November 30, 2010: A glitch fixed in TRACE_08: when tracing a field line from space to Earth, its final footpoint coordinates {XF,YF,ZF} were not included in the last record of output arrays XX,YY,ZZ.

November 27, 2009: A minor bug fixed in RECALC_08. In the case of non-radial solar wind, a warning message popped up at each call of the subroutine. Now it is displayed only once, as originally intended.

GEOPACK: earlier corrections/updates:

April 21, 2008: because of a bug in the subroutine STEP_08, in the first version of Geopack-2008 the parameter DSMAX had no effect on the field line tracing stepsize, if made parallel to the magnetic field direction (e.g., from southern to northern hemisphere). Earlier versions of the source file should thus be replaced by the most recent one. Thanks are due to Dr. S. Dubyagin (University of St.-Petersburg) for pointing out the problem.

August 29, 1997: Subroutine RECALC: a correction was made to the calculation of the quantity DJ (lines #176-177). The earlier version of the FORTRAN code contained a bug, which resulted in a small (less than 4*10^-7 degrees) error in the calculated orientation of Earth's rotation axis. Owing to the extremely small magnitude of the error, its effect can be ignored in all applications. (Note that the angles returned by the subroutine SUN are accurate within a margin of 6*10^-3 degrees, i.e., 4 orders of magnitude wider). Thanks are due to Dr. A. Petrukovich (IKI, Moscow) for pointing out the mistake.

July 15, 1999: Subroutine RECALC: in the comment line (#30), the input parameter ISEC was erroneously referred to as "seconds of day" . Its actual meaning is SECONDS OF MINUTE. The correction was prompted by courtesy of Drs. S. Petrinec (Univ. of Washington), G. Le, and Prof. C. T. Russell (IGPP UCLA).

January 5, 2001: An update of the GEOPACK has been made, including an extension of the IGRF and RECALC subroutines (main field coefficients for the epoch 2000 were added); for more details, see comments preceding the FORTRAN subroutine listings.

April 3, 2001: Subroutine RECALC (updated version): lines 98-99 should read

        ELSEIF (IY.LT.2000) THEN                        !1995-2000
           F2=(FLOAT(IY)+FLOAT(IDAY)/365.-1995.)/5.
  

May 14, 2002: Line 29 of the subroutine IGRF was modified as follows:

           SAVE MA,IYR,IPR,G,H,REC
  

In the earlier version, the last three variables were missing, which caused run-time errors with the g77 compiler 3.0.4 on SUN OS 5.8 (Solaris 8). (Courtesy of Dr. Sean Chen, RITSS, Code 632, NASA GSFC).

To avoid possible problems of the same kind in the subroutines DIP and RECALC, the following changes were also made:

Line 20 of the s/r DIP should read:

      SAVE M,PSI,SPS,CPS
  

Line 38 of the s/r RECALC should read:

      SAVE IYE,IDE,IPR,SQ,SQQ,SQR
  

In the earlier version, the last term in the parentheses (second line) was incorrectly specified as 1990. (Courtesy of Dr. Scott Boardsen, RITSS Corp.)

May 9, 2006: Subroutine RECALC: interpolation of the coefficients (between labels 50 and 105 is now made through the last element of the arrays G(105) and H(105) (previously made only through N=66, which in some (rare) cases caused runtime errors)

T01_01 model: corrections/updates of the source code:

February 13, 2002: In some cases with abnormally low values of the solar wind pressure, the previous version of the model code crashed at low altitudes, due to inaccurate scaling of the Birkeland currents. In the corrected version, the problem has been eliminated. Courtesy of Dr. Mattias Foerster, (Max-Planck Institute, Germany).

May 14, 2002: An update was made of the FUNCTION AP(R,SINT,COST), to eliminate possible floating underflows of the variables DEXP1, DEXP2, and DEXP3, becoming exponentially small at large geocentric distances.

June 24, 2006: Replaced coefficients in: (i) DATA statement in FUNCTION AP, (ii) DATA C_SY statement in SUBROUTINE FULL_RC, and (iii) DATA A statement in SUBROUTINE T01_01. This correction was needed because of a bug found in the symmetric ring current module. In all cases tested, the largest discrepancy due to the error did not exceed a few nT and was found in the innermost magnetosphere (R<=2).

TS04 model code corrections/updates:

March 14, 2005: 79 -> 69 in the line 102 of the source code: should read CALL EXTERN (IOPGEN,IOPTT,IOPB,IOPR,A,69,...

(caused compilation problems with some Fortran compilers). Courtesy of Dr. S. Sazykin, Rice University.

June 24, 2006: Replaced coefficients in: (i) DATA statement in FUNCTION AP, (ii) DATA C_SY statement in SUBROUTINE FULL_RC, and (iii) DATA A statement in SUBROUTINE T04_S. This correction was needed because of a bug found in the symmetric ring current module. Its impact was found to be significant (up to ~20 nT) only in the innermost magnetosphere (R<=2) and only for extremely disturbed conditions; otherwise, the correction resulted in only minor (up to a few nT) changes of the inner field.

Nikolai Tsyganenko - Institute of Physics, University of St.-Petersburg, Russian Federation (tsyganenko@geo.phys.spbu.ru)