Wednesday, July 1, 2020

Changes in light emissions during the COVID-19 shutdowns

At the start of the lockdown periods in Europe and North America, a few people speculated about what it might mean for changes in light emissions. I was skeptical that we would see much of anything.

Shortly afterwards, NASA showed some results for Hubei province in China:

We are still waiting for the monthly satellite data composites for February-April to come out, but I still don't expect to see much in most places. However, it turns out that there were some pretty dramatic changes in at least a few places around the world. The best example I've seen of this comes from Poland. Check out these two photos of Kraków, Poland, that Mateusz Windak posted on a Polish forum:

This work by Mateusz Windak is licensed under a Creative Commons Attribution 4.0 International License.
Click for the full size version.

Here's what Mateusz had to say about the photos:

The animation shows differences before and after midnight, when the Cracow city lights are switched off. After the switch off, only private light sources and the highway in the foreground remained. The lights were turned off for a month and a half from 00:00 to 04:00 each night due to the lack of traffic during the lockdown period of the COVID-19 pandemic. Currently, Cracow has stopped shutting down the lights, but some other cities are still continuing this practice.
The photos were taken from Wieliczka, a neighboring city, using a Mavic Air drone. They are four panel panoramas, the settings for each photograph are: 2s, ISO1600, f/2.8.

Thanks for sharing your photos Mateusz! I find it really impressive that in addition to the lights going off, you can also see the decrease in sky brightness

Friday, June 19, 2020

Panel discussion about light pollution and dark skies

A while back I was invited to be part of a panel discussion titled "Light Pollution & the Dark Sky Movement". Check it out:

[d]arc thoughts: Episode 4 | Light Pollution & the Dark Sky Movement from [d]arc media on Vimeo.

Thursday, June 11, 2020

Loss of the Night app video tutorials

A group of undergraduate students from Worcester Polytechnic Institute recently produced tutorial videos for using the Loss of the Night app. They made separate videos for the Android and iOS versions. Feel free to share them!



Big thanks to Aaron, Kai, Ryan, and Vinit for producing these videos!

Tuesday, June 9, 2020

Removing natural light from nighttime satellite images

My student Jacqueline Coesfeld recently developed an analysis that will be interesting for readers of the blog who work with satellite data. I summarized it in a series of tweets. You can click on the tweet to read the thread, or see the unrolled version here.

Tuesday, March 24, 2020

Street view with the lights out

Ken Walczak from the Adler Planetarium in Chicago recently sent me a set of two cool photos I want to share. Here's what he had to say about them:

I got home after an evening event and didn't realize at first why there was such a different quality to the ambience of our neighborhood. It took a moment to realize that due to a film shoot, the city had turned off all the lights on the main streets. I walked the dog, said "Hi" to neighbors walking down the street, just as normal and safe as a typical night. I snapped some photos. I set the exposure to try and capture the visual quality of the experience (1sec f/3.5 at ISO 800). The lights came back on later in the night and I shot with the scene with the same settings.

This work by Ken Walczak is licensed under
a Creative Commons Attribution 4.0 International License.

This work by Ken Walczak is licensed under
a Creative Commons Attribution 4.0 International License.

Animated version

In an area with significant skyglow, I have often found that you actually have the better vision in unlit areas (e.g. parks, rooftops) than on lit city streets. The reason alleys in big cities are dark is not generally because there isn't enough light for you to see, but because of either glaring lights creating more contrast than your visual system can deal with, or else because you've left a really bright street and your eyes haven't had a chance to adapt.

In case you didn't see it already, check out the other amazing photo of clouds glowing over the Chicago skyline Ken sent me a few months ago.

Wednesday, February 26, 2020

How not to light a staircase

Staircases represent a tripping hazard, and it is therefore not unreasonable to illuminate them under certain circumstances. However, as I have shown in the past, poorly designed illumination is actually much worse than no illumination at all. A (literally) brilliant example of this comes from the "Lustgarten" park in Potsdam:

This work by Christopher Kyba is licensed under a
Creative Commons Attribution 4.0 International License.

These lights combine both the wrong direction (shining horizontally rather than downward) with being extremely bright, which makes them terribly glaring, and makes it harder, rather than easier, to safely ascend the staircase. (Right after I took this photo, I misjudged the position of the first step and stumbled slightly, even though I was trying to shield my eyes from the glare).

What could be done instead? One interesting application I have seen is LEDs installed on the underside of a handrail. Unfortunately, I don't have any photos of a really good execution of this. Here is an example of an attempt that didn't quite work out:

This work by Christopher Kyba is licensed under a
Creative Commons Attribution 4.0 International License.

It's better than the example above, but is still a problem, because it's still too bright near the lamps. What pedestrians need is accent illumination that helps them separate the stairs from the surroundings. When you have most of the scene lit with ~0.1 lux from skyglow and other distant light sources, and you then put 50 lux directly under the handrail, you've now created a 500:1 contrast. Your visual system doesn't deal well with that level of contrast, and it makes it harder to see.

For areas that are not brightly lit, moonlight provides a good benchmark. A typical full moon night is around 0.1-0.2 lux, so if you're illuminating a fairly light concrete, you really shouldn't need any more than that.

Monday, December 2, 2019

Second action of the Nachtlicht-BüHNE outdoor lighting project

Nachtlicht-BüHNE is a co-Designed Citizen Science project related to light at night, funded via the CitizenScience@Helmholtz initiative of the Helmholtz Association in Germany.
Within the project, citizen scientists will be developing two apps. The first, coordinated by the German space agency (DLR), is about improving the system for reporting sightings of fireballs in Germany. The second, coordinated by the German Research Centre for Geosciences (GFZ), is about creating large-scale inventories of all outdoor lighting. This means we are interested not just in street lighting, but also private lights, lighting for advertisement, industrial lighting and so on. For more detailed background, see the previous blog post from September 5th

This work by Markus Schönrock is licensed under
a Creative Commons Attribution 4.0 International License.
To get to the point: there are no exhaustive databases categorizing all outdoor light sources (in contrast to the situation for streetlights, for which there are some public databases). In order to understand the radiance observed by satellites, we therefore need information about what types of light sources are producing the light. We would also like to understand how the types of light source change depending on whether the satellite is viewing a small town, suburban areas, or city centers. This project aims to develop an app to allow citizen scientists to record and count outdoor luminaires, creating one of the first datasets of this kind…you can be part of it!
Before we can develop the app, we need to figure out how to make it work well. We are planning on producing two versions app the app: a Basic version with the minimum requirements, and a Pro version that allows more detailed information to be recorded. We are starting with the Pro version, and this is where we need your help right now:
We have a special form for you to record the data. You should send us your record by January 19th, 2020 to Also, please feel free to send us your feedback about what worked, what didn’t, and any problems and further ideas you have about our approach.
Here are the detailed instructions:
  • Find a friend
  • Print out our paper form
  • Have a look at the form from start to end, so you that you are familiar with the different categories and attributes of the luminaires. (For example, we also count illuminated windows, which might at first sound crazy, but actually does not take that long and turns out to be really important.)
  • Choose a starting and ending location (typically a single street from one intersection to another).  
  • Get a pencil or pen, and then take the form and meet your friend at the starting location.
  • Write down the time you started.
  • Start walking slowly down the street, recording ALL the light sources you can see. We find it works best for one person to spot the lights, and the other to record which type they are.
  • When you reach your stopping point (eg. the next intersection), write down the time you finished
  • Optional: do another street! (But with a new form, please!)
  • When you get home, type your observations into this Excel form, and save it.
    • Record the coordinates of your segment on the form in decimal degrees (WGS84, for example 52.5163N, 13.3777E). You can find the coordinates using Google maps or Openstreetmap.
  • Send us your data and your thoughts about this paper version to by January 5th 2020.

Finally, some additional hints regarding the lamp categories:
  • Most of the time we count the number of luminaires we see – if a street lamp more luminaires (e.g. a  candle tree with two arms) we count each of them individually (so two)
  • Sometimes counting is not enough in categories where the luminaires vary greatly in size (e.g. illuminated signs outside of shops). In this case, we have three size categories, and you count the times the luminaire fits into them (e.g. a sign roughly as big as two hands counts as two hand-sized signs)
  • Full cutoff luminaires: this is a lighting fixture that projects all of its light in a downward direction – it can sometimes be tricky to tell if a street light is very high up. Here some hints to help you with determining if a light is full cutoff:
    • The light source is fully inside of the fixture, it does not hang below it.
    • The glass below the lamp is entirely flat, not curved
    • The fixture is directed straight downwards (not tilted at an angle) 
    • No light should shine above the height of the lamp (if a nearby house or tree is lit at a level higher than the lamp, it’s not full cutoff) 
  • Normal versus bright: humans cannot reliably estimate small differences in brightness, but we can identify particularly glaring light sources. So we want to record separately light sources that seem particularly bright to you compared to the rest in your street segment (e.g. the lighting at a zebra crossing, an extraordinarily bright shop window or sign etc.) 
Paper form
Excel form
Do not make observations alone! With two people, you will do a better job of keeping track of your surroundings. But it’s also much easier and a lot more fun to do it with a friend or family member! 

This work by Markus Schönrock is licensed under
a Creative Commons Attribution 4.0 International License.