Add +proj=mod_krovak projection method for Modified Krovak that applies to S-JTSK/05 in the Czech Republic#4021
Conversation
|
There is an inconsistency between "Krovak Modified" and "Modified Krovak". Guidance Note 7-2 mostly uses "Krovak Modified", with a few exceptions (e.g. "Modified Krovak East North") An advantage of Another more convoluted method is to keep only |
yeah, I considered krovak_mod at first, but there was the risk of not noticing the _mod suffix, so I switched the mod_krovak to make things obvious. That said, if you feel that krovak_mod is preferable, I could switch to that
I also considered that, but same reason, it is easy to miss it (and if people would copy&paste PROJ strings with +proj=krovak +modified to older PROJ releases, the +modified would be silently ignored, whereas +proj=mod_krovak will just error out). Furthermore, technically/formally, modified krovak takes extra parameters (the correction coefficients), which are currently hard-coded, because there's no point in requiring users to type them, given that the S-JTSK/05 is the single instance where this projection method is typed. I was also wondering if we should set the default values of x_0 and y_0 to 5000000, which is what is used for S-JTSK/05, but was undecided, so I kept the 0 default value. |
mwtoews
left a comment
There was a problem hiding this comment.
Re: "Krovak Modified" and "Modified Krovak". Looking further, there is no obvious "correct" form.
- In Czech it is "modifikovaném Křovákově", never the other way around
- English would normally prefer "Modified Krovak" too
So sure, let's stick with "Modified Krovak". See related suggested changes...
…es to S-JTSK/05 in the Czech Republic Validated with the test point given in EPSG Guidance Note 7-2 Also validated with CUZK online calculator at https://geoportal.cuzk.cz/(S(g4ipaut0ckzb5lellu153lvf))/Default.aspx?head_tab=sekce-01-gp&mode=TextMeta&text=wcts&menu=19 Unfortunately this calculator does not offer 'elementary' transformations, and cannot just validate Modified Krovak alone. The best we can do is to use ETRS89 <--> S-JTSK/05 / Krovak modified, which involves a Helmert transformation as well for the datum transformation. * Validation of reverse modified Krovak: Online calculator: - Souřadnice/input: -5568990.91 -6050538.71 100 - Source CRS: S-JTSK/05 + Bpv (-Y-XH /east-north) - Destination CRS: ETRS89 (BLh / DEG) - Výsledek/result: 50.208297081 16.848326835 143.762 With PROJ: echo -5568990.91 -6050538.71 100 | bin/cs2cs -d 9 EPSG:5516 EPSG:4937 --3d 50.208297081 16.848326833 143.143233569 Note the different on longitude is totally neglectable: 0.1 mm, as computed by: $ echo 50.208297081 16.848326835 50.208297081 16.848326833 | bin/geod -F "%.4f" -I +ellps=GRS80 -90d 90d 0.0001 and that when outputing Krovak modified, the online calculator displays Y/X with a precision of 1 cm. The difference on Z is significant, but can be explained from the fact that the online calculator using Baltic Height for S-JTSK/05 and thus apply a Baltic Height<-->ETRS89 geoid that we don't apply here. * Validation of forward Krovak using same point: Online calculator: - Souřadnice/input: 50.208297081 16.848326835 143.762 - Source: ETRS89 (BLh / DEG) - Destination: S-JTSK/05 + Bpv (-Y-XH /east-north) - Výsledek/result: -5568990.91 -6050538.71 100 With PROJ: $ echo 50.208297081 16.848326835 143.762 | bin/cs2cs -d 2 EPSG:4937 EPSG:5516 --3d -5568990.91 -6050538.71 100.62
Validated with the test point given in EPSG Guidance Note 7-2
Also validated with CUZK online calculator at https://geoportal.cuzk.cz/(S(g4ipaut0ckzb5lellu153lvf))/Default.aspx?head_tab=sekce-01-gp&mode=TextMeta&text=wcts&menu=19
Unfortunately this calculator does not offer 'elementary' transformations, and cannot just validate Modified Krovak alone. The best we can do is to use ETRS89 <--> S-JTSK/05 / Krovak modified, which involves a Helmert transformation as well for the datum transformation.
Validation of reverse modified Krovak:
Online calculator:
With PROJ:
Note the different on longitude is totally neglectable: 0.1 mm, as computed by:
and that when outputing Krovak modified, the online calculator displays Y/X with
a precision of 1 cm.
The difference on Z is significant, but can be explained from the fact that
the online calculator using Baltic Height for S-JTSK/05 and thus apply a
Baltic Height<-->ETRS89 geoid that we don't apply here.
Validation of forward Krovak using same point:
Online calculator:
With PROJ: