Bug Description
When running with LVT running mode: "Data intercomparison" and "NLDAS2" as the second data type,
LVT gives an error "NLDAS2 data directory: not defined" even though it is defined in the config file.
This happens the SECOND time it LVT tries to find this line. In NLDAS2_dataMod.F90, it loops
k=1,i where i is 2 (the second data type) and it fails on the second attempt.
Putting the same line in the config file twice did not help.
Some of the other readers do not have this loop, and just set the variable (e.g. Merra2obs(i)%odir)
for the index i (which is 1 or 2 but does not change within the subroutine).
Steps to Replicate
See above.
Trying to fix but not sure what the intended behavior is. I don't know why the NLDAS2obs() structure is an array in the first
place. Is it intended that you can have NLDAS2 for both the first and second data types? If so, there needs to be an example
of how to set up the config file.
Bug Description
When running with LVT running mode: "Data intercomparison" and "NLDAS2" as the second data type,
LVT gives an error "NLDAS2 data directory: not defined" even though it is defined in the config file.
This happens the SECOND time it LVT tries to find this line. In NLDAS2_dataMod.F90, it loops
k=1,i where i is 2 (the second data type) and it fails on the second attempt.
Putting the same line in the config file twice did not help.
Some of the other readers do not have this loop, and just set the variable (e.g. Merra2obs(i)%odir)
for the index i (which is 1 or 2 but does not change within the subroutine).
Steps to Replicate
See above.
Trying to fix but not sure what the intended behavior is. I don't know why the NLDAS2obs() structure is an array in the first
place. Is it intended that you can have NLDAS2 for both the first and second data types? If so, there needs to be an example
of how to set up the config file.