search

CARTAvis / carta

To CARTA users, this repo holds the CARTA release packages. Please use this repo to log bugs and feature requests. These will be triaged by the development team and prioritised as necessary in the development cycles.
19 stars 0 forks source link

Documentation: Requested changes, and question on velocity reference systems #56

Open jehibbard opened 3 years ago

jehibbard commented 3 years ago

Is github the proper place to request changes to the CARTA documentation? It looks more like a software ticketing system than a user helpdesk for such requests, but let me know what the project prefers.

Document request #1: In the section on "Match images in world coordinates" (https://carta.readthedocs.io/en/latest/image_visualization.html#match-images-in-world-coordinates) you should mention how to match cubes either in frequency, radial velocity, or channel (the only way I could find to do this is by using the "Preferences", under Global settings, and the screenshot in the documentation for this does not extend to show this part of the settings).

Question on Velocity References Systems Speaking of this, while doing a tutorial on CARTA for NRAO, I found that sometimes the products in the ALMA archive (may apply to other archives) have different redshifts for the same object (or one product has a redshift, another doesn’t, or the line is a doublet and some average is used....), which makes spectral registration not work properly (especially if registering by Vrad). Is there a way to overwrite the header “on the fly” when exploring remote products (i.e., by entering you own redshift or rest frequency for each file)? I couldn’t find anything in the documentation. If this is not there, let me know and I'll submit a helpdesk ticket for new feature.

Document request #2 (from NRAO helpdesk ticket from a user): In the section on the statistics widget (https://carta.readthedocs.io/en/latest/analysis_tools.html#statistics-widget): say how flux density is calculated. e.g., for images in units of Jy/beam flux is calculated as [sum of all pixels, in Jy/beam ] [Region area, expressed as number of beams] where the Region area, expressed as number of beams is the region area (=Npix dx" * dy") divided by the area of the synthesized elliptical beam (calculated by integrating under an elliptical beam = pi FWHMx" x FWHMy" / 4 ln 2) (or if its calculated using the CASA imstat tool, say that so we know where to go for more information) (if other units are calculated simply as the Sum of all pixels, then say so). Are pixels counted if their center is in the region, or is there an attempt to properly calculate fractional pixels?

kswang1029 commented 3 years ago

@jehibbard

Question on Velocity References Systems Speaking of this, while doing a tutorial on CARTA for NRAO, I found that sometimes the products in the ALMA archive (may apply to other archives) have different redshifts for the same object (or one product has a redshift, another doesn’t, or the line is a doublet and some average is used....), which makes spectral registration not work properly (especially if registering by Vrad). Is there a way to overwrite the header “on the fly” when exploring remote products (i.e., by entering you own redshift or rest frequency for each file)? I couldn’t find anything in the documentation. If this is not there, let me know and I'll submit a helpdesk ticket for new feature.

Regarding this, please help to file an issue in the frontend repo.

kswang1029 commented 3 years ago

Document request #2 (from NRAO helpdesk ticket from a user): In the section on the statistics widget (https://carta.readthedocs.io/en/latest/analysis_tools.html#statistics-widget): say how flux density is calculated. e.g., for images in units of Jy/beam flux is calculated as [sum of all pixels, in Jy/beam ] [Region area, expressed as number of beams] where the Region area, expressed as number of beams is the region area (=Npix dx" * dy") divided by the area of the synthesized elliptical beam (calculated by integrating under an elliptical beam = pi FWHMx" x FWHMy" / 4 ln 2) (or if its calculated using the CASA imstat tool, say that so we know where to go for more information) (if other units are calculated simply as the Sum of all pixels, then say so). Are pixels counted if their center is in the region, or is there an attempt to properly calculate fractional pixels?

If my memory is right, we adopt the same approach as CASA-imstat. @pford could you confirm please?

kswang1029 commented 3 years ago

@jehibbard regarding documentation request 1 and 2, I will improve the user manual accordingly.

pford commented 3 years ago

For non-HDF5 images, we use casacore LatticeStatistics for all statistics calculations, the same code used for imstat. CARTA has HDF5-specific code for statistics since per-plane and per-image stats are stored in the image.

jehibbard commented 3 years ago

@kswang1029 - more generally, would you like me to submit documentation suggestions this way? And for questions that come in from the NRAO user base - should I have the helpdesk filter those through me, or point them to this ticketing system?

It looks like you already opened the ticket for modifying the velocity reference. I'll make any comments there.

kswang1029 commented 3 years ago

@jehibbard i suggest we post issues and feature suggestions here.

I just filed the issue for changing how spectral axis should be matched. Please help to file the issue for changing reference frequency for velocity conversion on the fly.