|
Satellite nocturnal images of the Earth are a useful way to identify urbanization. Nighttime lights have been used in a range of scientific contributions, including studies on building human development indices and on the identification of megalopolises and impacted landscapes. However, the study of the area and the internal structure of urban systems by nighttime light imagery has had a fundamental limitation to date: the low spatial resolution of satellite sensors. DMSP Operational Linescan System (OLS), with a 2.7 km/pixel footprint, has been gathering global low-light imaging data for over 40 years. The 2011 launch by NASA and the NOAA of the Suomi National Polar Partnership (SNPP) satellite, with the Visible Infrared Imaging Radiometer Suite (VIIRS) sensor on board, has led to a significant improvement. This instrument has a better spatial resolution (742 m/pixel), onboard calibration, a greater radiometric range, and fewer saturation and blooming problems than DMSP-OLS data. The launch of Luojia 1-01 in June 2018 has increased expectations. Its high-resolution nocturnal images (130 meters/pixel) allow a better in-depth study of the landscape impacted by urbanization. The objective of this work is to analyze the ability of different nighttime light sensors to delineate urbanized and built-up areas, as well as their effectiveness in typifying and classifying different types of urban developments. The case studies are the three major Chinese megacities: the Guangdong Bay Area (GBA), the Shanghai agglomeration and the Beijing- Tianjin-Hebei (BTH) metropolitan region. For the delimitation of the urbanized and built-up areas that make up the three mentioned megalopolises, the usefulness of Zipf's Law is evaluated. The research question proposed in this paper is whether urban development occurs spontaneously following Zipf's Law, or more generally Pareto's distribution. Initial results confirm the usefulness of applying Zipf's law to DMPS-OLS sensor images to identify the urbanized areas of major Chinese megacities. However, sensors with a higher spatial and radiometric resolution, such as NPP-VIIRS or Luojia 1-01, show greater complexity. The application of Zipf's Law to the images obtained by these sensors allows the identification of the densest and most compact built-up areas. However, they do not show a clear opportunity to identify and delimit the urbanized areas that compose the metropolis.
|
