Proposal for a FITS convention for region based "skymaps" #170
Conversation
adonath
requested review from
registerrier,
lmohrmann,
bkhelifi,
TarekHC,
maxnoe and
jknodlseder
|
I have requested reviews from the people, that have been active in this repo for the last time. Possibly I forgot someone... |
|
For now I have added it to the skymap section, however it could possibly be a sub-section of spectra as well. The "skymap" section seemed more natural to me, as the format definition relies on the "bands HDU" and has the support for region definitions as well. So it's a format proposal for an N-dimensional map with a single spatial bin, defined by the region. The main argument for me is that the format does not strictly apply to a "spectrum" in a classical sense. E.g. it can be used as well to store a reduced instrument response function, e.g. a PSF with a true energy and rad axis or an RMF matrix with a true and reconstructed energy axis. |
| * Specify how the non-spatial bins are spaced and recommended to be interpolated. | ||
| Possible values are: | ||
| * ``"lin"``: linear spacing e.g. used for time or offset axes | ||
| * ``"log"``: logarithimic spacing e.g. used for energy axes |
There was a problem hiding this comment.
Does it matter here if log is log10 or ln? log is ambiguous but I guess for interpolation it is not important
There was a problem hiding this comment.
In yesterdays aswg call, the need for some trigonometric scheme, e.g. cos for zenith dependence was discussed
There was a problem hiding this comment.
No, for the interpolation the type of the log doesn't matter. I agree a trigonometric scheme is useful as well, but probably mostly for DL3. For the "skymaps" (DL4) the data is typically binned or project in sky coordinates already.
I think currently the information on the bin spacing is not part of the "gadf" DL3 specification. E.g. in Gammapy we add it "by hand": https://github.com/gammapy/gammapy/blob/master/gammapy/irf/io.py#L5 In general it would be a useful addition!
adonath
force-pushed
the
region_sky_maps
branch
from
25a60e4
to
e173046
Compare
|
Following up on a private discussion with @registerrier (see also the PR with the prototype implementation in Gammapy gammapy/gammapy#3231), I updated the proposal as following:
This allows to store multiple "ND spectra" with the same axes specification and let them share a single bands HDU. |
|
The addition of this format is indeed useful. The OGIP format only supports parts of the use cases that this supports as it is limited to purely spectral information (counts or counts rate as a function of channel). It is not totally clear to me where this belongs, though. This convention and the one describing skymaps (https://github.com/open-gamma-ray-astro/gamma-astro-data-formats/blob/master/source/skymaps/index.rst) are not necessarily fully connected to gamma-ray format. They simply describe a FITS storage convention for multidimensional quantities. Should we describe them here or propose them as a set of conventions for the registry of FITS conventions that is maintained by the IAU FITS working group? For instance in the section "Conventions applicable to specific types of data" . Any opinion about this @cboisson and others? |
In Gammapy we already use the GADF sky maps formats for HEALPix and WCS. In Gammapy there is now the use-case to handle "spectral" data, that was integrated / averaged over a certain region with additional extra axes, e.g. time.
This pull request proposes a new format for these "region based" maps. Here is the core of the proposal:
I hope the general need for such as format is uncontroversial, as it will allow to store e.g. time dependent counts spectra together with complex region definitions. Already now there are many use-cases where this format would be useful.
Currently the description is rather short, but as noted above the proposed format relies almost exclusively on already defined HDU definitions.