Tuesday, April 22, 2014

The Loss of the Night app is 1 year old!

The Loss of the Night app went into the play store on April 22, 2013, so today is the app's first birthday!

Since that time, we started this blog, we held our first Flashmob for Science, we upgraded the app to support 11 languages, and most importantly, we've had over 11,000 observations submitted! Most of the observations were taken by people using the app under non-ideal conditions (e.g. with clouds or the moon out), but tonight we will probably reach our 1,500th observation with data that we can use for analysis!

The app's birthday means that we are going to start to have data taken at the same location in multiple years. The major scientific goal of this project to understand how the sky is changing as the world transitions to LED lamps. This is something we can't measure with the current Earth night observing satellites because they are not sensitive to the blue light that is typical of the most energy efficient LEDs. (And also because they look at the ground, not the sky!) We can only get these measurements by having people from around the world look up.

I think that 1,500 observations in the first year is a great start, but we really need your help to expand further! The more observations we have, the more accurate our measurement of the change in the night sky will be, and the more quickly we will be able to report it. So please, tell your friends and family about the app, post a link to the download site on Facebook, and the next time you have a clear night sky in your city, please go out and do a measurement!

Other ways you can help:
  • We would love to hear back about your experience with the app, especially if you had any problems, or if you have recommendations for how we can improve it! Please comment here
  • We are looking for translators for future releases.
  • We want photos of good and bad lighting for the view from your app series.
  • We are looking for a landscape photo taken in an area with no light pollution on an overcast moonlit night.
You can now view the data you submit on the GLOBE at Night map page. It should appear there within a few minutes of your data being transferred. If you zoom in and find your point, you'll get a bit of feedback about how bright the sky was, and the range over which you are uncertain whether stars are visible or not. (Please note that this is a preliminary analysis, and in some specific cases the algorithm that calculates the range doesn't work properly.) This information will be available to you immediately in the next release of the app (we are still working on securing funding).

Thanks again to everyone who has contributed data! Without you, there would be no project!

Dark Sky Meter app is currently available for $0.99 download!

Great news, tell all your friends with iPhones and iPads!

Our "Loss of the Night" app was publicly financed, and is available for free download for Android devices:

Monday, April 21, 2014

Representations of the moon in art

It's day 2 of International Dark Sky Week, and the moon is fading towards a crescent, visible only in the early morning. That means that it's a great time to use the Loss of the Night or Dark Sky Meter apps, or to contribute to GLOBE at Night using the regular webapp. To celebrate the crescent moon, here's a post about depictions of the crescent moon in art.

If you saw a painting of a scene that included a coniferous tree filled with blossoms or fruit, it would be incongruous, and unless it was done intentionally it would probably somewhat spoil the painting. Similarly, if shadows are drawn incorrectly it can distract from the image, which is why shading is taught to aspiring artists early on. Unfortunately, these lessons often don't extend to the shading of the moon, and representations of the moon in contemporary art are often (or more likely usually) incorrect. (Check out these tips to learn how to draw the moon correctly!)

The moon is most often depicted as a crescent, and there are two problems that frequently pop up. The first is that the real crescent moon is lit from the very top to very bottom:

Crescent Moon by Thomas Bresson from the Wikimedia Commons

Many artistic depictions of the crescent moon extend considerably past the top and bottom of the moon:

Golden Crescent Moon

Golden Crescent Moon by Dcubillas from Wikimedia Commons

The second problem has to do with the orientation of the moon and the surroundings. If you think of the crescent moon as an arrowhead, it always points towards the sun, and because the crescent moon occurs near the new moon, it's always close to the sun in the sky. So a "D" shaped crescent moon is low in the sky and sinking after sunset (in the Northern Hemisphere), and a "C" shaped crescent moon rises shortly before the sun. (Apparently, right handed artists prefer to draw the C shape, making their images look like early morning rather than evening).

Once you know a bit about the moon, incorrect depictions jump right out at you, and can spoil the art. For example, in the generally excellent "Conqueror" series of books about Ghengis Khan, the moonrise or moonset were often incorrectly described, which isn't so pleasant in "historical fiction". In one particularly bad case, the crescent moon rose shortly after the sun set (not possible!), and the opposing general took it to be a good omen. Since the plot could hang on a detail like this, it's unfortunate when more care isn't taken.

I have suspected for quite a while that before the advent of electric lighting, people would have been more familiar with the comings and goings of the moon, and it's made me wonder if depictions of the moon in art are were more likely to be correct in the past.

Over Easter, I visited St Mary's Church in Stralsund Germany to listen to an organ concert. While we waited for the concert to begin, I took photos of two of the church's paintings that caught my eye because of the presence of the crescent moon. The first photo shows a burial at sea:

"Seebestattung" (Burial at sea) - unknown artist, ca 1900, oil on canvas
The crescent moon is correctly depicted: it points towards the sunset in the East. We can tell it's sunset and not sunrise because the painting most likely represents a burial in the Northern Hemisphere. Since the moon is never near the northern horizon, it follows that the right hand side of the painting must be the East.

I don't know the name or any details of the second image:

Here you can see that the crescent moon is again correctly depicted, with the lit part running exactly from top to bottom, and the moon "pointing" straight to the sun. Some artistic license is used here, because stars are shown with the moon while the sun is up, but (I think) the two sides represent before and after Adam and Eve ate from the tree. (It's also interesting to me that living in Eden is depicted as being in the night, while being cast out of Eden is under the sun.)

Of course, artists are free to depict the moon (and pine trees) any way they want. But if it's going to be depicted differently from nature, artists should be aware that some segment of the population will note the incongruity, and it may spoil the work for them.

Friday, March 21, 2014

GLOBE at Night turns 8 years old!

The Loss of the Night app came about because of my research into the citizen science data from the GLOBE at Night project. The first GLOBE at Night data came in on March 22, 2006, so a few hours from now GLOBE at Night is going to turn 8 years old! Congratulations to Connie Walker, and everyone else on the GLOBE at Night team at NOAO!

The third GLOBE at Night campaign of 2014 begins tonight, and even if you prefer to use the Loss of the Night app or the Dark Sky Meter app, I hope that you will take a moment to take part in GLOBE at Night this month. The reason is that the time series for GLOBE at Night stretches back further, and observations taken using the same method are much easier to compare to each other. Since the original GLOBE at Night campaign began in March, additional observations from March each year are more important than those in any other month.

So if it's clear where you live, please take a few minutes to go outside and look up at the stars in the next ten days!

Wednesday, March 5, 2014

Skyglow over South Africa

Markus Pössel from the German Haus der Astronomie recently returned from South Africa, and realized that some of the photos he took nicely demonstrate the effect of skyglow. What's really great about the photos is that they not only have the same camera settings, he also set up the shot to be of the same part of the sky. He gave me permission to re-post the photos from his SciLogs post here.

He took three photos and overlaid them one above the other. The top photo shows the sky over Grahamstown (population 70,000). The middle photo has the same settings as the top, but was shot in Sutherland, near the South African Astronomical Observatory. The bottom photo is the same location as the middle photo, but with an exposure time twice as long to reveal more details.

Creative Commons License
An example of light pollution by Markus Pössel is licensed under a
Creative Commons Attribution-ShareAlike 4.0 International License.
The top two photos have identical camera settings,
the bottom photo had an exposure time that was twice as long.

The area of the sky that was photographed. The Southern
Cross is at the bottom. Constellation overlay via astronomy.net

You can get the full resolution version of the images at Markus's SciLogs post.

If you have a good image about skyglow or lighting, please send it to me, and I'll share it on the blog! You can see the entire light pollution photo series by clicking on the "view from your app" label below.

Monday, March 3, 2014

Multiple observations at the same location

I recently got a question via twitter:

Thanks for the question Aashay!

The answer is that multiple observations are extremely valuable to us, especially at this early stage. There are several reasons for this:
  • Observations from multiple people at the exact same time are very useful, because it helps us understand how different people perceive the night sky and how they interact with the app.
  • Observations from the same person at different times on a single night help us understand curfew effects (the sky over most cities gets darker as the night goes on).
  • Observations from the same person on different days at about the same time within the same month help us understand how consistent the app's results are with changing atmospheric conditions and clarity, as well as with the changing stars that it asks you for (there is some randomness built in). 
  • Observations from the same person several months apart are helpful for us because the stars change with the seasons. We are also trying to understand which stars are easier to decide upon and which are problematic, so the more stars you observe the better.
And finally, and most importantly, the main purpose of the app is to track long term changes in skyglow worldwide. For this to work, you need to sample the same location in a future year.

So please feel free to make observations as often as you like and observe as many stars as you like (going past 7 stars is helpful). But most of all, please plan to come back to the same location in the future and see if the sky has changed!

Wednesday, February 26, 2014

Passed 10,000 submitted observations!

The Loss of the Night app project passed an arbitrary (yet still exciting) milestone at the end of January: The 10,000th observation submitted to the GLOBE at Night server.  The number includes submissions with the test version of the app: we will probably cross 10,000 submissions by people running release 1.0.0 or higher tonight.

I promised to send a holographic postcard to the citizen scientist who submitted the complete observation closest to #10,000, but unfortunately the user is in Japan, and I think my (English) email informing the person that he or she has won a prize and asking for his or her mailing address may have been flagged as spam...  I am still trying to contact the contributor, but for now, let's celebrate observation #10,005, which was the second closest to #10,000.

Data from observation #10005

The observation was made inside of Luxembourg City, and was especially useful because 15 stars were observed instead of the minimum 7. Stars which were seen are marked in black, stars which were invisible are empty red circles, and stars that the user wasn't sure about are blue. The best fit for the naked eye limiting magnitude was 2.55 (solid line), with an estimated range of +/-0.6 over which the observer was likely to have trouble deciding whether the stars were visible or not.

In the next version of the app, we intend to provide instant feedback to you about what the naked eye limiting magnitude is at your site, and how consistent your results were. We will also do a much better job of sampling both sides of the estimated NELM, and we will get rid of most of the pesky stars that are hard to identify (did you hear that Draco?).

Locations where the app has been used.
Above is the most recent plot of all of the places on Earth that people have used the app (up to February 5). The black dots mark the 1208 places where the observations can be used for our analysis (no clouds, no twilight, no moonlight, and at least 7 stars observed), and the red dots show locations where this criteria wasn't met. It's fantastic that so many of you from around the world are contributing to the project, and I hope that we'll continue to increase add new citizen scientists as the year rolls on! Thank you for participation!