6 89 8 0 4 80 Cnb 7 6 89 1&2 - 1 5 3 102 45 Cnb 26 7 89 2 missing 2 16 Had 18 9 89 4 missing 3 46 Had 20 9 89 3 missing 0 2 Had 21 9 89 6 missing 1 8 Had 15 11 89 6 missing 1 8 Cnb 16 1 90 7 2 3 45 Cnb 22 2 90 5 missing 3 32 Cnb 26 2 90 8 missing 4 46 Cnb 21 5 90 2 4 1 12 The monthly lists issued by ISGI contain, we believe, some errors probably due to incorrect entry of K indices. These are not listed here, since the ISGI lists are provisional. Ellen Clarke and David Kerridge, 1991