Observing with Street Lights

Observing with Street Lights
Dark sky sites not always necessary to see the Milky Way (This image was taken ouside of a B&B in Julian, CA)

Wednesday, December 6, 2017

Super moon time again!; Finally got another useful Photoshop lesson, Hooray!; Measuring the night sky brightness (m = 14.3) right here in city lights

Greetings from Palmia Observatory


Well this has been a cloudy week, but a few stars and the super moon were able to peek through.  Check out the images of the super moon below and the availability of a few stars and clear regions of space allowed for an attempt at measuring the sky brightness in these city lights polluted skies.


This view of the super moon was taken with just the easily transported camera tripod and 600mm telephoto lens.  It is so bright that just a 1/1000 second exposure is needed.  This image seems to be more yellowish that I remember and might be just an artefact of enhancing the dynamic range of the image in Photoshop, even though no change in the color was intentional.


Super Moon, 600mm, 1/1000 second (Source: Palmia Observatory)
Super Moon, 600mm telephoto, 1/1000 second (Source: Palmia Observatory)


The day after the super moon another opportunity to capture the moon just between the trees came up and this was too good to pass up.  However, I found that the exposure time to capture the background trees, which in this case was 1/3 seconds, was way too long to capture the post full moon, which only required 1/1000 second exposure.  I wanted to have one photograph that showed both the detail of the moon, at 1/1000 second, and the background trees, etc., at 1/3 second.  What to do?  Well, I finally bit the bullet and signed up for a an hour lesson in using Photoshop.  This turned out to be a very productive investment.  I signed up with www.TakeLessons.com and arranged for Photoshop Expert, Zack, to come out and let me in on the secrets of how to do that.

Check out the image below which shows my first attempt, with some coaching from Zack, of cutting and pasting the 1/1000 second exposed moon image into the 1/3 second exposed background image.

Composite image with background 1/3 second and moon at 1/1000 second (Source: Palmia Observatory)
Composite image, 300mm, with background 1/3 second and moon at 1/1000 second (Source: Palmia Observatory)

Hey, I think it looks pretty cool, as my first attempt!  This composite image is just what your eye would have seen and no intentional change in the size of the moon was done.  There are a few rough edges around the circular cutout of the overexposed moon, but it mostly works.  Thanks for all your help, Zack!  By the way, if you recognize the need for getting a little bit more help in Photoshop, I can certainly recommend Zack, who displayed a great level of experience and professionalism and ability to gently guide the amateur user of Photoshop.


Ok, the other opportunity that came about was a chance to again try to measure the sky brightness.  The method was to compare the brightness of an area of sky that contained a known magnitude star, but was not expected to contain any Milky Way or ordinary meteorological clouds.  The selected location included Vega, a known magnitude 0.0 star, at this time located in the West toward the Los Angels lighted polluted sky.  The screenshot below shows that the AIP4WIN Star Image Tool finds a total ADU camera value of 130,064 for the 300mm focal length, 1/30 second exposure with Vega off to the left.  The profile line across the image shows no other obvious objects in this short exposure.

Using Vega (m = 0.0) in this 1/30 second exposure as the reference star (Source: Palmia Observatory)
Using Vega (m = 0.0) in this 1/30 second exposure as the reference star (Source: Palmia Observatory)



Next, the camera exposure time was manually controlled for 90 seconds.  In the image below, you can see star trails associated with this long exposure on a non-tracking tripod.  Vega is the heavily saturated track near the left edge of the image.

There is also some strange dark structure that apparently is some sort of camera artefact, but I have no idea what it is.  Anybody out there with a clue as to what that is?

90 second exposure, 300mm of Vega (Source: Palmia Observatory)
90 second exposure, 300mm of Vega (Source: Palmia Observatory)


Anyway we can analyze this photo image and get an estimate of how bright the sky is.  Look at the yellow profile line in the AIP4WIN screenshot below and notice how the selected line neatly bypasses any stars, but still goes across the dark structure.  The amplitude along the profile line shows up in the panel on the right hand side.  Hey, this bell curve type shape is good news in that this type of curve is how a normal lens responds to a uniformly illuminated background, due to vignetting, in which the center of the lens is more intensely illuminated than is the sides of the lens.  Note also that the dark band structure again shows up as a decrease in illumination when the profile line crosses it.


Profile line for sky near Vega 90 second exposure (Source: Palmia Observatory)
Profile line for sky near Vega 90 second exposure (Source: Palmia Observatory)


Now the AIP4WIN data file for that selected profile line, with over 5000 data points, can be averaged to give us the average sky brightness value.  The chart below shows the filtered version of that profile line, which has overall average of 640 camera ADU (Analog Digital Units).  This value is the average over the profile line after the camera offset dark 90 image value of 2046 ADU is subtracted.  The nominal bias value that is associated with each camera pixel is 2048 ADU for no light at all, but the measured value for 90 seconds turned out to be 2046.  Yes, I was outside in the dark, and yes, I just put the lens cap on for 90 seconds and let the camera count for 90 seconds.

The filtered (moving average over four pixels) land adjusted for the zero offset ight curve is shown below.

Profile line for sky near Vega 90 second exposure (Source: Palmia Observatory)
Profile line for sky near Vega 90 second exposure (Source: Palmia Observatory)

So with this average sky brightness value of 640 ADU, we can now use the reference star Vega to calculate what the sky brightness is in magnitude units.  So starting with Vega measured value and adjusting for the shorter exposure time, we find that if Vega had been exposed for 90 seconds instead of just 1/30 second, then the total ADU units would have been 130,064 * 30 * 90 = 351,172,800 ADU.  So this average sky is 351,172,800 / 640 = 548,707 times dimmer than the magnitude 0.0 star, Vega.  Converting this difference to magnitude, we see that the differential magnitude = 2.5 log10 (548,707) = 14.3.  Hey, this first measurement of the sky brightness in the direction of Los Angeles is magnitude = 14.3, since Vega is magnitude = 0.0.

No wonder we can't see the Milky Way here in these skies because the sky is much brighter than the estimated Milky Way brightness of about 20!  Check out the November 12, 2017 blog post where this measured estimate of the Milky Way was made using an image taken in Casper, WY at the time of the total solar eclipse.

Now this first estimate contains some other issues that have not been corrected for, such as the fact that the measurement of Vega is off to one side of the camera image and it itself will contain some vignetting error.  But this first estimate seems ok as a first pass attempt.  I wonder how the measurement would have differed if I had picked other reference stars in the East, North and South?  Wouldn't the sky away from Los Angeles be found to be a bit darker?  Oh well, I didn't consider these issues at the time and didn't make these other measurements.  Maybe one of you other amateurs can make the measurements and let us all know?


Until next time,


www.palmiaobservatory.com







No comments:

Post a Comment