You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
In geogrid the transformations between geographic and computational coordinates for lat-lon projections appear to apply the rotation by stand_lon twice. This occurs in the subroutines set_cassini, llij_casini and ijll_cassini. When unravelled, the subroutine rotate_coords(ilat,ilon,olat,olon,lat_np,lon_np,lon_0,direction), for direction -1 (geographic to computational coordinates), successively applies the following transformations:
A rotation about the z-axis by lon_np,
A rotation about the new (rotated) y-axis by 90 - lat_np
A rotation about the new (rotated) z-axis by lon_0
When direction == 1 the reverse transformation is applied.
When applied in llij_cassini for example,
the second line of code, after rotate_coords is called, has the effect of rotating again by stand_lon. A similar thing occurs in subroutine set_cassini, and the reverse occurs in ijll_cassini.
I am not sure that this is a serious issue. When pole_lat, pole_lon and stand_lon are set according to recommendation in Section 3 of the Users Guide, the centre of the computational domain is over the equator, which is important, but is rotated E-W and does not lie over the 180 deg meridian. I think that fixing this issue might break many peoples namelists and post-processing code. I am not sure if this spurious rotation has any other implications, so I am bringing it to your attention.
The text was updated successfully, but these errors were encountered:
In geogrid the transformations between geographic and computational coordinates for lat-lon projections appear to apply the rotation by stand_lon twice. This occurs in the subroutines
set_cassini
,llij_casini
andijll_cassini
. When unravelled, the subroutinerotate_coords(ilat,ilon,olat,olon,lat_np,lon_np,lon_0,direction)
, for direction -1 (geographic to computational coordinates), successively applies the following transformations:When direction == 1 the reverse transformation is applied.
When applied in
llij_cassini
for example,the second line of code, after
rotate_coords
is called, has the effect of rotating again bystand_lon
. A similar thing occurs in subroutineset_cassini
, and the reverse occurs inijll_cassini
.I am not sure that this is a serious issue. When
pole_lat
,pole_lon
andstand_lon
are set according to recommendation in Section 3 of the Users Guide, the centre of the computational domain is over the equator, which is important, but is rotated E-W and does not lie over the 180 deg meridian. I think that fixing this issue might break many peoples namelists and post-processing code. I am not sure if this spurious rotation has any other implications, so I am bringing it to your attention.The text was updated successfully, but these errors were encountered: