#3 T04_s input parameter type mismatch

[Sorin Zaharia]

I was obtaining dubious results for L* with T04S (~dipolar values) for positions in the SM equatorial plane from L=3 to L=9, regardless of solar wind conditions. I am using Absoft and Gfortran compilers on a Mac. A colleague was getting the same strange results on Linux with the PGF compiler.

The problem is a mismatch between the types passed to Tsyganenko04.f and those of the calling arguments; the first declaration line in T04_s (file Tsyganenko04.f)

REAL PARMOD(10),PS,X,Y,Z,BX,BY,BZ,tilt

declares those variables single precision, while they are defined as double precision in the calling routine (init_nouveau.f). This led to strange PARMOD values in Tsyganenko04.f (most of them zero, with some W values even negative) and incorrect T04S field.

Changing the above line to

REAL*8 PARMOD(10),PS,X,Y,Z,BX,BY,BZ,tilt

fixed the problem and led to realistic L* values for T04S.