Optimization of sensor locations for a light pollution monitoring network

•An objective function is needed to evaluate the suitability of the near-optimum location of light pollution sensors.•The representativeness of a set of locations are related not only to the light intensity over a satellite image, but also to the environmental vulnerability of each location.•Measurements in the most vulnerable areas of a city may optimally represent a wide region of the studied territory. Light Pollution is an environmental problem that needs to be retrieved by experimental means. However, to the best of our knowledge, there is no methodology nor a quantitative procedure to determine an optimal light pollution monitoring network. In this work, we propose a methodology for locating sensors in a light pollution monitoring network by formulating an optimization problem. We introduce an objective function that measures the representativeness of a set of locations using the spatial semi-variance over an image, and different levels of monitoring needs according to the environmental vulnerability of each location. To apply the methodology to a region of interest, we consider three inputs: a Nighttime-Light Image NTLI, an Environmental Vulnerability Map, and a constrained number of sensors. The output is a set of coordinates to locate sensors that consider the intensity of luminosity in nighttime images and its environmental impact. A case study shows that the methodology locates sensors in the most vulnerable areas in which measurements may optimally represent a wide region of the studied territory.