Milky Way Visibility Maps
Sarah Kurelowech
Carnegie Mellon University - Department of Physics
August 9th, 2020
Project Information:
I, Sarah Kurelowech, am conducting research on the points around a city or town in which the Milky Way can be seen. I am conducting this research under Diane Turnshek and Stephen Quick, both of Carnegie Mellon University.
I hope to analyze light pollution maps in order to gain an understanding of the shape that the visibility boundary makes around a city. I also wish to determine if the shape is in any way influenced by the terrain surrounding the city and, if so, how it is influenced.
The research began on July 24th, 2020 and will continue through the end of the summer session. I would like to have the project continue longer than the scheduled class time, if possible.
Objectives:
1. Determine where the Milky Way can be seen at points around a city.
2. Analyze how the terrain around a city influences the visibility of the Milky Way.
Methodology:
1. Determine and define at what light pollution level the Milky Way can be seen.
2. Determine which cities to analyze.
3. Look up these cities using multiple light pollution maps.
4. Determine which colors on the map correspond to the defined light pollution level.
5. Take a screenshot of the map being sure to include the area in which the Milky Way can be seen.
6. Carefully trace the boundary line.
7. Repeat for all desired cities.
8. Analyze the results.
Internship Role:
During my time working on this project, I created maps with boundary lines indicating the point in which the Milky Way can begin to be seen around a city. In addition to these maps, I measured the distance between the city center and the boundary for each cardinal direction.
To begin my tasks, I first determined how to tell where the Milky Way was visible. On the Bortle Scale, at a value of 5, the Milky Way is barely visible and appears washed away. I decided to use 5 as the value at which I would say the Milky Way is visible. I assumed that most people’s eyesight would be decent enough at this value to see even a washed away Milky Way. In my opinion, this is considered visible.
I then had to convert this to units of mag/arcsec^2 so that the values would correspond to those on the light pollution maps. 5 on the Bortle Scale was found to be 19.1-29.4 mag/arcsec^2. I used this range of values to look at the legend on the light pollution map and determine the color in which the Milky Way could be seen.
I began by using two light pollution maps1 and compared their outputs to determine the optimal map to use for the project. In comparison, the first map source that I used appeared to
have more detail about small-town light pollution. On the other hand, it seemed that the second link took the city data and then extrapolated it out to the small towns without actually taking data in the small towns. Since I am gathering images for both small towns and large cities, it is necessary that the small-town maps be as accurate as possible. The following figure shows the difference between the two sources in a small-town case.
I have been able to gather map screenshots for 23 cities so far. With these screenshots, I drew the boundaries of where the Milky Way could be seen. These images can be seen in the results section. I then used the map to determine the distance between the city center and the edge of the boundary in each of the cardinal directions. This is to give a sense of really how far out of the city a person must be to see the Milky Way. This data can also be found in the results section. After this, I took map screenshots of the terrain in each of the cities, which are also visible in the results section. I use the terrain to explain the shape of the boundary.
Results:
In the following figure are the light pollution maps for the cities that I chose along with the Milky Way boundary lines and the surrounding terrain. It is important to note that the Milky Way could possibly be visible inside of these boundaries. More research is needed to determine why this happens or to define better boundary lines.
I then put the distance data into an Excel spreadsheet and plotted the distances I measured using the light pollution map tools. The table defines each city number and the plot shows the distance from the city center in each cardinal direction.
As we can see, there is a trend where the bigger cities have their boundaries further out than small towns. It was also found that there was a direct correlation between the terrain and the shape of the boundary line. Typically, the boundary line follows the shape of the terrain.
These findings are documented in the Excel spreadsheet entitled Dark_Sky_Research in the Data
Sheet tab. Only initial comments were made regarding the boundary lines’ connection to the surrounding terrain as the project had a time crunch.
Further Research:
There is much more research to be done here. I highlight the things that the next researcher should do to continue this project below.
Use the Python program edge_detect to get more accurate boundary lines. This program was written by Ela Gulsen. Any questions regarding the program, please contact egulsen@andrew.cmu.edu. I have included the image that the program outputted for the city of Pittsburgh. As we can see, the boundary line is much cleaner and more well defined than in the hand-drawn version shown earlier in the report.
Determine a mathematical correlation between different terrain features and boundary lines. For example, how much does a water feature dampen Milky Way visibility in comparison to a mountain range?
Overlay the light pollution maps and boundaries on the terrain maps. This would be a great visual tool to see just how much the terrain defines the boundaries.
Determine if the size of the city is proportional to the size of the boundary.
Collect data on more cities. This will allow for a more comprehensive understanding of the subject.
Knowledge Gained:
By completing this project, I was able to learn more about the way research is conducted. I have never had the opportunity to conduct research here at CMU because I was never really interested in it, but after this course, I think I have more of an understanding and appreciation of research.
Additionally, I was able to connect with more people at Carnegie Mellon from various departments. Networking is a large part of professional development, and professional development is very important for a career. For example, I have had the chance to work with Ela Gulsen, a math major at CMU. She has been able to write a Python program to draw the boundary lines with more detail and precision. This program will be used in the next stages of this project.