Notes on GOES FITS files containing 2 channel XRS data. Kim Tolbert 26-Jun-2012 Rewrote all GOES 14 and 15 FITS files (they start 2-Dec-2009) to fix the following three issues (the YOHKOH GOES files should also be rewritten for the new understanding in the meaning of the times, i.e. shift to midpoint of interval, and the base time error. Sam Freeland may or may not rewrite them) : Rodney Vierek at NOAA tells us that the time in the GOES 13,14,15 NOAA files correspond to 1.024 seconds after the end time of the 2.048s exposure. We were treating those times as the midpoint of the exposure interval. (The 1.024s offset is apparently an error at NOAA, which they plan to correct. They will post the correction when made on their web site.) Now the times in the files are the midpoint of the interval for each data point (I subtracted 2.048 from the times in the NOAA file, 1.024 to get to end of interval, another 1.024 to get to center of interval). Added comment to Comment section in 'FLUXES' extension to indicate times are center of interval. In addition, there were two errors in writing the FITS files. 1. The times were stored as sec relative to the earliest time in the file, but the base time in the header was the start of day. Changed so that times stored are relative to start of day. 2. In routine that reads NOAA file (goes_day_ascii), seconds relative to earliest time in file were being returned as a float. This caused random errors in the msec portion of the times stored in the FITS file. Now return as double to preserve msec. Not that important, just confusing since the times didn't match what was in the input files. 24-May-2011 We're only getting the GOES 15 data since Sep 2010, but noticed that there's also GOES 13 data since May 2010. Looked at a few of those files (from early 2011) to see if we should be getting them too, but the data looked completely different, and decided we shouldn't. Also the NOAA site says GOES 15 is the primary for this period, and there's no secondary. 23-May-2011 Previously, all files from 04-Jan-1980 to 18-Aug-1994 were labeled as being from GOES 6 (both in the file name, and in the TELESCOP header keyword). Amy pointed out that GOES 6 wasn't launched until 1983. Cross-checked with GOES files that are now online at http://satdat.ngdc.noaa.gov/sem/goes/data/full/xrays/ and determined which sat they really were. Didn't just copy their files (which have a mod date of 2004 - don't know when they were created and made available) to replace ours because they seemed to have many more gaps than our files for the same periods. I changed the name of the files, and the TELESCOP header keyword for all of the files through 31-jul-1984 according to this: 4-Jan-1980 through 30-Apr-1983 GOES 2 1-May-1983 through 30-May-1983 GOES 5 2-Jun-1983 through 30-Jun-1983 GOES 6 1-Jul-1983 through 31-Jul-1984 GOES 5 1-Aug-1984 through 18-Aug-1994 GOES 6 GOES 6 main mission was over in 1992, but they kept the XRS data going until Nov 1994. To figure out what the correct sat was for 1980-1984 data, I used the NOAA files at http://satdat.ngdc.noaa.gov/sem/goes/data/full/xrays/ (I'm assuming their files are labeled correctly!). I manually copied their files to my computer. Their files are in the same format as the pre-1980 files, so to read them using the GOES object, I modified gfits_r to force it to use the gfits_r_pre_1980 routine by doing the following: 1. at line 148 added (before retry_gfits_r: label): sat = 5 ; or whatever the number for the satellite corresponds to the file name 2. at line 208 added (just before if nfile gt 0 test): nfile = 1 3. at line 218 (just before the if count gt 0 test before it reads the files) added this: pre_1980=1 & file='X0538408.FIT' & count = 1 (and modified the file name for each file I wanted to cross-check) Difference in temperature: Temp: Time, Y: (1980/06/24 06:36:52.127, 12.757) GOES 2 Time, Y: (1980/06/24 06:36:35.105, 14.620) using GOES 6 response Emis.M: Time, Y: (1980/06/24 13:12:09.805, 0.46178) GOES 2 Time, Y: (1980/06/24 13:09:53.402, 0.37682) using GOES 6 response Temp: Time, Y: (1980/06/24 13:01:01.543, 11.423) GOES 2 Time, Y: (1980/06/24 13:01:37.633, 13.058) using GOES 6 response Emis M: Time, Y: (1981/06/22 15:10:49.008, 0.60372) GOES 2 Time, Y: (1981/06/22 15:12:06.887, 0.60141) using GOES 6 response Temp: Time, Y: (1981/06/22 14:54:33.664, 14.582) GOES 2 Time, Y: (1981/06/22 14:54:13.250, 16.806) using GOES 6 response Emis. M: Time, Y: (1984/06/21 14:20:10.191, 0.26987) GOES 5 Time, Y: (1984/06/21 14:15:51.203, 0.29874) using GOES 6 response Temp: Time, Y: (1984/06/21 13:51:15.750, 8.8608) GOES 5 Time, Y: (1984/06/21 13:51:19.754, 8.8049) using GOES 6 response