Assessing FY-3D MERSI-Ⅱ Observations for Vegetation Dynamics Monitoring: A Performance Test of Land Surface Reflectance

• Published in: IEEE transactions on geoscience and remote sensing Vol. 62; p. 1
• Main Authors: Yang, Kai, Yan, Kai, Zhang, Xinjian, Zhong, Run, Chi, Haojing, Liu, Jinxiu, Ma, Xuanlong, Wang, Yuanyuan
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.01.2024; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Algorithms; Dynamics; EVI2; FY-3D MERSI-II; GVF; land surface reflectance; Meteorological satellites; MODIS; Monitoring; Performance testing; Performance tests; Reflectance; Reflectivity; Satellite broadcasting; Sensors; Spatial discrimination; Spatial resolution; Spectroradiometers; Temporal resolution; Vegetation; vegetation dynamics; Vegetation index; Vegetation mapping
• ISSN: 0196-2892; 1558-0644
• DOI: 10.1109/TGRS.2023.3348997

Medium-resolution satellites have been instrumental in monitoring global vegetation dynamics over the past decades. The Fengyun (FY) 3D satellite, a second-generation medium-resolution polar-orbiting meteorological satellite launched by the China National Meteorological Administration in 2017, plays a pivotal role in the low-orbiting group network for meteorological, oceanic, and land surface observations. MERSI-II, a key component of FY-3D designed with inspiration from MODIS, holds significant yet untapped potential for analyzing vegetation dynamics. This study embarks on a systematic analysis of FY-3D MERSI-II's applicability in vegetation research, comparing it with Aqua MODIS. First, the spectrums of MERSI-II and MODIS are very close to each other, and compared to MODIS, MERSI-II is slightly overestimated in Red and slightly underestimated in NIR bands; both reflectance products maintain a good temporal stability when examined through desert sites, with the data being more fluctuating when the observation angle is larger; and the data availability for MERSI-II is slightly lower than that of MODIS, due to its more stringent cloud detection algorithm. Finally, the results of EVI2 (enhanced vegetation index with two bands) and the vegetation parameter GVF (green vegetation fraction) show that MERSI-II is also capable of monitoring vegetation dynamics with an optimal temporal resolution of 12 days and a spatial resolution of 2 km. Our comprehensive assessment confirms the remarkable capability of FY-3D MERSI-II in dynamic vegetation monitoring and underscores the need to make the most of its valuable observations. Our findings support the advancement of vegetation monitoring techniques and aid in adjusting the optimal spatial and temporal resolution of related products.