Exploring the ecological potential of SDGSAT-1 MII and TIS data: Methods, applications, and comparisons

Abstract

As global urbanization accelerates and ecological challenges intensify, effective monitoring and assessing ecological conditions have become critical for sustainable development. Remote sensing technologies play an increasingly crucial role in this context. The Sustainable Development Goals Science Satellite 1 (SDGSAT-1), a next-generation remote sensing satellite, provides 10-m spatial resolution and multispectral imaging capabilities, offering new opportunities for ecological monitoring. This study explores the ecological potential of SDGSAT-1 data, focusing on the comprehensive assessment of urban heat islands (UHI), urban vegetation coverage, and regional ecological conditions. This is achieved through a detailed comparison with the widely-used Landsat-8/9 data. The study develops several methodologies for cloud detection, atmospheric correction, and land dryness retrieval. Validation shows that the cloud removal effect achieved by the proposed SDGSAT Cloud Mask (SCM) algorithm is comparable to, or slightly better than, those of the CFMask algorithm for Landsat-9 and the machine learning-based S2cloudless algorithm for Sentinel-2A, with F1 scores greater than 0.92. The results show that the monitoring of regional ecological conditions by SDGSAT-1 is very similar to that of Landsat-8/9, with differences generally under 5 %. Because SDGSAT-1's multispectral and thermal infrared imagery has higher spatial resolution than Landsat-8/9, it can detect 5.6 % more vegetation area and 2.6 times larger high-temperature areas within urban environments than Landsat data. SDGSAT-1's finer resolution enables more detailed ecological assessments, supporting urban sustainability applications. However, due to the lack of shortwave infrared bands in the SDGSAT-1 imagery, it is less effective than Landsat-8/9 in interpreting land surface dryness and moisture content.