SID Report 12/2015

Number of SID signals grouped by classification received at LS22 in Constance (Germany).

Number of SID signals grouped by classification received at LS22 in Constance (Germany).

The number of SID events I could register in December 2015 continues to drop.  This is also true for the normalized number of SID signals where I correct for varying daytime duration and number of days per month.

Normalized number of SID signals grouped by classification received at LS22 in Constance (Germany).

Normalized number of SID signals grouped by classification received at LS22 in Constance (Germany).

You may want to check out the GitHub project or the RPubs document to see how I calculated the data.

Number SID signals grouped by signal classification received at station LS22 in Constance (Germany) for December 2015.

Number SID signals grouped by signal classification received at station LS22 in Constance (Germany) for December 2015.

I classified the importance of a SID signal in regard to its  duration:

Classification1-11+22+33+
Duration [min]< 1919 - 25 26 - 3233 - 4546 - 8586 - 125> 125

My SID monitor LS22 received signals from the VLF transmitters DHO, GBZ, GQD, and ICV.

Monitored VLF transmitters DHO, GBZ, GQD, ICV, and HWU from station LS22.

Monitored VLF transmitters DHO, GBZ, GQD, ICV, and HWU from station LS22.

Lunt LS 60 THa

Lunt LS60THa on an AstroTrac mount.

Lunt LS60THa on an AstroTrac mount.

The Dawes limit is 2.3 arcsec and the Raleigh limit is 5.5 arcsec for Hα each. Useful magnification is in the range of [15, 90] assuming an observer’s pupil up to 4 mm.

NameFocal Length [mm]MagnificationField of View [°]
Kasai68332
77138
95649
12.54068
182897
2520135
Lunt LS105084

You may want to check out the GitHub project or the RPubs document to see how I calculated the data.

Hofheim Instruments 12

Hofheim Instruments 12” f/5 Travel Dobson set up for the stars.

Hofheim Instruments 12” f/5 Travel Dobson set up for the stars.

A simplified calculation for the star limit gives a value of 14.4 mag for this refractor. The Dawes limit to split double stars is 0.4 arcsec and the Rayleigh limit is 0.9 arcsec at scotopic vision maximum sensitivity at 510 nm each. Useful magnification is in the range of [43, 450] assuming an observer’s pupil up to 7 mm.

NameFocal Length [mm]MagnificationField of View [armin]
Kasai 53009
625011
721413
916716
12.512022
188332
256045
Meade MA1212518
Pentax XL5.228814
721418
10.514327
1410736
Pentax XW207556
Widescan III1311544
207567
3050101

You may want to check out the GitHub project or the RPubs document to see how I calculated the data.

M 29 (NGC 6913)

Open star cluster M29 (NGC 6913). Exposure time 6x2min. Planewave 20" CDK, FLI ProLine PL11002M CCD, Planewave Ascension 200HR. North is left, East is to the bottom.

Open star cluster M 29 (NGC 6913). Exposure time 6x2min. Planewave 20" CDK, FLI ProLine PL11002M CCD, Planewave Ascension 200HR. North is left, East is to the bottom.

History

Charles Messier observed this open cluster first in the night from June 29 to 30, 1764 and added it as number 29 to his catalog: “[…] I have discovered a cluster of six or seven very small stars which are below Gamma Cygni, & which one sees with an ordinary [non-achromatic] refractor of 3 feet & a half [FL] in the form of a nebula […]” [8].

Bode saw it on on December 5, 1774 and reported it as number 69 [1].

John Herschel detected it in sweep 200 and logged it as number 2078: “A coarse cluster of 8 large stars (10 m), and a dozen or 20 smaller in a roundish form. (Milky Way.)” [5]

Dreyer add it as number 6913 to NGC and commented: “Cl, P, lC, st L and S” (Cluster, poor, little compressed, large [bright] and small [faint] stars.) [3].

Distance

Distribution of open cluster distance . The red line marks the distance to M29

Distribution of open cluster distance. The red line marks the distance to M 29

Distribution of open cluster distance using a sample size of 10 pc. The red line marks the value of 1.15 kpc for M 29. It is a relatively near object as only 518 (25.4%) clusters out of 2038 with known value are closer to Sun. [2]. Newer research has updated the distance to 1.54 kpc [11].

Number of Stars

Distribution of number of stars in open clusters. The red line marks the number of stars for M29.

Distribution of number of stars in open clusters. The red line marks the number of stars for M 29.

Distribution of number of stars in open clusters using a sampling size of 10 members. Messier once only saw 6-7 stars but the modern value is 131 for M 29 (red line). Research has found that 1374 (65.4%) clusters out of 2102 with known value have less member stars [2].

Diameter

Distribution of diameter of open clusters using a sampling size of 2 arcmin. The red line marks the diameter for M29.

Distribution of diameter of open clusters using a sampling size of 2 arcmin. The red line marks the diameter for M 29.

Distribution of diameters for open clusters using a sampling size of 2 arcmin. The red line marks the value of 10 arcmin for M 29. It is a quite large object as 1451 (67.1%) clusters out of 2161 with known value are smaller [2].

Age

Distribution of age of open clusters . The red line marks the value for M29.

Distribution of age of open clusters . The red line marks the value for M 29.

Distribution of age of open clusters using a sampling size of 1 log(yr). The red line marks the value of 7.111 log(yr) for M 29. It is a quite young object as only 249 (12.4%) clusters out of 2010 with known value are younger [2].

Brightness

M 29 has a visual brightness of 7.1 mag [4]. But the visual extinction is variable across the cluster with an average value of 2.97mag [11].

Not so Field Stars

Annotated open star cluster M29 (NGC 6913). Exposure time 6x2min. Planewave 20" CDK, FLI ProLine PL11002M CCD, Planewave Ascension 200HR. North is left, East is to the bottom.

Annotated open star cluster M 29 (NGC 6913). Exposure time 6x2min. Planewave 20" CDK, FLI ProLine PL11002M CCD, Planewave Ascension 200HR. North is left, East is to the bottom.

Two bright stars do not belong to M 29. Namely HDE 229238 is a background, and HD 194378 (= V2031 Cyg, WDS 20238+3830) is a foreground star [11].

Variable Stars Nearby

NamePosition J2000.0Brightness Range [mag]TypePeriod [d]
V1332 Cyg21 17 13.35 +44 54 50.0
14.9 - 15.8 pLB
V1382 Cyg 20 22 54.40 +38 10 22.617.5 - <18.2 pLB
V1384 Cyg20 24 28.64 +38 30 43.016.8 - 18.1 pL
V 2031 Cyg20 23 51.01 +38 29 34.38.53 - 8.67 VEA2.70465

In my view of field of M 29 are several variable stars listed in GCVS [9]. The AAVSO Light Curve Generator does not provide any data for the variable stars from table above.

Double Stars Nearby

Name
Components
Position J2000.0
Distance [arcsec]
Position Angle [°]
Brightness [mag]
WDS 20229+3829
AB
20 22 56.71
+38 29 03.2
9.50
294
12.70, 13.90
WDS 20229+3829
BC

2.50
152
13.90, 14.70
WDS 20236+3817

20 23 38.85
+38 17 03.6
1.90
230
11.97, 12.06
WDS 20238+3830
AB
20 23 51.01
+38 29 34.3
38.60
320
8.65, 12.12
WDS 20238+3830
AC

65.80
296
8.65, 9.34
WDS 20238+3830
AD

54.60
119
8.65, 11.50
WDS 20238+3830
BE

14.80
263
12.12, 14.08
WDS 20238+3830
BF

16.00
323
12.12, 13.70
WDS 20238+3830
BG

14.80
30
12.12, 14.40
WDS 20240+3820

20 23 59.54
+38 19 30.6
1.30
259
10.95, 12.80
WDS 20242+3819

20 24 10.71
+38 19 11.4
30.80
341
9.16, 10.00
WDS 20243+3811

20 24 18.57
+38 10 30.9
1.90
203
10.53, 11.97

Near to M 29 are several double stars listed in WDS [7]. The most interesting one to me is a system build out of 7 components, namely WDS20238+3830.

WDS20238+3830 components A-G. Exposure time 6x2min. Planewave 20" CDK, FLI ProLine PL11002M CCD, Planewave Ascension 200HR. North is left, East is to the bottom.

WDS20238+3830 components A-G. Exposure time 6x2min. Planewave 20" CDK, FLI ProLine PL11002M CCD, Planewave Ascension 200HR. North is left, East is to the bottom.

Using DynamicPSF module from PixInsight I created a list of star centers and PSF parameters, with star id 1  component A, star id 2  component B, …, star id 7  component G.

View of Dynamic PSF module of PixInsight for WDS20238+3830.

View of Dynamic PSF module of PixInsight for WDS20238+3830.

With the help of WCSTools I converted the pixel coordinates of the star centers to sky coordinates. I then used the right ascension and declination of the star’s center to calculate distance and position angle for WDS20238+3830 [10] .

Components
r
Error(r)
P
Error(P)
[arcsec]
 [°]
AB
38.65
0.05
322.67
0.04
AC
65.88
0.03
298.65
0.02
AD
54.50
0.04
121.26
0.04
BE
14.70
0.01
266.16
0.03
BF
16.07
0.01
325.39
0.03
BG
14.66
0.01
33.36
0.06

Visual Observations

6” Refractor

To me it looks like the stars of this cluster are forming the greek letter π. I see about 20 stars (x45). NELM: 5.8 mag

References

[1] Bode, J.E., 1777. Ueber einige neuentdeckte Nebelsterne und einem vollständigen Verzeichnisse der bisher bekannten. In Astronomisches Jahrbuch oder Ephemeriden für das Jahr 1779. Berlin: Astronomisches Jahrbuch, pp. 65–71. Available at: http://messier.seds.org/xtra/similar/bode.html.
[2] Dias, W.S. et al., 2002. New catalogue of optically visible open clusters and candidates. Astronomy and Astrophysics, 389(3), pp.871–873. Available at: http://www.aanda.org/index.php?option=com_article&access=bibcode&bibcode=2002A%252526A...389..871D.
[3] Dreyer, J.L.E., 1888. A New General Catalogue of Nebulæ and Clusters of Stars, being the Catalogue of the late Sir John F. W. Herschel, Bart, revised, corrected, and enlarged. Memoirs of the Royal Astronomical Society, 49
[4] Frommert, H. & Kronberg, C., 2007. Messier 29. SEDS. Available at: http://messier.seds.org/m/m029.html [Accessed August 20, 2014].
[5] Herschel, J., 1833. Observations of nebulæ and clusters of stars, made at Slough, with a twenty-feet reflector, between the years 1825 and 1833. Philosophical Transactions of the Royal Society of …, pp.359–505. Available at: http://www.jstor.org/stable/108003.
[6] Kharchenko, N.V. et al., 2005. Astrophysical parameters of Galactic open clusters. Astronomy and Astrophysics, 438(3), pp.1163–1173.
[7] Mason, B.D. et al., 2014. The Washington Visual Double Star Catalog (Mason+ 2001-2014). VizieR On-line Data Catalog.
[8] Messier, C., 1774. Catalogue des Nébuleuses & des amas d'Étoiles, que l'on découvre parmi les Étoiles fixes sur l“horizon de Paris; observées à l”Observatoire de la Marine, avec differens instruments. Mémoires de l'Académie Royale des Sciences, pp.435–461. Available at: http://messier.seds.org/xtra/history/m-cat71.html.
[9] Samus, N.N., Durlevich, O.V. & al, E., 2009. General Catalogue of Variable Stars (Samus+ 2007-2013). VizieR On-line Data Catalog.
[10] Smolinski, J. & Osborn, W., 2006. Measurement of Double Stars With a CCD Camera: Two Methods. Rev Mex AA (Serie de Conferencias).
[11] Straizys, V. et al., 2014. The Enigma of the Open Cluster M29 (NGC 6913) Solved. arXiv.org, astro-ph.SR.

External Links

M 29 - NED
M 29 - SEDS
M 29 - SIMBAD
M 29 - WEBDA
M 29 - Wikipedia
M 29 - WikiSky

Acknowledgment 

For this blog post I used data from the International Variable Star Index (VSX) database, operated at AAVSO, Cambridge, Massachusetts, USA, the SIMBAD database, operated at CDS, Strasbourg, France, and the Washington Double Star Catalog maintained at the U.S. Naval Observatory.

 

Setting up a Raspberry Pi for SID Processing

1. Download the Raspian image from Raspberry Pi Foundation

2. Insert the SD card in the reader and identify it on your Mac

$ diskutil list
/dev/disk8 (internal, physical):
   #:                       TYPE NAME                    SIZE       IDENTIFIER
   0:     FDisk_partition_scheme                        *7.9 GB     disk8
   1:             Windows_FAT_32 boot                    58.7 MB    disk8s1
   2:                      Linux                         7.9 GB     disk8s2

3. Unmount the disk from your Mac, e.g. disk8

$ diskutil unmountDisk /dev/disk8
Unmount of all volumes on disk8 was successful

4. Write the ISO image to the (raw!) disk on your Mac

$ sudo dd bs=1m if=2015-09-24-raspbian-jessie.img of=/dev/rdisk8

This will take some time. You may want to press CTRL+T to get some intermediate information

load: 1.65  cmd: dd 10305 uninterruptible 0.00u 1.61s
2537+0 records in
2536+0 records out
2659188736 bytes transferred in 139.614439 secs (19046660 bytes/sec)

5. Unmount the SD card from your Mac and remove it

6. Insert the SD card into your Raspberry Pi and boot it

7. Configure the Raspberry Pi system to your need (hostname, expand disk, …)

8. Update the system

$ sudo apt-get update
$ sudo apt-get upgrade

9. Fork the SuperSID project and clone it to your Raspberry Pi, e.g.

$ git clone https://github.com/rrogge/supersid.git SuperSID
$ git remote add upstream https://github.com/ericgibert/supersid.git

10. Install python modules on Raspberry Pi

$ sudo apt-get install python-matplotlib
$ sudo apt-get install python-alsaaudio

11. Set up Raspberry Pi USB audio card, e.g. adjust gain for the input (Mic)

$ alsamixer

Set Up Raspberry Pi USB Audio Card

12. Edit the SuperSID configuration file, e.g. my station is called LS22

$ cd ~/SuperSID/Config
$ cp supersid.cfg supersid.ls22.csg
$ vi supersid.ls22.cfg

13. Start SuperSID monitor

$ cd ~/SuperSID/supersid
$ ./supersid.py ../Config/supersid.ls22.cfg

Optimal Focal Ratio for Imaging

From Rayleigh criterion I get the resolution for a telescope r_T with focal length L, aperture D for light of wavelength \lambda

r_T = 1.220\,L \ /\ D\,\lambda

and from Nyquist-Shannon sampling theorem I get the resolution of a camera  r_C with pixel size s

r_C= 2\,s

For optimal imaging results I set the resolution of the telescope equal to the resolution of the camera and so I get the optimal focal ratio for a camera at a given wavelength

 N = 2\,s \ /\ 1.220\, \lambda

Optimal focal ration for imaging with cameras of three different pixel sizes.

Optimal focal ration for imaging with cameras of three different pixel sizes.

Lunar Eclipse September 28, 2015

Lunar eclipse on September 28, 2015. (Self built refractor based on a TMB 80/600 objective, Canon 10D, Losmandy G11, Berlebach Planet)

Composite of lunar eclipse on September 28, 2015. Self built refractor based on a TMB 80/600 objective, Canon 10D, Losmandy G11, Berlebach Planet.

To calculate the size of Earth’s shadow I took one image before (#1923, 1:35 UT) and one after (#2072, 4:07 UT) the umbral phase of the lunar eclipse but with Earth’s shadow visible on the Moon surface using a self-built refractor based on a TMB 80/600 objective and a Canon 10D DSLR.

Self built refractor based on a TMB 80/600 objective, Canon 10D, Losmandy G11, Berlebach Planet.

Self built refractor based on a TMB 80/600 objective, Canon 10D, Losmandy G11, Berlebach Planet.

To convert the Canon raw images I used PixInsight (PI) version 1.08.03.1123 and applied the VNG debayer algorithm (RGGB pattern and multi-resolution support for noise evaluation).

Images of the partially eclipsed moon at September 28, 2015 06:07 UT (left, #2072) and 03:35 UT (right, #1923) made during the then ongoing lunar eclipse. Self built refractor based on a TMB 80/600 objective, Canon 10D, Losmandy G11, Berlebach Planet.

Images of the partially eclipsed moon at September 28, 2015 06:07 UT (left, #2072) and 03:35 UT (right, #1923) made during the then ongoing lunar eclipse. Self built refractor based on a TMB 80/600 objective, Canon 10D, Losmandy G11, Berlebach Planet.

To prepare image binarization I plotted a line profile through the Moon’s and Earth’s shadow centers. After inspecting the plot I set the binarization threshold to an intensity value of 0.002.

Intensity profile for a line through the centers of Moon and Earth’s shadow for the images of partially eclipsed moon at 06:07 UT (blue, #2072) and 03:35 UT (red, #1923). Binarization threshold is marked as black line.

Intensity profile for a line through the centers of Moon and Earth’s shadow for the images of partially eclipsed moon at 04:07 UT (blue, #2072) and 01:35 UT (red, #1923). Binarization threshold is marked as black line.

For edge detection I applied a simple Laplacian operator on the binarized images.

Edge detection result after I applied a Laplacian operator on the binarized image of the partially eclipsed Moon at 06:07 UT (left, #2072) and 03:35 UT (right #1972).

Edge detection result after I applied a Laplacian operator on the binarized image of the partially eclipsed Moon at 04:07 UT (left, #2072) and 01:35 UT (right #1972).

Together with the known size of the Moon (3474 km) fitting circles to the detected edges gave me a diameter of Earth’s shadow of 7828 +- 61 km for image #1972 at 01:35 UT and 7592 +- 77 km for image #2072 at 04:07 UT. And indeed a check with HORIZONS from JPL shows that the Moon has moved outward by 1866  km between the two images and so the shadow diameter decreased.

Fitted circles to the Moon’s border and Earth’s shadow for the images of the partially eclipsed Moon at 01:35 UT (black, red) and 04:07 UT (blue, orange). The Earth’s shadow diameter decreased because the Moon moved out between the two image

Fitted circles to the Moon’s border and Earth’s shadow for the images of the partially eclipsed Moon at 01:35 UT (black, red) and 04:07 UT (blue, orange). The Earth’s shadow diameter decreased because the Moon moved out between the two image

The error values of the Earth's shadow diameter I got by calculating the standard deviation of detected edges from fitted circle.

Installing SExtractor on Mac OS X Yosemite

Installing SExtractor 2.19.5. on OS X 10.10.5 using MacPorts 2.3.3 is actually reported as being somewhat tricky. Main reason for this is that the ATLAS library wants an FORTRAN compiler by default (Atlas could not detect any fortran compiler. If you really don’t need the fortran interface to be built, please use the +nofortran option, else install a fortran compiler (e.g. gcc4X) before building Atlas). But even an installed FORTRAN compiler e.g. GCC 4.8 doesn’t make MacPorts happy. In addition to this problems the installation is known of being a somewhat lengthly procedure (> 6 hours).

To stick with MacPorts anyway I succeeded with the following protocol:

$ sudo launchctl unload -w /System/Library/LaunchDaemons/com.apple.metadata.mds.plist
$ sudo port -v install sextractor +nofortran
$ sudo launchctl load -w /System/Library/LaunchDaemons/com.apple.metadata.mds.plist

Line 1 Switched off spotlight search to speed up compiling
Line 2 Install sextractor. As I have no FORTRAN compiler installed I complied only the C API of the ATLAS library. Compiling an installing all the stuff took about 4 hours on my 3.5 GHz i7 iMac w/ 16GB (this has some room for improvements as the compilation process only took about 10% of the system load, e.g. the compilation is not running in parallel but only on one core).
Line 3 Enabled spotlight again

SID Signal Processing

During day time the sunlight (extreme UV and X-ray) ionizes atoms and molecules in the ionosphere, in the night the ions and electrons recombine. This is why we find electron density is raised during the day and lowered in the night for the D, E, and  for the F1 region of the ionosphere.

Plot of electrons density and temperature versus height of a virtual ionosphere (IRI 2012) at 0:00 (blue line) and 12:00 UT (red line). Data of the reference ionosphere (IRI-2012) by NASA.

Plot of electrons density and temperature versus height of a virtual ionosphere (IRI 2012) at 0:00 (blue line) and 12:00 UT (red line). Data of the reference ionosphere (IRI-2012) by NASA.

During the day a VLF transmitter sky wave is partly absorbed by the D and E layers and reflected by the F1 layer. In the night the it gets its way to the F1 and F2 combined F layer where it’s reflected. At the receiver site the sky wave interferes with the surface wave resulting in the final VLF transmitter signal strength, this is true for day and night time.

A VLF transmitter signal on the day side is partially absorbed by the E and D layers and reflected by the F1 layer. While on the night side the E, D, and F1 layers don’t exist and so the signal is reflected by the F2 layer. Image of Earth “The Blue Marble” by NASA.

A VLF transmitter signal on the day side is partially absorbed by the E and D layers and reflected by the F1 layer. While on the night side the E, D, and F1 layers don’t exist and so the signal is reflected by the F2 layer. Image of Earth “The Blue Marble” by NASA.

A plot of the VLF transmitter signal strength for 24 hours shows therefore a typical diurnal day/night pattern. Depending on the geographical positions of the VLF transmitters and the receiver site the night/day and day/night patterns show up at different points in time.

VLF transmitter signal for GQD (green) and ICV (purple) showing a typical diurnal day/night pattern.

VLF transmitter signal for GQD (green) and ICV (purple) showing a typical diurnal day/night pattern.

A flare on the Sun dramatically impacts the ionization of D layer and its electron density may be raised by up to two magnitudes. So the sky wave is suddenly not longer absorbed but reflected by the D layer and the signal strength of the VLF transmitter changes dramatically. It gets higher or lower depending on the phase change between the two waves.

VLF transmitter signal for GQD (green) and ICV (purple). ICV was offline from 09:30 to 10:45 UT. Three solar events (#8690, #8700, and #8740) are visible.

VLF transmitter signal for GQD (green) and ICV (purple). ICV was offline from 09:30 to 10:45 UT. Three solar events (#8690, #8700, and #8740) are visible.

To collect SID data I use an 1 m x 1 m wireframe antenna, a SuperSID monitor from Stanford Solar Center as amplifier, a cheap and simple USB A/D converter, and a Raspberry Pi as data host running 7/24. The SuperSID software samples the data from USB A/D converter over 5 seconds, filters it for VLF transmitter frequencies, and finally writes it to daily files. To create plots from the VLF transmitter signals I transfer the data files to a data processing host on a regularly base and run a R script on it.

SID Workflow #1

While inspecting the VLF transmitter signal plots I manually extract SID signals to a spreadsheet. From SID start and end time I calculate the duration which I use to classify the importance of a signal:

Classification1-11+22+33+
Duration [min]< 1919 - 25 26 - 3233 - 4546 - 8586 - 125> 125

To create a monthly report  I then run a R script on the spreadsheet data. This gives me a plot of  SID signals by classification and month so that I can  see quickly how Sun activity evolves.

SID Workflow #2

You may want to check out the SID Signal Processing GitHub project for R and shell script sources.