Runtime errors on some systems when reading Atkinson tables

[pmpowers-usgs]

from @JMAltekruse and @AHajiSoltani

When I try to run the CEUS scripts on Linux, I get a fortran runtime error, which, based on the log file occurs after it preps the TP05 subroutine. Here’s the error:

At line 7975 of file src/hazgridXnga13l.f
Fortran runtime error: Substring out of bounds: lower bound (-13) of ‘fname’ is less than one

The line that fails is in routine GailTable, which appears to be used for AB08, AB06’, and Pz11 only. The error is on line 7975:
write(*,9) fname(i2:i3), itype(i)
If I change line 7927 from:
i2=index(fname,'.rev')-13 !names changed to *.rev when including 1.5s coefs
to this (which is consistent with this line in hazFXnga13l.f, for example
i2=max(1,index(fname,'.rev')-13) !names changed to *.rev when including 1.5s coefs
the CEUS files run without errors in the log file or terminal – I’m not certain what the output file is supposed to look like.

Note: when compiling hazgridXnga13L.f using the unmodified make file, I see this error:

src/hazgridXnga13l.f:469.19

      common/gnome/name
                   1
Warning: Named COMON block ‘gnome’ at (1) shall be of the same size

My fortan is basic at best, but I don’t know where the gnome common block is set, or why in this instance it contains ‘name’ and in the GailTable routine the gnome common block contains ‘fname’, and neither variable seems to be assigned a value anywhere that I’ve seen.

cc: @ashumway-usgs

[pmpowers-usgs]

These appear to be resolved in current versions of the code. The *.rev tables with 1.5s values have not yet been ported to nshmp-haz.