Jump to content


Photo

Dark Sky Meter Live map now online

  • Please log in to reply
38 replies to this topic

#1 Nop

Nop

    Vendor (iPushto)

  • -----
  • Vendors
  • Posts: 121
  • Joined: 12 Feb 2006
  • Loc: Netherlands, Valkenburg

Posted 03 June 2013 - 02:06 AM

Dear all,
Our Dark Sky Meter map is online. The map contains all valid night sky measurements made with the Dark Sky Meter for iPhone. It's fully automated.
Feel free to explore
http://www.darkskymeter.com/map/
Regards
Norbert

PS Note that this is a 'semi-live' map. All the data is refreshed every hour.

#2 cuivienor

cuivienor

    Viking 1

  • -----
  • Posts: 562
  • Joined: 06 Oct 2010
  • Loc: Tokyo, Japan

Posted 03 June 2013 - 03:18 AM

Hi Nop,

This is excellent! Thank you, I'm already waiting for some new neat data in Tokyo - now if the app can also measure the frequency of light pollution, maybe we could find cities where phase of the streetlights is the same, and this skyglow peaks could be filtered out by blocking out light at the peaks :)

OK the above was a bit "out there", some on a more feasible angle, I think this app could also turn into a great way to find nice observation places - my understanding is that locations are rounded to 800 meters. Wouldn't it be a good idea to allow users to disclose the exact location (from a public park for example) should they chose to do so? This would turn into a great dictionary of observation sites.

This would be particularly useful for city dwellers like me, in Tokyo, to know where we could travel by train to which park to get the best skies possible! Under that mode, if submitters could also rate their location (streetlights, public lights, etc.) it would make it even more powerful.

Yannick

#3 dcbrown73

dcbrown73

    Explorer 1

  • -----
  • Posts: 64
  • Joined: 23 Jun 2012
  • Loc: Lower Connecticut

Posted 03 June 2013 - 08:09 AM

Pretty cool. Though, I'm an Android guy. I don't have any iOS devices.

#4 REC

REC

    Fly Me to the Moon

  • *****
  • Posts: 5480
  • Joined: 20 Oct 2010
  • Loc: NC

Posted 03 June 2013 - 08:25 AM

Oh, that's cool. Just saw my post of 18.42 here in Gastonia, NC.

Bob

#5 pfile

pfile

    Gemini

  • -----
  • Posts: 3169
  • Joined: 14 Jun 2009

Posted 03 June 2013 - 01:40 PM

i wish you would fuzz the locations on the map a little bit... bit of a privacy concern.

#6 Nop

Nop

    Vendor (iPushto)

  • -----
  • Vendors
  • Posts: 121
  • Joined: 12 Feb 2006
  • Loc: Netherlands, Valkenburg

Posted 04 June 2013 - 01:33 AM

Thanks!
A quick note on location:

- The locations are rounded (about 110 meters for privacy reasons.
- Uploading is not obligatory. The app doesn't do any 'sneaky uploading':) . So you are in control.
- If you don't want your location on the map. Simply erase it from the 'My Data' menu (swipe and delete)

#7 Qwickdraw

Qwickdraw

    Surveyor 1

  • *****
  • Posts: 1731
  • Joined: 03 Mar 2012
  • Loc: Ann Arbor, MI

Posted 04 June 2013 - 04:43 AM

It would be nice if the moon phase could somehow be calculated out of the number so that all locations show actual LP factors regardless of moon phase.

#8 Nop

Nop

    Vendor (iPushto)

  • -----
  • Vendors
  • Posts: 121
  • Joined: 12 Feb 2006
  • Loc: Netherlands, Valkenburg

Posted 04 June 2013 - 06:16 AM

Yes, we are discussing that. But, it is and will be an estimated result. There's a lot of side effects measuring with moon (dirty sensor, bad transparency etc).

#9 derangedhermit

derangedhermit

    Apollo

  • -----
  • Posts: 1158
  • Joined: 07 Oct 2009
  • Loc: USA

Posted 04 June 2013 - 09:03 AM

If I uncheck the markers for data that appears less useable, there's not much left. Wrong/no time, low angle, clouds...you're getting many more less useful readings than useful ones. Any way to imrpove those percentages?

#10 Nop

Nop

    Vendor (iPushto)

  • -----
  • Vendors
  • Posts: 121
  • Joined: 12 Feb 2006
  • Loc: Netherlands, Valkenburg

Posted 05 June 2013 - 03:58 AM

@derangedhermit
It will need some time. Keep in mind that the app is released only 2 months ago and for the Northern hemisphere the nights are getting shorter. Here in Europe the weather has been terrible. So there's a lack of good readings. I'll expect to have a good number of readings coming this fall and more amateur astronomers will provide useful SQM measurements. And I hope we get a lot of users on the southern hemisphere.
With over 2000 downloads worldwide (lite+pro) the app has a good potential.

#11 derangedhermit

derangedhermit

    Apollo

  • -----
  • Posts: 1158
  • Joined: 07 Oct 2009
  • Loc: USA

Posted 05 June 2013 - 07:38 AM

I know it will. I hope the app or server provides feedback to users about no/wrong time, pointing angle - whatever it can to improve the readings.

#12 audioaficionado

audioaficionado

    Apollo

  • -----
  • Posts: 1014
  • Joined: 24 May 2012
  • Loc: Medford, Orygun, USA

Posted 07 June 2013 - 03:09 PM

Dark Sky Meter for iPhone


There are also a lot of Android phone users who could be recruited to add more data points.

#13 Dark Sky Scott

Dark Sky Scott

    Vostok 1

  • -----
  • Posts: 127
  • Joined: 06 Jan 2012
  • Loc: Tucson, Arizona

Posted 07 June 2013 - 06:59 PM

Android phones have a lot of variability in their cameras. A different app was developed for them. Check it out here:
https://play.google....lux.welovestars

#14 StarDaddy

StarDaddy

    Lift Off

  • -----
  • Posts: 18
  • Joined: 10 May 2013
  • Loc: Eastern Shore, Maryland

Posted 14 June 2013 - 06:57 PM

i wish you would fuzz the locations on the map a little bit... bit of a privacy concern.


=========

Don't worry my friend, the NSA knows exactly where you are! (and the fact that we are having this correspondence)

#15 Mauro Da Lio

Mauro Da Lio

    Viking 1

  • -----
  • Posts: 571
  • Joined: 12 Sep 2004

Posted 04 August 2013 - 04:31 PM

I have installed the Pro version yesterday. Here is a comparison with a "hardware" SQM.

a) Today at my home.
I took 5 SQM readings, which were: 18.88, 18.89, 18.87, 18.87, 18.87. Mean 18.876 and SD < 0.01
I took 5 Dark Sky Meter readings which were: 18.78, 18.82, 18.73, 18.75, 18.84. Mean 18.784, SD 0.046.
The difference between the two means is less than 0.1mpsas, ie less than the claimed accuracy of standard SQM (mine is a little bit optimistic). HOWEVER the repeatability of the Dark Sky Meter is nearly 5 times worse.

b) Yesterday at dark site.
The SQM was averaging at 21.6, with deviations of the order of 0.1. But the DSM, in such low lights, had very large fluctuations, from 21.4 to 22.2 (nearly one magnitude).

It seems that in low lights (ie what is more interesting for me) the DSM is loosing accuracy.

Hopefully tomorrow I will be on a dark site again and I will record carefully the readings of both.

#16 derangedhermit

derangedhermit

    Apollo

  • -----
  • Posts: 1158
  • Joined: 07 Oct 2009
  • Loc: USA

Posted 04 August 2013 - 05:21 PM

Mauro: is the close agreement with the SQM after you "calibrated" the app against the SQM? Is the SQM the narrower FOV version, or the original?

Also, very good information, thank you; and I look forward to your further tests.

#17 Mauro Da Lio

Mauro Da Lio

    Viking 1

  • -----
  • Posts: 571
  • Joined: 12 Sep 2004

Posted 05 August 2013 - 05:35 AM

No, I did no particular calibration. My SQM is on average 0.1 mg optimistic and is the standard version. However the sky at my home is pretty uniformly polluted so the difference in the FOV is in influential.

However, despite the agreement at ~SQM 19, there is no longer agreement above SQM 21, and, what is worse, the DSM gives non repeatable readings.

#18 Tony Flanders

Tony Flanders

    Voyager 1

  • *****
  • Posts: 11303
  • Joined: 18 May 2006
  • Loc: Cambridge, MA, USA

Posted 05 August 2013 - 08:50 AM

However, despite the agreement at ~SQM 19, there is no longer agreement above SQM 21, and, what is worse, the DSM gives non repeatable readings.


That's exactly what I would expect. The DSM has to rely on the built-in camera, which is designed for compactness and high pixel count -- precisely the qualities that make a camera work poorly at low light levels. On the whole, it's amazing that it works as well as it does.

#19 Mauro Da Lio

Mauro Da Lio

    Viking 1

  • -----
  • Posts: 571
  • Joined: 12 Sep 2004

Posted 06 August 2013 - 04:31 AM

This is quite obvious. There is no detail about how the sensor data are processed, but I believe that by lumping all the pixel together one may reduce noise and increase signal. After all the CCD of the iPhone is similar in size to the sensor element of the SQM. So it looks to me that a clever processing might obtain similar results.

Use in Dark sites is important. Moreover another useful feature might be that of collecting readings at intervals the whole night and producing a chart (similar to the SQM-LE).

#20 derangedhermit

derangedhermit

    Apollo

  • -----
  • Posts: 1158
  • Joined: 07 Oct 2009
  • Loc: USA

Posted 06 August 2013 - 05:46 PM

Mauro, the phone gets *hot* inside, the SQM does not. The heat you know makes a lot of noise in the image.

Kill all the unneeded apps and services, see if there is a way to turn off the mobile radio hardware (big problem), turn off the display hardware (big problem), etc. Wrap a cold pack around it and let it chill.

It may help in both variance and going deeper.

#21 Mauro Da Lio

Mauro Da Lio

    Viking 1

  • -----
  • Posts: 571
  • Joined: 12 Sep 2004

Posted 07 August 2013 - 05:32 AM

That is a good advice! I will shut down everything not needed (I think the the "airplane use" switch shuts down every radio link) and test it for repeatability the next time I go to a dark site.

#22 derangedhermit

derangedhermit

    Apollo

  • -----
  • Posts: 1158
  • Joined: 07 Oct 2009
  • Loc: USA

Posted 08 August 2013 - 05:20 PM

That is a good advice! I will shut down everything not needed (I think the the "airplane use" switch shuts down every radio link) and test it for repeatability the next time I go to a dark site.

Since you will likely be at the site for a while, you might also at one point try turning the phone off for 30 min to let it cool to ambient, then turn on and immediately try it.

#23 Mauro Da Lio

Mauro Da Lio

    Viking 1

  • -----
  • Posts: 571
  • Joined: 12 Sep 2004

Posted 11 August 2013 - 12:13 PM

Yesterday I went to this site in the middle of the Alps. http://goo.gl/maps/eqOFW

The sky was clear. A traditional (non-L) read abut 21.64 in the first part of the night aiming the Cygnus (at zenith overhead). The SQM was mine, which is probably 0.1 mags optimistic (so the correct readings might have been ~ 21.55).

Turning the iPhone off did not work much probably because as soon as the iphone is turned on and used to gather data, it undergoes a temperature transient. However the ambient temperature was ~6 °C and i worked in a way to keep the iPhone activity constant and at minimum.

I got the following three series:

a) between 00.55 and 00.58: 21.77, 22.03, 21.93, 21.80, 22.07, 21.92, 21.96, 21.77, 21.72, 21.71, 22.00. i.e. 21.88 +/- 0.39 (mean 21.88 standard deviation 0.13).

b) between 02.10 and 02.15: 21.99, 21.50, 21.82, 21.54, 22.26, 21.62, 21.85, 22.06.
i.e. 21.83 +/- 0.80 (mean 21.83 standard deviation 0.27)

c) between 03.24 and 03.26: 21.84, 21.82, 21.54, 21.48, 22.07, 22.07.
ie: 21.80+/- 0.76 (mean 21.80 standard deviation 0.25)

The mean value is consistent with a systematic error of say (+0.25 mags) which may be corrected in the calibration settings. Fluctuations are large (typically +/- 0.8 mags at 3 sigma) but can be averaged.

It might be useful having a systems that takes a series of reading (necessarily with only one dark) and averages.

#24 Mauro Da Lio

Mauro Da Lio

    Viking 1

  • -----
  • Posts: 571
  • Joined: 12 Sep 2004

Posted 12 August 2013 - 03:38 PM

Tonight I made 6 measurements in my backyard.

Between hours 22.15 and 22.19: 18.95, 18.83, 18.85, 18.94, 18.91, 18.91.
ie 18.9 +/- 0.15 (Mean 18.9, standard deviation 0.05

My SQM reads: 19.08, 19.11, 19.13, 19.11, 19.12.
ie 19.1 +/- 0.05 (Mean 19.11, standard deviation 0.019

#25 derangedhermit

derangedhermit

    Apollo

  • -----
  • Posts: 1158
  • Joined: 07 Oct 2009
  • Loc: USA

Posted 12 August 2013 - 11:54 PM

If I understand your results
app = 21.84, SQM = 21.64 (app +0.2)
app = 18.9, SQM = 19.1 (app -0.2)

So the app calibration needs to have at least two offset points, since it seems to be at best a linear slope difference v. the SQM, not a constant offset, and 0.4 mag over a range of 2.6 mag is, to me at least, significant.

If this is repeatable, of course.






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics