Two-Dimensional High-Resolution Mapping and Modeling of Light Pollution in Urban Environments
Location
Poster #7
Start Date
2-5-2025 10:00 AM
Department
Biology
Abstract
Discussions surrounding pollution usually revolve around environmental contamination of non-renewable natural resources. While we are familiar with excess greenhouse gases, heavy metals, or plastics, excess amounts of light are not often thought of as a pollutant. The significant expansion in the quality and variety of lighting technology combined with increased affordability of outdoor lighting has allowed for proliferation in its use, especially in urban areas. Greater amounts of people and wildlife will be impacted by photopollution and its negative physical and mental health effects. Currently, photopollution is primarily determined from low resolution satellite images and research is limited by the lack of a low cost, scalable, high resolution techniques to measure the implications that photopollution has on urban microenvironments. Novel techniques are needed to bridge the gap in our understanding of the role that photopollution plays in microenvironments and microhabitats and improve our knowledge of nocturnal ecology in urban areas like Chicago. Drone images will be captured with a camera from a height of 350 feet above the NEIU campus following the FAA regulations for recreational drone use in Chicago. Light level data collected for each photograph will be converted into a raster plot and transformed to grayscale for analysis. All analyses will be performed using R Statistical Software (v4.1.2; R Core Team 2021). Briefly, two-dimensional models of urban space at night will be created by first compiling the lux levels measured during image acquisition and associating them with the corresponding pixel value of the grayscale raster image. Various regression models will be applied to the data and evaluated using the coefficient of determination. The best model configuration will be chosen and used to predict the remaining unknown pixels in the raster. To provide such a technique we plan to study the impact of photopollution by: Utilizing high resolution photographs from low-altitude drone flights to collect photopollution data in multiple areas of Chicago. Developing high resolution models of photopollution microenvironments in Chicago using R statistical Software.
Faculty Sponsor
Aaron Schirmer
Two-Dimensional High-Resolution Mapping and Modeling of Light Pollution in Urban Environments
Poster #7
Discussions surrounding pollution usually revolve around environmental contamination of non-renewable natural resources. While we are familiar with excess greenhouse gases, heavy metals, or plastics, excess amounts of light are not often thought of as a pollutant. The significant expansion in the quality and variety of lighting technology combined with increased affordability of outdoor lighting has allowed for proliferation in its use, especially in urban areas. Greater amounts of people and wildlife will be impacted by photopollution and its negative physical and mental health effects. Currently, photopollution is primarily determined from low resolution satellite images and research is limited by the lack of a low cost, scalable, high resolution techniques to measure the implications that photopollution has on urban microenvironments. Novel techniques are needed to bridge the gap in our understanding of the role that photopollution plays in microenvironments and microhabitats and improve our knowledge of nocturnal ecology in urban areas like Chicago. Drone images will be captured with a camera from a height of 350 feet above the NEIU campus following the FAA regulations for recreational drone use in Chicago. Light level data collected for each photograph will be converted into a raster plot and transformed to grayscale for analysis. All analyses will be performed using R Statistical Software (v4.1.2; R Core Team 2021). Briefly, two-dimensional models of urban space at night will be created by first compiling the lux levels measured during image acquisition and associating them with the corresponding pixel value of the grayscale raster image. Various regression models will be applied to the data and evaluated using the coefficient of determination. The best model configuration will be chosen and used to predict the remaining unknown pixels in the raster. To provide such a technique we plan to study the impact of photopollution by: Utilizing high resolution photographs from low-altitude drone flights to collect photopollution data in multiple areas of Chicago. Developing high resolution models of photopollution microenvironments in Chicago using R statistical Software.