Jump to content

  •  

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Photo

Not exactly the Imaging X-ray Polarimetry Explorer, but it's the first astronomical polarimetry image for me

  • Please log in to reply
8 replies to this topic

#1 Organic Astrochemist

Organic Astrochemist

    Apollo

  • *****
  • topic starter
  • Posts: 1,179
  • Joined: 10 Jan 2015

Posted 02 January 2022 - 11:02 PM

Recently, NASA launched the Imaging X-ray Polarimetry Explorer which, along with the JWST, has new and unique observing capabilities. 

IXPE

 

More humbly, I have been struggling for over a year with my QHY550P camera with a polarization filter.

 

I finally learned enough Python (and about Windows 10) to enable me to create an image of the angle of polarization.

 

I used a celestron C90 piggybacking on my Atlas mount to capture this image of the Orion Nebula, in 360 seconds at 100 gain. I was using a V-band filter so no Hα. 

OrionNebula3bcropped.png

 

The polanalyser code required some debugging and optimization, but it is able to demosaic the image into four channels, one for each polarization angel 0°, 45°, 90° and 135°. The values of four contiguous pixels are then used to calculate the degree of linear polarization and the angle of linear polarization. The code then converts the angle value into a color based on the following color wheel:

AoLP_wheel.jpg

 


  • gregj888 likes this

#2 Organic Astrochemist

Organic Astrochemist

    Apollo

  • *****
  • topic starter
  • Posts: 1,179
  • Joined: 10 Jan 2015

Posted 02 January 2022 - 11:24 PM

And here is the angle of polarization image

OrionNebula3b_AoLPcropped.png

I didn't have time to properly calibrate the image with darks and flats, so that is certainly the next step. But as a proof-of-principle I think this is a success.  Whenever there are changes in intensity that are smaller than the four pixel scale, the values of polarization will be affected and this will generate erroneous values. The yellow blobs in the middle are the trapezium. In between the pink and the yellow pixels that run from the center to the top are the red pixels that correspond to 0° angle of polarization (I have to calibrate this too with the sky). Going counterclockwise the colors are red to yellow to green to blue to pink to red to yellow to green to cyan to blue to pink. In other words, the angle of polarization is symmetric around θ1 Orionis C, the brightest star in the trapezium which is primarily responsible for ionizing the nebula. This is consistent with the literature

 

 


  • gregj888 and mwr like this

#3 mwr

mwr

    Ranger 4

  • -----
  • Posts: 375
  • Joined: 12 May 2018

Posted 08 January 2022 - 08:23 AM

The yellow blobs in the middle are the trapezium. In between the pink and the yellow pixels that run from the center to the top are the red pixels that correspond to 0° angle of polarization (I have to calibrate this too with the sky). Going counterclockwise the colors are red to yellow to green to blue to pink to red to yellow to green to cyan to blue to pink. In other words, the angle of polarization is symmetric around θ1 Orionis C, the brightest star in the trapezium which is primarily responsible for ionizing the nebula. This is consistent with the literature

Really nice work! Can you give the degree of polarization as well? Is there a software available that allows to draw the polarization vectors (lenght=degree of polarization; direction=angle of polarization)?



#4 Organic Astrochemist

Organic Astrochemist

    Apollo

  • *****
  • topic starter
  • Posts: 1,179
  • Joined: 10 Jan 2015

Posted 12 January 2022 - 03:31 PM

That is how they display the data in the paper I cited. Theoretically it should be doable in Python, but I'm not there yet. What I can do is use the color to show the angle and the intensity to show the degree of linear polarization. But there are some problems with astronomical images. The program must have predefined maximum values for each pixel so if I stack or saturate the stars it doesn't return the correct DoLP. So I still have some work to do. Here is an example with a terrestrial image.

Here is an image of a table with light reflecting off it.

TableI.jpg

 

Here is the degree of linear polarization DoLP

Table_DoLP.jpg

 

 

 



#5 Organic Astrochemist

Organic Astrochemist

    Apollo

  • *****
  • topic starter
  • Posts: 1,179
  • Joined: 10 Jan 2015

Posted 12 January 2022 - 03:33 PM

Here is the angle of polarization for each group of four pixels AoLP

Table_AoLP.jpg

 

 

 



#6 Organic Astrochemist

Organic Astrochemist

    Apollo

  • *****
  • topic starter
  • Posts: 1,179
  • Joined: 10 Jan 2015

Posted 12 January 2022 - 03:34 PM

And here is the angle of polarization, but scaled to the degree of polarization, which shows the color for the angle but the intensity is given by the degree of polarization,

Table_AoLPDoLP.jpg

 

This effectively has the same information as the graphic in that professional paper.


  • gregj888 and mwr like this

#7 Octans

Octans

    Mariner 2

  • -----
  • Posts: 279
  • Joined: 13 Aug 2009

Posted 12 January 2022 - 06:05 PM

One nice way to visualize this kind of data is to put it into HSV or HSL color, with polarization angle as the hue, degree of polarization as the saturation, and the intensity as the value/lightness. I think that may be roughly how the fairly famous Hubble picture of the Egg Nebula was constructed: https://hubblesite.o...1305-Image.html


Edited by Octans, 12 January 2022 - 06:08 PM.


#8 mwr

mwr

    Ranger 4

  • -----
  • Posts: 375
  • Joined: 12 May 2018

Posted 13 January 2022 - 05:07 AM

That is how they display the data in the paper I cited. Theoretically it should be doable in Python, but I'm not there yet.

I can still vaguely remember a polarisation map of comet Neowise and have just found the corresponding page at the BAA:

 

https://britastro.or...15#comment-9015

https://britastro.or...29#comment-9029

 

"The polarisation functions in IRIS are used to produce the various maps."

 

May be you can give it a try. IRIS is available here: http://www.astrosurf...s-software.html

 

The tutorial describes the polarization analysis here: http://www.astrosurf...13/doc34_us.htm



#9 morsanimi

morsanimi

    Lift Off

  • -----
  • Posts: 17
  • Joined: 16 Dec 2011

Posted 14 January 2022 - 02:43 PM

Maybe here, there could be sth of use.

 

https://github.com/a...plpy/issues/128




CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.


Recent Topics






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics