Introduction: One of the fundamental areas of the environmental field is analyzing the effects of various types of pollution. However, one specific type of pollution that does not get enough attention in the environmental field is light pollution. Light pollution is defined as the excess glow of artificial lights and can include urban sky glow, light trespass, glare, and clutter (Light pollution—what is it and why is it important to know?). The most obvious impact of light pollution on the environment is the inability to see fainter stars and the Milky Way galaxy in the night sky over urban areas. While this may seem insignificant compared to the benefits of light at night, recent research has suggested that light pollution may disrupt the circadian rhythm of humans and animals, particularly animals that are more active at night (Light pollution—what is it and why is it important to know?).
In recent years, many cities are in the process of converting their streetlights to LED lights, which emit blue light and are more energy-efficient. However, these lights are suspected of causing more light pollution than traditional artificial lighting. A new report issued in 2016 by the American Medical Association finds that exposure to LED lights, which are five times more disruptive to circadian rhythm than traditional lights, is associated with increased risks for cancer, diabetes, heart disease, and obesity (AMA Report Affirms Human Health Impacts from LEDs). This study focuses on the spatial extent of both LED streetlights and light pollution. The ultimate goal of the project is to reduce the amount of light pollution, while also respecting energy and money savings.
Objectives: The purpose of this project is to determine if a correlation exists between the spatial location of LED streetlights and the spatial location of high light pollution levels. If such a correlation is found, data on energy and monetary savings can be analyzed in an effort to find balance between saving energy and emitting less light pollution.
Scope: The city of Los Angeles (CA) has been selected as the city to do this project on. It seems Los Angeles is one of very few cities with open data on streetlight locations by type. What sets Los Angeles apart from other cities is that it also has data available on energy and money savings associated with these streetlights. While light pollution affects every urban area, one large city should be enough of a region to determine the spatial correlation between LED streetlights and light pollution. Once this initial study has been done, the same type of study can be done on other cities with sufficient data on streetlights, in an effort to reduce light pollution.
Point vector data exists on the location of every streetlight in the city of Los Angeles, including whether the streetlight is LED (Streetlight Locations). The city of Los Angeles also has data on energy and monetary savings associated with converting to LED lights (Citywide LED Streetlight Savings). This data does not have spatial information and is divided by council district. For this dataset to be used, a vector shapefile of Los Angeles council districts must be found, and the data must be joined to it (by attribute—council district number). Satellite imagery will be analyzed and a raster dataset of light pollution over the Los Angeles area will be created. This raster dataset can easily be converted to vector data (contours) if the need arises.
Tasks: The first task should be to generate a raster map of the Los Angeles area using satellite imagery. This method was done by Oakland County (MI) to generate such a map for the Detroit area. Oakland County used data from NASA and NOAA to generate their map (Light Pollution in Oakland County). It is advisable to contact the people who did this analysis to find out exactly what their method was, then use it on nighttime satellite imagery of the Los Angeles area to generate a raster map of light pollution for this area.
The second task is to integrate this raster map with the vector map of streetlights in Los Angeles, and determine if a correlation exists between areas with the highest amount of light pollution and areas dominated by LED streetlights. The raster map of light pollution may be converted into a vector map to make this task easier, if necessary. Select the number of LED vs. non-LED streetlights contained within areas of high vs. low light pollution by selecting all data that contain “LED” in the “LAMPA”, “LAMPB”, “LAMPC”, “LAMPD”, “LAMPE”, and “LAMPF” data columns. The easiest way to tell them apart from other lights in the analysis may be to export the selected data as a new layer, then do the same for all data that do not contain “LED” in any column. The original point vector layer can be removed, and the streetlight data are now expressed in two layers—one showing LED lights, and one showing all other lights. Using these data to create graphs may be the best method to determine whether a correlation exists. Graphs may show the number of LED vs. non-LED lights in areas with high, medium, and low relative light pollution, for example. The determination of whether or not this correlation exists should be the primary goal of this project. An optional, more detailed analysis may be done using the wattage numbers, which are also provided in the dataset. However, it may be better to save this type of analysis for a subsequent investigation.
Mapping the data is a good way to visually show the results. The light pollution map should show color differences as symbols for levels of light pollution. LED and non-LED streetlights can also be made different colors to distinguish them from each other, as well as from levels of light pollution. The result would be a nice visual map of LED and non-LED lights over light pollution levels for the city.
Once these steps are done, the answer to the fundamental question should be found. If there is no correlation between LED streetlights and higher levels of light pollution, or if there is a correlation between LED streetlights and lower levels of light pollution, this investigation is complete with the conclusion that LED streetlights not only save energy, but they are also better at limiting light pollution. Based on the current research of this topic, however, the most likely result is a correlation between LED streetlights and higher levels of light pollution. If this indeed turns out to be the case, this project can be expanded to determine the spatial extent of the city’s savings in both money and carbon emissions, and how much more light pollution is being emitted as a result.
The data on the city’s energy savings is already available, but since it contains no spatial identifiers, it must be joined to a layer that contains Los Angeles council district numbers and their spatial location. Once this is done, the savings (both money and energy) can be visualized in GIS. An analysis can be done on each district to see the number of LED lights (by selecting the number of point vectors contained in each district polygon), average level of light pollution in each district, and monetary and carbon savings. This analysis could find that one of the 15 Los Angeles council districts is saving more energy while emitting less light pollution. If this is the case, the logical next step would be to present these findings and find out what makes this district different than the others. This analysis could also find that the more energy/money saved is directly related to higher levels of light pollution, in which case a more detailed analysis would be needed to determine appropriate levels of light pollution while taking money and emissions into account. Either way, this project is designed to be the first step of a large-scale investigation on light pollution, especially if it can be repeated for other cities around the nation and world.
Once all results have been found, the final data, maps, and conclusions can be integrated into a report. The types of potential formats for this report include a manuscript, poster, PowerPoint presentation, or a website. The type of format will depend on the target audience of this study, and may depend on the results of the investigation.
Resources: Most of the data required for this project is free and available on the internet. The main vector dataset of Los Angeles streetlights is available from data.gov, the government’s open data portal. The dataset of energy savings is also available from this website. If this data is to be used, it must be merged with a dataset containing the spatial location of Los Angeles council districts. There seem to be a number of shapefiles available for download that contain this information.
The one dataset that will have to be created is the light pollution raster map. Oakland County made their map by analyzing imagery from NASA satellites. An image over the nighttime Los Angeles area will have to be acquired to make this dataset. In order for the map to be consistent with the streetlight data, the image must have been taken in late 2016 or early 2017. The raw satellite data will be converted to a map of light pollution using methods learned from Oakland County.
Time Schedule: In the first two weeks, the people from Oakland County should be contacted about the exact methods used to create the light pollution map from satellite imagery. This time should also be used to download every other dataset that will be needed, and ensure they are placed in a working directory. By the end of this time, the complete methodology used to create such maps should be known.
The next step involves the acquisition of a satellite image over nighttime Los Angeles from NASA, and converting this image into a raster map of light pollution levels using remote sensing techniques and raster mathematics. The exact timeframe to perform this step is unknown, and will only be known after Oakland County has been contacted. A tentative timetable for this step is one month, but this is only an estimate, and should be refined after talking with the people from Oakland County.
After the map has been made, one week will be devoted to determining if a correlation exists between light pollution levels and LED streetlights. If there is sufficient correlation, the next week of the project will be devoted to studying energy and money savings versus light pollution levels. The final week of the investigation will be used to gather the results into a formal report.
References
“Light pollution—what is it and why is it important to know?”. Dark Skies Awareness. http://www.darkskiesawareness.org/faq-what-is-lp.php . Web. April 14, 2017.
“AMA Report Affirms Human Health Impacts from LEDs”. International Dark Sky Association. http://darksky.org/ama-report-affirms-human-health-impacts-from-leds/ . Web. April 14, 2017.
“Streetlight Locations”. City of Los Angeles. https://catalog.data.gov/dataset/streetlight-locations . Web. April 14, 2017.
“Citywide LED Streetlight Savings”. City of Los Angeles. https://catalog.data.gov/dataset/citywide-led-streetlight-savings . Web. April 14, 2017.
“Light Pollution in Oakland County”. Oakland County. https://oakgov.maps.arcgis.com/apps/MapSeries/index.html?appid=b4a5748a18fe49a890b2285fe0abe381 . Web. April 14, 2017.