Jump to content


Photo

Dark Sky Meter Live map now online

  • Please log in to reply
38 replies to this topic

#26 Mauro Da Lio

Mauro Da Lio

    Viking 1

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

Posted 13 August 2013 - 02:38 AM

Yes, but there are two calibration settings for above and below 21. However it is not clear how they works (maybe there are two different algorithms?).

#27 Nop

Nop

    Vendor (iPushto)

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

Posted 15 August 2013 - 08:22 AM

@Mauro,
Your measurements and comparisons are great. I'm wondering what kind of device are you using?
Please note that a unihedron SQM is made for dark sky measurements, with dedicated hardware. The DSM software is made to correct a lot of camera-specific items like gamma correction, dark noise etc. Because an iPhone programmer doesn't have direct access to hardware and cannot stat things like temperature the reading will always be a little fuzzy compared to a dedicated hardware SQM meter.

#28 Nop

Nop

    Vendor (iPushto)

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

Posted 15 August 2013 - 08:25 AM

And a little update about the map. The map visible in the app is live. The map on the website ( www.darkskymeter.com/map ) is updated weekly. This is because i am validating measurements. The map now contains 1500+ measurements. The amount of measurements is growing rapidly because most of the users are in the Northern Hemisphere and darkness is coming back :)

#29 Mauro Da Lio

Mauro Da Lio

    Viking 1

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

Posted 16 August 2013 - 04:09 AM

Hi No, I think the DSM has an advantage compared to the SQM, which is immediate transparent data logging (try to input measurements by hand to the Unihedron Database!).

One feature that would be very useful wold be the possibility to query all reading in a location (all within a give radius) and plot them against time.

I am using a standard SQM. Mine seems to be a bit optimistic. Compared to others it reads on average 0.05-0.1 mags darker.

As for the SQM if it were possible to close the camera diaphragm and shot (ie to take a dark automatically) then it would also be possible to take 5 darks and 5 readings in a row, and by averaging to get a much better estimate.

The spectral response of the SQM is likely to be different from that of the camera, and thus the two devices will respond differently to polluted skie (mostly polluted in red) and pristine skies. So far it seems that if I take 5 readings and average them, and correct for the bias i can get a good estimate of the sky.

#30 derangedhermit

derangedhermit

    Apollo

  • -----
  • Posts: 1153
  • Joined: 07 Oct 2009
  • Loc: USA

Posted 16 August 2013 - 09:33 AM

One feature that would be very useful wold be the possibility to query all reading in a location (all within a give radius) and plot them against time.
[...]
As for the SQM if it were possible to close the camera diaphragm and shot (ie to take a dark automatically) then it would also be possible to take 5 darks and 5 readings in a row, and by averaging to get a much better estimate.
[...]
So far it seems that if I take 5 readings and average them, and correct for the bias i can get a good estimate of the sky.

I too hope this feature comes, and also building a map as the number of readings greatly increases.

I think Mauro meant DSM app above where he wrote SQM. I too hope you can add a function where you automate taking 5 consecutive readings and averaging them together. Then, if calibrated, you seem to get a quite consistent result.

It would be nice to hear about what the calibration actually does to readings of various levels.

#31 Mauro Da Lio

Mauro Da Lio

    Viking 1

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

Posted 16 August 2013 - 11:47 AM

Well, if the dark can be automated, nothing prevents to take continuous measurements (may the battery life L;-) thus getting something like the SQM-LE logs.

#32 Nop

Nop

    Vendor (iPushto)

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

Posted 27 August 2013 - 05:13 AM

Yes, more readings and averaging will make the app much more reliable. The problem still is that Apple doesn't allow us to control exposure and shutter directly, otherwise we could trigger a series of measurements. I'll think about it. The consequence is that the reading will take much longer than a unihedron reading.
We are aware that the number of pins grows every day. Almost 300 readings a week now. So we might port the map to fusion tables (if possible). And yes, i really like to experiment with a query tool. But it will take some time.
FYI: we've split the R, G, B channels in the database, the separate values will be used for evaluation by scientists.

#33 Mauro Da Lio

Mauro Da Lio

    Viking 1

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

Posted 27 August 2013 - 11:47 AM

Hi,

Splitting the RGB channels may bee a good way to evaluate the light pollution. Polluting light is redder that the natural airglow (which is green). Thus the G-R, B-R information is related to artificial skyglow.
See here one example. This was may last might (the one related to the readings above).
http://oi39.tinypic.com/2cxj85u.jpg
This is another site ~30 km south (nearer to populated areas)
http://oi41.tinypic.com/5kgaq8.jpg
The two pictures point the same direction (Aquarius). You can notice the greenish background (airglow) in the first and the reddish background in the second. The sky was 21.6-21.7 in the first (Cygnus Milky Way) whereas it was 21.4-21.5 in the second one year before.

However there is no to link the G minus R reading to the amount of artificial skyglow to my knowledge.

PS There apps like NightCamera, ProCam and Camera+ that look like being able to control the shutter speed. In Particular Night Camera allows to set the exposure time up to 1 s and reduce di ISO speed. If it takes longer than the SQM that is not much important compared to advantage of having automatic logging (and in the future the ability to study LP in one location). This is a feature that can be obtained only with a SQM-LE and it is not so simple (I mean I do not want to bring my MacBook at minus 20° to find out that the battery dies).

#34 Nop

Nop

    Vendor (iPushto)

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

Posted 28 August 2013 - 03:53 AM

Hi Mauro,
Thanks for your interesting links! My app sums up all channels (except l). It might be interesting to experiment with addition of only G-R or R-B, and see if results get better. Apple probits long exposures (by simply not letting us developers control the exposure time). My workaround is stacking (and averaging) single frames. I suspect that other apps do that too. Harro found another interesting 'issue'. The FOV of the DSM is so small that a single bright star (say Wega) can influence the reading with 0.2 . So we'll start experimenting with a new version and a slightly wider FOV and separate channels. ( I already started with a gamma correction). that way we hope to improve the readings and the reliability. Which is not bad wih multiple measurements.

#35 Mauro Da Lio

Mauro Da Lio

    Viking 1

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

Posted 28 August 2013 - 05:13 AM

Hi, I suspect there is some workaround for the exposure time. Have a look at Night Camera: you can set up to 1 s time (may write them).

As for the FOV issue, why not summing up all pixels in say two-thirds central area of the detector (leave out the border which might be more noisy and suffers vignetting). By summing the signal of more pixel you may hopefully reduce noise.

Having three readings for R, G and B might be a plus compared to the SQM. In particular the difference G-B might be very informative of the actual amount of artificial light.

I am waiting to see the update.

#36 Nop

Nop

    Vendor (iPushto)

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

Posted 10 September 2013 - 02:43 PM

@Mauro, thanks for the info. If you want to be a beta tester please let me know, i'll be happy to send you test versions!
@All
The map has been updated and zoom levels (groupings) have been implemented. Now with more than 2000 measurements! Dark skies are returning :)

#37 the1andonlyfinn

the1andonlyfinn

    Explorer 1

  • -----
  • Posts: 58
  • Joined: 17 Jul 2011
  • Loc: Rancho Cucamonga, California

Posted 27 September 2013 - 05:39 PM

Question: I downloaded the Pro app and have submitted readings for Rancho Cucamonga California and Fontana California, they show up on MY map on the app, but online neither reading shows up. How long before they go onto the "big map" ?

#38 Nop

Nop

    Vendor (iPushto)

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

Posted 30 September 2013 - 02:21 AM

I'm updating the map on the site by hand (mostly every day).
So stay tuned. I'm working on an automated process for validation and submitting. Until then the map is 'semi-live'

#39 the1andonlyfinn

the1andonlyfinn

    Explorer 1

  • -----
  • Posts: 58
  • Joined: 17 Jul 2011
  • Loc: Rancho Cucamonga, California

Posted 30 September 2013 - 08:57 AM

Thanks! You might see a bump in readings from Southern California: I've been wandering about the Inland Empire to "flesh it out".






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics