Page Comparison

...

Flow

Image Added

Chips grids a uvfits file, producing multiple .dat that can be used for crosspower and 1d power spectrum analysis.

The naming convention for .dat files usually contains the polarisation (xx or yy) and unique extension (ext) to identify that file.

.dat files from multiple uvfits files can be combined together.

arguments:

• ext - unique name for the grid
• eorband - either 0 (low, 139-170MHz), 1 (high, 167-198MHz)
• eorfield - either 0 (radec=0,-27 deg), 1 (radec=60,-30 deg)
• nchans - number of channels to analyse
• freq_idx_start - first channel number to analyse
• pol - polarization to analyse (xx or yy)
• period - integration time in seconds (e.g. 8.0)
• chanwidth - frequency resolution in Hz (e.g. 80000)
• lowfreq - first frequency in the data in Hz
• umax - max uvw (default 300)
• addsub - add (0) or subtract (1)
• nbins - number of power spectra bins (e.g. 80)
• bias_mode - one of 0/10/11/12 e.g. 0

environment variables:

• DATADIR
• INPUTDIR - where to look for input files
• OUTPUTDIR - where resulting files are written to
• OBSDIR
• OMP_NUM_THREADS - number of threads to use

binaries

• gridvisdiff - calculate the beam weights for an MWA observation
• prepare_diff - combine data over frequency
• combine_data - combine data over multiple gridded sets
• lssa_fg_simple - compute the LS spectral power (no kriging) using diff, tot and weights uvf binary files

Creating grids

gridvisdiff $uvfits $obsid $ext $eorband -f $eorfield
# for pol in xx yy; do
prepare_diff $ext $nchans $freq_idx_start $pol $ext $eorband -p $period -c $chanwidth -n $lowfreq -u umax
# this produces {vis_tot,vis_diff,noise_tot,noise_diff,weights}_${pol}_${ext}.dat in $OUTPUTDIR

Combining grids

combine grids with extensions ext1, ext2 together into a grid with extension $group_ext

# for pol in xx yy; do
# create a file containing all the exts to combine together prefixed with "${pol}."
export combinelist=combine_${pol}.${group_ext}.txt 
for ext in \
    "ext1" \
    "ext2" \
; do
    echo ${pol}.${ext} | tee -a ${combinelist}
done
# ensure {vis_tot,vis_diff,noise_tot,noise_diff,weights}_${pol}.${ext}.dat all exist in $OUTPUTDIR

combine_data ${combinelist} $nchans "${pol}.${group_ext}" $addsub
# this produces {crosspower,residpower,residpowerimag,totpower,flagpower,fg_num,outputweights}_${pol}_${bias_mode}.${ext}.dat in $OUTPUTDIR

Compute Power Spectra

# for pol in xx yy; do
lssa_fg_simple $ext $nchans $nbins $pol $maxu $ext $bias_mode $eorband

Plotting power spectra

see: https://github.com/JLBLine/plot_CHIPS

Garrawarla (new)

# load the chips module
module use /astro/mwaeor/software/modulefiles
module load chips/cmt

# optional optimization: do everything in nvmetmp, need to request enough --tmp from slurm
mkdir /nvmetmp/deleteme
cd /nvmetmp/deleteme

export DATADIR="$PWD"
export INPUTDIR="$PWD/"
export OUTPUTDIR="$PWD/"
export OBSDIR="$PWD/"

Garrawarla (old)

CHIPS is available on Garrawarla

...

See (page to be written soon) for a full overview of the plotting commands available. But a quick example of how to make a 2D cross-power spectra using outputs of the OzStar example is below. Note the output tag for the files is a combination of the --output_tag and --uvfits_dir options that were fed to run_CHIPS.py :

...

this will produce a plot called chips2D_yy_test_8dec_test_ozstar_chips_crosspower.png which looks like this:

this This is some random EoR1 data I found on OzStar so I have no idea of the quality or what peeling was done, but hey, it's a power spectrum!

...