This document is a draft of required/desired functionalities for IRISpy.
An object to read in, store and manipulate IRIS spectrograph files (raster or sit-and-stare) from the same OBS campaign.
IRIS spectrograph data have the following characteristics:
- They contain a number of non-contiguous 3D data cubes, one for each spectral window in OBS.
- Data from sit-and-stare OBS are stored in a single FITS file with a single set of WCS values.
- Data from rastering OBS are broken into multiple files, one for each raster scan. Each file has its own set of WCS values.
- Data cubes have the following dimensions:
- spectral dimension (Dimension along which wavelength varies.)
- slit dimension (Dimension along length of slit.)
- raster dimension (Dimension along which time and/or raster slit position varies.)
- Data cubes have a pseudo-dimension labelling the spectral window. This is referred to as a pseudo-dimension because data may want to be selected or manipulated based on this attribute. But the values along this pseudo-dimension are not numerical.
- The spectral dimension is associated with the following coordinate
systems:
- wavelength (or physical equivalent, e.g. frequency.)
- The slit dimension is associated with the following coordinate
systems:
- helio-projected latitude. This is dependent on the helio-projected longitude also associated with the raster dimension.
- slit pixel number
- The raster dimension is associated with the following coordinate
systems:
- helioprojected longitude. This is dependent on the helio-projected latitude also associated with the slit dimension.
- time
- raster scan position number, i.e. the slit position number within the scan sequence. Therefore spectra taken at the same position within each raster scan will have the same value. This makes it possible to identify spectra taken at nearly the same longitude (if solar rotation tracking off) or over nearly the same feature on the solar (if solar rotation tracking on).
- Read in and store IRIS raster/sit-and-stare files of the same OBS campaign.
- Store non-contiguous data cubes, one for each each spectral window.
- Truncate all spectral windows using criteria applied to common dimension(s) with a single command. Common dimensions are the slit and raster dimensions. The spectral dimension is not common. Spectral window is a common pseudo-dimension.
- Dimensions should be indexable by the coordinate systems associated with them.
- Both data and coordinate systems should be unit aware (desired).
- Convert value and unit of data from DN to photon counts and vice versa using gain and yield info.
- Convert value and unit of data from DN (or photon counts) to DN/s (or counts/s) by applying an exposure time correction.
- Calculate fractional uncertainty of intensity data.
- Convert intensity data from DN to radiance.
- Calculate and apply wavelngth correction due to orbital variation.
An object to read in, store and manipulate IRIS slit-jaw imager (SJI) data files from the same OBS campaign.
SJI provides spatially resolved images of the chromosphere and transition region in a number of passbands.
Lijia's suggestions based on the ‘User Guide to Data Analysis’: (http://iris.lmsal.com/itn26/itn26.pdf)
*routines that load data and display header in p14,32,33 of ‘user guide to data analysis’, ex. d=iris_load(my file) and d->showiness
*data query from remote sever if not already implemented in sunpy
*iris_get_mg_features_lev2 mg ii line variable component fitting, ‘moment’ fitting, gaussian fitting, red-blue asymmetry analysis (p38)
*rest wavelength calibration
*density calculation
*quick look tool like iris_xfiles
Combines information from the raster and the slitjaw into one object. Should either be able to output IRISSJIMap and IRISRaster objects or contain them. Should be able to open IRIS Level 3 data and permits the user to explore the connection between the slit-jaw imagers and the spectral data with one object.
- Make IRISSJIMap a subclass of SunPy Map or MapCube.