NIKA Data Reduction

Go back to the NIKA Main Wiki

Getting ready

To start the IDL session for the real time analysis, open a terminal and type:

   $ ssh -X nikaw-13@mrt-lx1
   $ ssh_sami
   > idl 

Shortly after a scan is done, the NIKA scientific data, the xml information generated by PAKO (and the AntennaIMBfits shortly after) are written on SAMI and can be processed. There are two types of observation: the “science” scans and the “calibration” scans. There is a specific routine to analyze each type of scan.

Calibration scans

All calibration scans are reduced by the same routine:

IDL> rta_reduce, day, scan_num

day is in the format YYYYMMDD, e.g. 20140218, scan_num is a simple integer.

Informations or actions to be taken by the observer appear on the plots and explicitely in IDL terminal window:

cont_pointing_cross

• The plots show 1D profiles of the azimuth or elevation scans at both wavelenght for the reference detector and a gaussian fit of the source along the timeline. The displayed offsets refer to the reference detectors and are not meant to be passed to PAKO.
• The maps show the source, a cross at the location of the reference pixel and gaussian fit of the source
• In the terminal window, the observer should follow the instruction (e.g.): B 2mm (MAP) (for PAKO) SET POINTING 3.0 0.7
• the pointing reference is the 2mm channel.
• If the sky conditions were correct, instructions at 1 and 2mm should be in good agreement.
• if a glitch corrupts the maps but the 1d profiles are ok, the user can follow the instruction B 2mm, : (TOI) (for PAKO) SET POINTING 3.0 0.7
• In good conditions, all instructions agree to better than 1 arcsec.

• If the pointing is far off, the default reference pixels may not see the source and the timelines may look bad. In the terminal window, you may read the ID of the closest kids to the source and pass them to rta_reduce as keywords e.g. rta_reduce, day, scan_num, numdet1=5, numdet2=494

cont_pointing_liss

• The routine displays one map for each band, an estimation of the flux where the source is fit (red), an estimation of the flux at the center of the map (green). It also provides a quick estimate of the NEFD and the map RMS for 10arcsec equivalent pixels and the values of Tau derived by NIKA.
• The observer should follow instructions given in the terminal, e.g.: B 2mm, : (MAP) (for PAKO) SET POINTING -4.7, 1.4

cont_focus_track

• you may change the reference kids as before e.g. rta_reduce, day, scan_num, numdet1=5, numdet2=494

• By default, optimize the focus at 1mm

cont_skydip

• No action is requested from the observer other than checking that the displayed histograms look reasonable. These results will be used by the NIKA team later on offline.

cont_focus_liss

• As soon as the data of a scan of this sequence are on Sami's disk, you may run rta_reduce, day, scan_num

• When at least 3 of the 5 scans are done, you may call focus_liss, day, [scan_num1, scan_num2, scan_num3] to have a first guess of the optimal focus

• When the 5 scans are done, call focus_liss with the complete list of scans and follow instructions in the terminal window and by default, optimize the focus at 1mm

cont_beammap

• Instead of rta_reduce, please call otf_geometry, day, scan_num

• The top plots compare the current derived beam maps to the reference ones
• The bottom plots display the distance between the current beam parameters and the reference ones in terms of centroids positions, FWHM and absolute calibration.

Science scans

Point source maps

• rta_reduce, day, scan_num

• If you want to coadd several scans, run combine_scans, scan_list. Note however that to allow to combine scans from different days, scan_list is here defined slightly differently, it is an array of strings of the format YYYYMMDDsScan_num, e.g. '20140127s290'

Diffuse emission

• rta_reduce, day, scan_num, /diffuse
• to combine several scans, use combine_scans in the same way as for point source maps

Data products

Data products for the 1st NIKA pool

Neel pipeline

The following diagram shows the NIKA data flow from the IRAM 30m telescope to the observers' accounts.