Stereoscopic hyperspectral remote sensing of the atmospheric environment: Innovation and prospects.

Traditional ground-based air sampling measurements of air quality have blind monitoring areas in the junctions between provinces, cities and urban and rural areas, and they lack the ability of vertical monitoring. Stereoscopic hyperspectral remote sensing techniques provide a promising strategy for improving our understanding of air pollution. Satellite and ground based hyperspectral remote sensing techniques have been demonstrated to have unparalleled technical advantages in monitoring the horizontal and vertical distributions of air pollutants compared to other monitoring techniques. However, to unveil the complex evolutions and processes of the atmospheric environment, the current stereoscopic hyperspectral remote sensing techniques still face several technical bottlenecks, such as a limited temporal resolution in horizontal space, a limited stereoscopic spatial resolution, the limited types of trace gases, the impact of cloud coverage, and the difficulty in nighttime monitoring. The new technical requirements mainly include the following changes: (1) from horizontal and vertical to grid-stereoscopic monitoring; (2) from kilometer to meter resolutions; and (3) from once a day to full-time monitoring with a high temporal resolution. In this article, we systematically review the recent advances in satellite- and ground-based hyperspectral remote sensing techniques, including China's first hyperspectral satellite GF-5, hardware, algorithms, and applications. Moreover, we discuss the broad application prospects of the unmanned aerial vehicle hyperspectral remote sensing monitoring system, the active hyperspectral remote sensing monitoring system, and machine learning in air pollution monitoring in the future. We recommend using the expected multi-means joint hyperspectral stereoscopic remote sensing monitoring mode to assist the effective monitoring and regulation of air pollution in the future. [ABSTRACT FROM AUTHOR]