Dyson Sphere search for amateurs  updated February 26, 2016  D. Carrigan carrigan@fnal.gov (subject line must be sensible)

Channeling Adv. Accel. Dyson/IA SETI Biography Bibliography Infrared
On the spread sheet to the right there are ten sources with best fits to a three point Planck distribution and an associated spectrogram in the IRAS Low Resolution Spectrometer (LRS). They have been selected as illustrative examples of pure Dyson Sphere look-a-likes from a sample of 3000 sources selected from the 250,000 source IRAS point source catalog that had reasonable flux values in the IRAS 12, 25, and 60 micron filters, and temperatures in the 150 to 500 degree Kelvin range. Of the 3000 10 to 20% have associated IRAS Low Resolution spectrographic information.
The Excel Spread Sheet for Dyson Spheres

You can search for pure Dyson Spheres using the steps below.

  1. Pick an IRAS source from the list on the Excel spread sheet (the row labeled  IRAS ID).  For more on IRAS see Introduction to IRAS.
  2. You are an astronomer now so you need to look at the source. Go to the amazing SIMBAD viewer to see the sky at the source. This is maintained by the Centre de Donnes Astronomiques de Strasbourg. In the entry box under "1 enter an identifier, type IRAS followed by a space and the IRAS source ID. Press the SUBMIT button five or six lines down. Shortly a page will appear with a lot or information about the IRAS source. Now it gets slightly tricky depending on your computer and its firewall. First try the Aladin Java applet button on the right. If you are lucky a picture will appear after a minute or so showing the sky over a square several arc minutes on a side. If you are unlucky you can use the Aladin Previewer. The porblem is the sky view is too large and there is not much flexibility. A third path if you are a computer whiz is to install the Aladin stand-alone viewer. Taking the lucky case wait until the picture appears. Right click on the picture and click grid to put a grid on. It helps to go down to the lower right and put the zoom at 2. Click the load button on the upper left to bring up the 2MASS and IRAS information. Get 2MASS by clicking the VizieR buttor on the right and then clicking 2MASS on the list that appears and then the submit button on the lower left.. There are lots of 2MASS sources so this takes awhile. Scroll down the list and repeat for IRAS. If you are lucky you will have your IRAS source near the center of the picture. Click on the IRAS source and information will appear at the bottom of the page. (This may take some practice.) A pure Dyson Sphere will not have an optical source in the survey optical picture.
  3. See if there is a 2MASS source nearby. Shift click to bring up its parameters at the bottom of the page. The xxx.xxxxxx 2MASS identifier is on the left. The values for the three IRAS filters in magnitudes appear in the third, fourth, and fifth boxes. These can be converted to values in Janskys by using the formulas in the 2MASS All Sky Explanatory Supplement VI.4a. Alternatively, the flux values for the three filters in magnitudes can be taken from the Excel spread sheet. They must still be converted to Janskys. Basically F = F0*10^(-M/2.5) where M is the magnitude and F0 = 1594 (for 1.235 micron filter), 1024 (1.662) and 666.7 (2.159). Plot  F (y axis) vs landa (microns) as x axis. Typically a good pure Dyson Sphere would be expected to have a small value.
  1. Plot the ratio F[i]/k[i] where F, in Janskys is the flux and k is the "color correction" for each of the four IRAS filters. In the IRAS database the fluxes are called FLUX[]. In the spread sheet these are at IRAS filter information. The wavelength (lamda) values are listed on the left and are the x values for the points. The color corrections are given just below the fluxes. Typically the color corrections are approximately 1. (Information on the color correction is given in Section VI.C.3 of the IRAS Explanatory Supplement.) The y values are the flux/color correction for each source.
  2. Plot the corrected raw spectra from the IRAS Low Resolution Spectrometer (LRS) times the wavelength in microns times 3.34E11 (10E20/c where c is the speed of light in m/second). The spectra are given in the chart but can be obtained for other IRAS sources using  Kevin Volk's Calgary page. Put in the IRAS ID and remember to use "corrected raw text". There are two spectra, one for shorter wavelengths, one for longer. They appear on the spread sheet at IRAS LRS - Calgary "corrected raw text". The associated wavelengths in microns are on the left.
  3. Fit the IRAS 12, 25, and 60 micron filters with a Planck distribution. Use the frequency version of the Planck distribution. This sounds strange since the x axis is in wavelength. However, this approach more easily matches the other distributions.  To do this it is necessary to fit the peak value (related to apparent magnitude of the source) and temperature. One way to estimate the temperature is to use so-called color-color temperatures. These two temperatures are given in the spread sheet under temperature and fit. The spead sheet gives a fit at the trial temperature, the coarse best fit that was done for the Dyson sphere search of the entire data base. A typical fit is shown on the "working" page of the spread sheet.
  4. Is the source a pure Dyson Sphere? For this search a pure Dyson Sphere is assumed to be thick enough to adsorb all the radiation from the star and reradiate it as a pure black body spectrum. If there are any absorbtion  or emission lines the source is not a pure Dyson Sphere. If the spectra from the IRAS filters, the LRS, and the 2MASS points do not match a Planck distribution it is not a pure Dyson Sphere. If it did closely match the Planck distribution (as a few post-AGB sources seem to) then one may need to really look in detail as other frequencies like radio bands to rule out other possibilities.