XPOL/EMIR Polarimetry at the 30m telescope

This site collects information relevant to polarimetric observations at the IRAM 30m telescope. The observing method XPOL makes use of the dual-polarization feature of EMIR and the cross-correlation capability of the VESPA. The method is described in the Technical Document No.2 (see below). The calibration of the instrumental phase is explained in detail in this presentation.

XPOL observations are possible in any of the 4 EMIR bands. Instrumental effects have been best characterised in the 3mm band (see Technical document No.3). The least experience exists at 0.9mm, but a preliminary Technical document (No. 4) is available upon request. Polarimetric observations of extended sources present the complex problem of contamination by the polarized sidelobes. Our current knowledge of these sidelobes is collected in the section 'Beam Maps' below. A strategy to estimate the influence of the polarized sidelobes on slightly extended linearly polarized sources is described in the scientific document No.3. Their influence on the circular polarization of an extended source is demonstrated in Scientific document No.4.

The POLAMI monitoring projects use E0/E2.

The 1mm NIKA2 polarimetry mode is currently being commissioning with a lot of supporting XPOL observations. <6-Jul-2020 CK>

News

• 18-Sep-2020: CT: The XPOL session on 1-Sep-2020 shows Vi = 3.24 +- 0.15%. Vi stays high, also after the repair of the position reference sensor (D13 -- M6N).
• 18-Sep-2020: SN: The thick vertex window was installed on September 25th, 2013. Well ahead of the NIKA2 installation. The new membrane is 1.4m in diameter compared to the old 1m diameter window.

• 17-Sep-2020: CT creates new plot of Vi between 2014 and now, showing the 3 jumps seen already in the report of 2018, but also a factor 3 increase of the scatter between XPOL sessions after the last jump, which occurred in November 2015. thum-vi-timeline-17Sep2020.pdf All later data (up to and including 1-Sep-2020) show elevated scatter of the circular instrumental polarisation Vi:

  • before step 1	Vip = -0.26 +- 0.13
    between steps 1/2	Vip = +1.09 +- 0.11
    between steps 2/3	Vip = +1.82 +- 0.06
    after step 3	Vip = +2.88 +- 0.34

The period over which these three steps occurred can only be constrained by the XPOL observations taken before and after:

• Step 1:	2-7 April 2015	Installation of new M3, M4. Positional translation of EMIR optics and XPOL calibration unit.
  Step 2:	25-Sep - 07-Oct-2015	Installation of NIKA2
  Step 3:	24-Nov - 22-Dec-2015	Installation of new E090 mixers and OMT

• 29-Jul-2020: SN reports on observations of linear and circular instrumental polarisation in the 3mm, 2mm, and 1mm EMIR bands, between about 95 and 263 GHz: instrumental-polarisation-emir-navarro-29jul2020.pdf.

• 08-Jun-2018: SN reports results of tests: We see a 2-3% (negative) Stokes V on non polarized sources. The effect is similar on position switching (moderate throw) and wobbler switching. We have a completely different Stokes V when using a 1 degree reference offset in elevation (Pc is about 0.5% and positive). That should tell us something. We have to make more tests (in two weeks time hopefully) with different reference positions. Stokes V is zero on scans with an absorber in the path (using the 1dB step trick to have a non zero Stokes I). This is contradicting an old test we made. I'm trying to retrieve the old data. The level of circular polarization has a smooth/small change while the antenna is going down in elevation. navarro-xpol-08Jun2018.pdf CT: As the 2mm channel does not show CP, these findings indicate an offset in the location or orientation of the 3mm receiver module, or, alternatively, an offset introduced in the warm optics in front of the Rx which bring the 3mm and 1mm beams together.

• February 2018: Clemens Thum & Ivan Agudo monitored the time evolution of the instrumental circular polarisation. The 3mm instrumental circular polarisation has increased in 3 steps of roughly 1% each one: after the installation of new receiver cabin optics in April 2015, after the installation of NIKA2 in October 2015, and after the installation of the 3mm OMT end of November 2015. thum-xpol-v-stokes.pdf

• 17-Jun-2017: Santiago Navarro reports on observations of Mars. The instrumental circular polarization is of the order of 2% for the E090 and doesn't change significantly with the presence of the D12 dichroic. For the E150 band we don't see signs of instrumental circular polarization, with or without the D12 dichroic inserted. navarro-22jun2017-Xpol_E090_E150.pdf

• 6-Jun-2017: Marta Alves reports on 1mm CN 2-1 XPOL/Zeeman observations of Orion A. These observations and follow-up work indicate that the instrumental polarization is very difficult to characterise and calibrate.

• "The strong instrumental polarization of V-Stokes is something that we know from the beginning of 2016. It was first found from our Sgr A* monitoring program by inspecting the data from Mars, and was reported to Santiago Navarro. The effect is stronger as the observing frequency decreases. At 86 GHz reaches a level of ~3%, whereas at 150 GHz is ~2%, and at 230 GHz is not really evident from our Mars data." <Ivan Agudo, 7-Dec-2016>

• On 15-Jun-2016, G.Paubert reported on tests. He finds a "frequency dependent leakage of I into both U and V in the receiver cabin" at 2mm with E150. This is from looking at a single onoff with the wobbler on Mars.

• On 9-May 2016, F.Herpin sent an observers report, saying that the strong leakage of I into V at 3mm with E090 severely hinders Zeeman observations to determine the magnetic field along the line of sight. The level of the instrumental circular polarisation (Stokes V) reaches 1.4%. This is based on observations of CN lines in IRC+10216 in March 2016. Comparison IRC+10216 spectra of 2006 do not show this leakage into V.

• The 3mm band setup (but also 2mm) was upgraded in November 2015: the mixers were exchanged against NOEMA type mixers, one orthomode transducer was installed to split polarizations. The performance of XPOL was evaluated as reported by A.Sievers on 1-April 2016 here. The instrumental circular polarisation was found to be rather large, at a level of 2%. The V map of Uranus now shows a central maximum, while previously it had looked very different, exhibiting two lobes (cf. Mars map at 3mm of April 2015 (see below) *after* the installation of the new receiver cabin optics).

• The external mirrors of EMIR, including the polarizer grid and the external cold load, had to be adapted to the overall new receiver cabin optics in April 2015. After re-installation, polarimetry was commissioned. See a preliminary report by A.Sievers of 19-Oct 2015 here. See also the general commissioning report by H.Ungerechts of 29-May-2015 here.

Observing

An XPOL observation is made in the wobbler-switching observing mode. This PaKo script sets up a simple XPOL observation at 3mm. More complex setups where two EMIR bands or two VESPA subbands are recorded are possible. The VESPA polarimetric capabilities are concisely described in Table 4 of the VESPA user guide. Observers are advised to check the good performance of XPOL by observing a polarization angle calibrator, the Crab nebula or the steep spectrum quasar 3C286 (see Scientific publications No. 1 and 5 below).

A separate note contains information on how to estimate the statistical error of an polarimetric observation. In addition, the note describes Zeeman observations and gives an estimate of the sensitivity for magnetic fields.

Beam maps

(under construction)

The following table lists the beam maps obtained with EMIR as of September 2012. The source name provides a link to the Stokes IQUV maps. The (x.y) cross on these maps indicates the horizontal and vertical directions in the Nasmyth cabin. Note that not all maps have been obtained under adequate weather conditions which can be judged from the appearance of non-gaussian features on the Stokes I maps. Copies of these maps are available in GILDAS format.

Technical documents

1. Versatile IF polarimeter at the IRAM 30m telescope

   • Thum et al. 2003 including the first mm observation of the polarization of the moon's limb

2. XPOL—the Correlation Polarimeter at the IRAM 30-m Telescope

   • Thum et al. 2008, PASP, 120, 777

3. Polarimetry with EMIR/XPOL - commisioning report (August 2010)

   • Wiesemeyer & Thum 2010, IRAM Report

4. Polarimetry with EMIR/XPOL and the new optics - commissioning report (October 2015)

   • Sievers et al. 2015, IRAM Report

5. Polarimetry with EMIR/XPOL: E090 with OMT - commissioning report (April 2016)

   • Sievers, 2016, IRAM Report

Scientific publications (selected for illustrating specific technical aspects)

1. Measurement of the Crab nebula polarization at 90 GHz as a calibrator for CMB experiments

   • Aumont et al. 2010

2. A sensitive upper limit to the circular polarization of the Crab nebula at λ3 mm

   • Wiesemeyer et al. 2011

3. Observations of the Goldreich-Kylafis effect in star-forming regions with XPOL at the IRAM 30 m telescope

   • Forbrich et al. 2008 This publication includes examples for assessing the magnitude of instrumental polarization due to sidelobes

4. A 3.5 mm Polarimetric Survey of Radio-loud Active Galactic Nuclei

   • Agudo et al. 2010

5. 3C 286: a bright, compact, stable, and highly polarized calibrator for millimeter-wavelength observations

   • Agudo et al. 2012

6. POLAMI: Polarimetric Monitoring of AGN at Millimetre Wavelengths - I. The programme, calibration and calibrator data products

   • Agudo, Thum et al. 2018

7. POLAMI: Polarimetric Monitoring of Active Galactic Nuclei at Millimetre Wavelengths - II. Widespread circular polarization

   • Thum, Agudo et al. 2018

8. POLAMI: Polarimetric Monitoring of Active Galactic Nuclei at Millimetre Wavelengths - III. Characterization of total flux density and polarization variability of relativistic jets

   • Agudo et al. 2018

Background material

1. IAU convention concerning Stokes Parameters

   • Radioastronomy, including the 30m telescope, uses these IAU conventions (see last page). Note however that according to this convention the sign of Stokes V is opposite to the usage in traditional physics as described by Born & Wolf in "Classical Optics"

2. The sign of Stokes V in 30m observations

   • with XPOL has been determined through several methods. They are described in a short note. It was verified that Stokes V spectra generated by the 30m data acquisition program have the correct sign according to the IAU convention.