Sensitivity of near real-time MODIS gross primary productivity in terrestrial forest based on eddy covariance measurements

• Published in: Chinese geographical science Vol. 25; no. 5; pp. 537 - 548
• Main Authors: Tang, Xuguang, Li, Hengpeng, Liu, Guihua, Li, Xinyan, Yao, Li, Xie, Jing, Chang, Shouzhi
• Format: Journal Article
• Language: English
• Published: Heidelberg Science Press 01.10.2015; Springer Nature B.V
• Subjects: Carbon; Carbon cycle; Climate change; Deciduous forests; Earth and Environmental Science; Environmental management; Forest ecosystems; Forest productivity; Forests; Geography; Land use; Mixed forests; Primary production; Terrestrial ecosystems; Tropical forests; evaluation; forest ecosystem; eddy covariance (EC); MOD17A2; FLUXNET community; gross primary productivity (GPP)
• ISSN: 1002-0063; 1993-064X
• DOI: 10.1007/s11769-015-0777-7

As an important product of Moderate Resolution Imaging Spectroradiometer (MODIS), MOD17A2 provides dramatic improvements in our ability to accurately and continuously monitor global terrestrial primary production, which is also significant in effort to advance scientific research and eco-environmental management. Over the past decades, forests have moderated climate change by sequestrating about one-quarter of the carbon emitted by human activities through fossil fuels burning and land use/land cover change. Thus, the carbon uptake by forests reduces the rate at which carbon accumulates in the atmosphere. However, the sensitivity of near real-time MODIS gross primary productivity (GPP) product is directly constrained by uncertainties in the modeling process, especially in complicated forest ecosystems. Although there have been plenty of studies to verify MODIS GPP with ground-based measurements using the eddy covariance (EC) technique, few have comprehensively validated the performance of MODIS estimates (Collection 5) across diverse forest types. Therefore, the present study examined the degree of correspondence between MODIS-derived GPP and EC-measured GPP at seasonal and interannual time scales for the main forest ecosystems, including evergreen broadleaf forest (EBF), evergreen needleleaf forest (ENF), deciduous broadleaf forest (DBF), and mixed forest (MF) relying on 16 flux towers with a total of 68 site-year datasets. Overall, site-specific evaluation of multi-year mean annual GPP estimates indicates that the current MOD17A2 product works highly effectively for MF and DBF, moderately effectively for ENF, and ineffectively for EBF. Except for tropical forest, MODIS estimates could capture the broad trends of GPP at 8-day time scale for all other sites surveyed. On the annual time scale, the best performance was observed in MF, followed by ENF, DBF, and EBF. Trend analyses also revealed the poor performance of MODIS GPP product in EBF and DBF. Thus, improvements in the sensitivity of MOD17A2 to forest productivity require continued efforts.