Version 9.3

This version corrects an error in the reference dates in the locations.dat file. There are also some code updates and more configure scripts for gfortran on Linux. Version 9.3 was released on March 23, 2010.

• Add K5 and VERA as allow values for DAR and RECORDER in the stations catalog. Add MARK5B as an allowed value for DISK. SCHED  doesn't actually do anything with these yet. It just doesn't die if they are there. March 25, 2010 RCW.

• Restrict the cases where the warning for excessive recording scans is issued to cases where the MARK5A recording system is in use. The LBA explicitly does not want it. March 25, 2010. Cormac Reynolds and RCW.

• The CELLS optimization routine is in a state of development so use at your own risk. I don't think anyone else is using it so I don't think this should cause trouble, but I'm leaving this note just in case. March 25, 2010 RCW.

• Fixed bug apparent when not all stations have the same epoch for their station position rates. When building the table of adjusted positions, it was using the reference date from the last station. RCW Mar 23, 2010.

• Add stations Mizusawa (VERA) & Yamaguchi to the stations, locations, and freq files. RCW Mar. 22, 2010.

• The GSFC 2009a geodetic solution is based on a different position reference date from previous solutions. When the new locations.dat was created, the new date was used for all stations. But many stations do not have updates from the GSFC solution, so they then contained incorrect reference dates. This creates station position errors when plate techtonics are taken into account. locations.dat has now been corrected to keep the old reference dates when the positions are not updated. Also fix typo in Petrov's name in the locations.dat file. March 18, 2010. RCW.

• Scripts to create makefiles for use with gfortran on Linux have been created for 32 and 64 bit machines. These do not support satellite tracking and won't until I obtain a gfortran version of the SPICE code. March 18, 2010. RCW

• The use of gfortran revealed some problems with unSAVEd variables in some optimization routines. Apparently g77 retains variable values between calls and gfortran does not. Fixed for the CELLS and CSUB routines. Note that the CSUB option seems to be broken for now for reasons that I have not yet tried to chase down. March 18, 2010 RCW.