Usability of noise-free daily satellite-observed green–red vegetation index values for monitoring ecosystem changes in Borneo

Abstract

We examined the usability of daily green–red vegetation index (GRVI) observations from the Terra and Aqua Moderate Resolution Imaging Spectroradiometer satellite on cloud-free days for monitoring ecosystem changes in Bornean tropical forests at a 500 m spatial resolution over 11 years (2003–2013). The number of observational days of cloud-free GRVI data in the southwest monsoon period (May–October; 1–5 days/month) was greater than that in the northeast monsoon period (November–April; 0–2 days/month). Spatial variation in the observation frequency was noticed, with Terra (morning) and Aqua (afternoon) data showing different geographic distribution patterns of cloud-free data. The observation frequency in the western Kalimantan mountains (Sarawak and Sabah) was 1–2 days/month greater than that in the eastern mountains (Kalimantan). The quality of cloud-free GRVI data was validated by using sky images taken at the same time as the satellite observations and canopy surface images in a tropical rainforest. In oil palm and acacia plantations and peatlands, which were mainly distributed in coastal regions of Sarawak and West and Central Kalimantan, the cloud-free daily GRVI value fell below zero owing to deforestation and forest degradation caused by forest fire and increased with replanting and vegetation recovery. These results indicate that daily cloud-free GRVI data from multiple satellites collected at different times of the day are required for accurate monitoring of intra- and interannual phenological variation and forest degradation attributed to changes in climatic conditions and deforestation caused by human activities in tropical ecosystems.

Acknowledgements

We thank T. Itioka and I. Asano (Kyoto University), K. Matsumoto (Ryukyu University), A. Yoneyama (Kochi University), K. Kawahara (Nagoya University), and all members of the Lambir Hills National Park site community for their assistance in the field. We thank the editor and the two anonymous reviewers for their kind and constructive comments. We thank the NASA Land Processes Distributed Active Archive Center in the Earth Observing System Data Gateway for providing the MODIS data.