The desaturation method of DMSP/OLS nighttime light data based on vector data: taking the rapidly urbanized China as an example. Issue 3 (4th March 2019)

• Record Type: Journal Article
• Authors: Zheng, Zihao; Chen, Yingbiao; Wu, Zhifeng; Ye, Xinyue; Guo, Guanhua; Qian, Qinglan
• Is Part Of: International journal of geographical information science. Volume 33:Issue 3(2019)
• Page Start: 431
• Page End: 453
• Publication Date: 2019-03-04
• Subjects: VDANTLI -- urbanization -- DMSP/OLS -- saturation -- China; Geography -- Data processing -- Periodicals; Information storage and retrieval systems -- Periodicals; Géomatique -- Périodiques; Systèmes d'information -- Périodiques; 910.285
• DOI: 10.1080/13658816.2018.1538516
• Languages: English
• ISSNs: 1365-8816

Abstract

ABSTRACT: The saturation of night light data caused by sensor defects conceals the differences and details of light luminance in the urban area, which greatly limits the application of Nighttime Light (NTL) in the study of urbanization. Although some methods have been proposed to mitigate the saturation of NTL data, the research of desaturation is worth further advancing due to its shortcomings in highlighting light differences and spatial resolution. Therefore, we propose a new spectral index, the Vector Data Adjusted NTL Index (VDANTLI) after analyzing the influence of different auxiliary parameters on saturation elimination. Then, we select the three most developed urban agglomerations in China and make a series of qualitative and quantitative analysis. The results of various assessments confirm that VDANTLI can effectively alleviate NTL saturation and enhance urban lighting differences. Moreover, by dynamically adjusting the length of the grid and single-phase vector data modified time-series NTL data, we further achieve the resolution-series and time-series VDANTLI. Therefore, compared with previous desaturation models, the VDANTLI can be used as an effective indicator and even an indispensable basic data for urbanization research and socio-economic spatial analysis because of its unique advantages in spatial resolution and the capability in time-series native NTL desaturation.