pablobuitron
commented
1 year ago I compiled my local CORSIKA without the CURVED option. When running the arti analysis script, I encountered conflicts due to previous compilations related to CORSIKA-LAGO, which takes "priority during compilation" on my machine. Perhaps a solution would be to have "local" versions in DOCKER, meaning a container that is compiled with CURVED and another one compiled without CURVED. Here's the link to the arti repo where it states that CORSIKA-LAGO already comes with this option compiled, and I'm also providing a brief summary of what the option does, for easy reference without having to always open the manual and search for it:
Within the CURVED atmosphere option for large zenith angles above ≈ 70◦
the Earth’s atmosphere is no longer assumed to be completely flat (as in the standard version for smaller zenith
angles). Rather the atmosphere is replaced by a ’sliding plane atmosphere’. Each times the
horizontal displacement of a particle exceeds a limit of 6 to 20 km (depending on altitude), a
transition to a new, locally plane atmosphere is performed. By these means the advantages of the
simpler transport formulas within a planar atmosphere are combined with the faster simulation
speed by avoiding the lengthy and more complicated treatment by using a true spherical system. Because of technical reasons only one observation level may be specified in the CURVED
option and the X and Y coordinates have a special definition in order to be able to reconstruct
the 3 coordinates (x,y,z) in a cartesian frame.
In the CURVED option the ionization energy loss, deflection within the Earth’s magnetic field,
and the generation of Cherenkov photons is enabled for charged hadronic primaries on their
path between entering the atmosphere and the first interaction. (The deflection is disabled in
the standard version without using keyword TSTART) The arrival time refers to the
start at the margin of the atmosphere, which is indicated by a negative value of element 7 of the
event header block. The NKG formulas do not take into account the curvature of the Earth’s surface. Therefore the
NKG output is suppressed in the CURVED option. As in the CEFFIC option the atmospheric absorption is treated only in a manner suited for planar atmospheres, you should not combine the CURVED option with CEFFIC. The CURVED option needs the atmospheres in the 5-layer model rather than interpolated from ATMEXT tables. Therefore in the CURVED option AATM, BATM, CATM values are fitted to the tabulated atmospheres and the table interpolation is disabled. To obtain this program version the CURVED option has to be selected when extracting the FORTRAN code from the source file.
caredg
@pablobuitron claims that the LAGO version of corsika is compiled with the CURVED flag. HOwever we need to be absolutely sure about that. What is the evidence for this claim beside some text written somewhere? What is the difference when this flag is not used (in the practical sene)? I.e., if you have a local version of corsika and compile it with and without this flag, do you see any difference when asking fluxes with angles between 70 to 90 deg?