Observing with NIKA2

Go to the NIKA2 main page.

Computer terminals for observations

NIKA2 pool observers have at their disposal three terminals for observing operations (also accessible remotely via a VNC server). Terminal mrt-remote:32 is devoted to observing tasks with NIKA2 and the terminal mrt-remote:30 is reserved for EMIR backup observations. In the mrt-remote:32 terminal the observer will find up to 7 active desktops. The desktop 1 contains a terminal with a running session of the observer user interface to the control system of the IRAM-30m telescope, nicknamed paKo (see next Section), the paKo display (light green window) with relevant information on the status of the telescope, an auxiliary terminal, and the Antenna Data Stream. A screenshot that illustrates what it looks like is shown below:

Starting paKo

To start the paKo session for regular observations, log into the pool account "nikaw-23" (ask the AoD/operator for the login information), open a terminal and type:

$ ssh -XY nikaw-23@mrt-lx1.iram.es
$ goNIKAnew
$ PAKO> set project [nikaw-23 or test]
$ PAKO> set dosubmit yes

Now you are ready to start observations.

Observations queue

To check what is currently in the observation's queue, open a terminal and type:

$ ssh -XY nikaw-23@mrt-lx1.iram.es
$ observationQueue &

This will open a file browser with a list of all the observation's files commanded, which are waiting to be executed by the telescope's system. To remove any observation's file from the queue, right-click on the corresponding xml file and select "delete". The display will be automatically refreshed.

Starting XEphem

XEphem is the software currently used at the IRAM 30m telescope to keep track of the position of astronomical sources on the sky. To start XEphem, open a terminal and type:

$ slogin mrt-lx3
$ xephem &

On the main Xephem window, click on the tab "RT" (for real-time) to activate source tracking in this software. Then, click on the tab "View" and open the "Sky View".

A map similar to the one shown below will appear. The "telescope icon" shows the coordinates the telescope is pointing at. The filled white circles are the pointing sources. The size of the circles is proportional to the flux of the target. Use the sliders on the border of the map for scaling the field of view.

To load the sources catalog of your project, click on the tab "Data" and go to "Files". In the new window go again to "Files" and select the catalog of the project that you plan to observe.

Ask the AoD or the Operator for further details.

Starting PIIC

PIIC is a quicklook (QL) monitor to process continuously the NIKA2 data. To start it, first ask the AoD to start the production of NIKA2-imbfits required to reduce the data with PIIC.

$ ssh -XY nikaw-23@mrt-lx3
$ cd ql
$ gagpiic
$ piic @monitorX 

with X = 2pf, for 2 mm-pointing/focus monitoring only, and X = 1, 2m, 3, the instrument array to be monitored.

Starting the NIKA2 IDL pipeline

Optionally, to have the IDL pipeline continuously processing observations as they are produced, open a terminal and type:

$ ssh -XY observer@nika2-a (ask the AoD for the password)
$ IDL
IDL> nk_rta, scanID    (e.g. '20201019s132')

where scanID is the corresponding identifier of the observation (as shown in the example) in the format YYYYMMDDsXXX, where YYYY is the year, MM the month, DD the day, and XXX the scan number.

Pointing

Under standard conditions, pointing corrections of the IRAM 30m telescope should be done every hour or a bit more with good stability (see below), except around dusk and dawn when more frequent pointing corrections may be necessary.

In order to correct the pointing of the telescope in a given part of the sky choose a nearby pointing source (using e.g. XEphem) and type:

PAKO> source 0133+476 /cat * (or "iram-J2000.sou" or "pointingIRAM30m.sou", or "CalibNIKA.sou" for secondary flux calibrators)

Now launch the nkpoint script:

PAKO> @ nkpoint mode

where the keyword "mode" can adopt the values b for bright sources and f for faint sources (the l option for very faint sources that require the Lissajous pattern is no longer used). To obtain the pointing corrections, use the results given by the 2pf PIIC monitor. Supplementarily, PIIC monitors 1 and 3 show the pointing corrections based on 1 mm data.

Alternatively, you can process the corresponding scan with the IDL pipeline: "IDL> nk_rta, scanID".

Once the pointing scan is processed, enter the new pointing corrections in azimuth (PnewX) and in elevation (PnewY) shown on the IDL prompt of the NIKA2 pipeline:

PAKO> set pointing PnewX PnewY

An example of PIIC-QL main plot of a standard pointing scan is shown in the following picture. Pointing correction, a proxy of the beam size (main-beam FWHM at 2 mm wavelength), and other data are highlighted:

Alternatively, the corresponding pointing scan number can be processed by the IDL pipeline by simply typing:

IDL> nk_rta, scanID (in the form 'YYYYMMDDsX')

The pipeline will recognize a pointing scan and the correction will appear in a while.

According to Perotto et al. 2020, lower limits of beam size (FWHM) with NIKA2 are 11.1 arcsec (1 mm, arrays 1 & 3 combined) and 17.6 arcsec (2 mm; array 2). If you happen to have noticeable broader beam sizes, (i) probably a new focus correction would be needed, or (ii) the pointing source is not a point-like one in the sense of the main beam size, or (iii) the FWHM is measured on scans acquired in the late afternoon when the main dish of the antenna is warmed and unrelaxed, or (iv) anomalous refraction/high-frequency atmospheric instabilities could be affecting the pointing scan. In case of doubts, ask the AoD.

Focus

The focus needs to be monitored and it needs to be corrected online. This should be done every two hours or so (~3h during very stable night-time conditions, ~1h or less near sunrise and sunset) in the Z direction. To check the quality of the focus run:

PAKO> @ focusOTF-Z fz  

Currently, this script will carry out 5 foci measurements in Z direction within a 1.6 mm range, centered at the fz value. Telescope focus for NIKA2 is currently around 0 to +1 mm in standard conditions. To obtain the focus corrections, use the results given by the 2pf PIIC-QL monitor. Supplementarily, PIIC-QL monitors 1 and 3 also show focus corrections from 1 mm data. It gives the best focus corrections from an order-2 fitting of the observed peak flux, and two different parameters from a 2d-gaussian fit of the source used for this purpose. Under expected conditions, these values should broadly converge. If this condition is not met, the average of the two mutually consistent values could be adopted as a focus correction. An example of a typical result of a focus sequence from the PIIC-QL monitor (2 mm) looks as follows:

Conversely, if the fz value is not a good pivot for the focus sequence or under stable weather conditions, the focus results would be like the plots in the figures below:

In general, after applying the new corrections, good fitting results are expected, but it is always safe to repeat pointing and focus observations once again to confirm whether the beam geometry is properly measured.

Alternatively, the corresponding scan number can be processed by the IDL pipeline:

IDL> nk_rta, scanID1    (for measurement 1)
IDL> nk_rta, scanID2    (for measurement 2)
IDL> nk_rta, scanID3    (for measurement 3)
IDL> nk_rta, scanID4    (for measurement 4)
IDL> nk_rta, scanID5    (for measurement 5)

Then, all measured foci need to be processed together by the "nk_focus_otf" routine of the IDL pipeline to perform the best focus fit in terms of the measured Flux, beam FWHM, and beam ellipticity of the KID's. This can be achieved by running the following instruction in the IDL prompt:

IDL> nk_focus_otf,'YYYYMMDDs'+strtrim(SCANNUMBER+indgen(5),2)

where the ScanID is in the format YYYYMMDDsSCANNUMBER (e.g. '20151019s132'). Once the "nk_focus_otf" routine has finished, check the focus correction values displayed in the IDL prompt under "Fluxes" for the three arrays (A1, A2, & A3) and calculate the average value that will be used to correct the focus. Enter the new focus value (Fnew) as follows:

PAKO> set focus Fnew 

Daily calibrations

Beam maps

This kind of scan is to be done once a day or every two days. It takes ~25 min, so to avoid accounting for the time spent on your project think about changing the project to:

PAKO> set project nikaw-23

Beam maps are designed to ensure the source is moved over all the detectors of the array, in order to characterize and calibrate them (field of view geometry, flat field, stability, etc). The aim is to calculate the actual pixel offsets in the focal plane (see figure below). Low opacity and good stability are mandatory, and Uranus, Mars (if apparent size is < 10”), or 0316+413 (3C83) can be used for that. To obtain a beam map go to a primary calibrator, get focus/pointing corrections, and launch the beammap1scan99sub.pako script in the normal paKo session:

PAKO> @ beammap1scan99sub  (to produce 99 sub-scans, standard paKo version)
OPTION = a   (ENTER OPTION: a = Az. scan, e = El. scan, l = lower than 60° in El., h = between 60° and 70° in El.)  
W-OTFMAP       /TOTF,  value 12.0 implies /speed 65.0 outside standard range 0.0 to 60.0
W-OTFMAP,  WARNING--CONDITION: Elevation must be less than 69.03 [deg]
OTFMAP 780''x 470.4'' (step 4.8) 
      > Number of subscans: 99 
      > Scanning speed: 65 arcsec/sec 
      > Time per subscan: 12 sec 
      > Execution time: 23.1 min           

Don't forget to set back to your project once the beam map is done!

Flux calibration scans

If conditions are stable and the beam sizes are close to the expected values, get flux calibrations during the day and the night as possible, taking advantage from a focus sequence on primary (Mercury -when is not so close to the Sun-, Mars -if the apparent size is < ~10 arcsec-, Uranus, Neptune) or secondary calibrators (NGC7027, NGC7538, CRL2688, CRL618, MWC349, W3OH, ALFORI in the CalibNIKA.sou catalogue). The AoD will ask the observer for this kind of scan, which takes ~4 min:

PAKO> set project nikaw-23

To run a flux calibration with NIKA2, select the desired standard source and type in paKo:

PAKO> @ calib_1scan

Skydips

This kind of scan is to be done two to four times a day. It takes ~6 min, so to avoid accounting for the time spent on your project think about changing the project to:

PAKO> set project nikaw-23

To run a skydip with NIKA2 type in paKo:

PAKO> @ skydip

The script will perform 11 measurements (30 [s] tracks) at different airmasses wrt. the position Az = 180°, El = 55° in the HORIZON coordinate system. The skydip takes ~6.4 minutes to be finished.

Don't forget to set back to your project once the sky dip is done!

Science targets

Observations of science targets are performed via on-the-fly (OTF). Observers may want to scan only in Azimuth, as scan directions other than Azimuthal induce instabilities/drifts especially for NIKA2, and increased noise due to: change of airmass; reaction to telescope acceleration; coupling to electric/magnetic/acoustic signals.

In any case, the prepared scripts allow for a variety of scanning directions. For example, for a 10'x15' OTF map, with a position angle of 25 degrees, a tilt angle of 0 degrees (both measured anticlockwise), in equatorial coordinates (radec, for the HORIZON coordinate system set azel), just type:

PAKO> @nkotf 10  15  25  0  20   40  radec

This command will produce a scan pattern as follows:

A sequence of several scans can be loaded using a script containing a sequence of nkotf maps. For example, the script observe_NGC4449.pako combines several 10'x15' OTF maps at different PA and tilt angles in the horizontal coordinate system (azel).

Note that in general, it is better to have the longer direction along the subscan to minimize the time loss due to overheads between subscans. Also for baseline subtraction in the data processing, it's always better to set a subscan length longer than the usable map you wish to have (typically ~1' more than the usable map; the optimal value depends on the map size and scanning speed). Scanning in azimuth is better for sky noise subtraction, but the best scanning strategy depends on your source shape and its az-el coverage during the integration time.

Atmosphere stability issues

At the IRAM 30m telescope the opacity at 225 GHz, tau_225GHz, is recorded by a dedicated radiometer and processed (see opacity monitor). From that, the precipitable water vapour, pwv = [tau_225GHz – 0.004]/0.058, can be estimated. These opacity estimations are plotted together with model (AM) predictions at different time scales, and they can be seen by accessing the https://weather.iram.es\opacity records & forecasts. This is a valuable tool for the observers/AoD for fine observation planning.

Together with opacity and the atmospheric transmission factor (modelled as exp⁡{-tau_225GHz * sin⁡[El.]}), the stability timescale of the atmosphere during the observations affects the baseline subtraction and the mean rms noise of a given scan. A rule-of-thumb that the observer can use as a quick approach is to have a look at the upper left plot that results from the PIIC-QL pipeline at 2 mm. As an empirical rough approach, if peak-to-peak variations of the timelines given in such a plot are below +/- ~6 Jy/beam, we can say that the atmosphere is stable in the sense of NIKA2 scan. For each project, a prescription about this issue is usually provided by the PIs of the projects to the observers. For the latter ones, it is a good practice to provide an estimation of the p2p variations, and complementarily the rms of the raw timelines, for selected scans in the logsheet of TAPAS. If we translate this rule to the 1 mm array, the threshold of these variations is -in general, but not always- about a factor 3 the value used for the 2 mm band. The following PIIC-QL plots are representative of high atmosphere stability. The first one corresponds to a scan for a pointing correction, and the second one to a science scan:

Contact: Angel Bongiovanni, Ioannis Myserlis, Stergios Amarantidis (NIKA2 NIKA2 Pool Management)

email: bongio@iram.es, imyserlis@iram.es, samarant@iram.es

Created: 2013.OCT.25, by Isreal Hermelo

Updated: 2017.FEB.05, by Pablo García

Updated: 2017.SEP.28, by Alessia Ritacco

Updated: 2017.OCT.23, by Samuel Leclercq

Updated: 2019.JAN.23, by Carsten Kramer

Last update: 2024.FEB.06, by Angel Bongiovanni