[A-CG36_29PM2] Science in the Arctic Region

[ACG36-12] Multidisciplinary in situ and satellite observations for accurate detection of phenology in sub- and Arctic ecosystems

*Shin NAGAI¹, Shunsuke TEI², Hideki KOBAYASHI¹, Hiroki IKAWA³, Taro NAKAI⁴, Yongwon KIM³, Rikie SUZUKI¹, Atsuko SUGIMOTO² (1.Research Institute for Global Change, Japan Agency for Marine-Earth Science and Technology, 2.Graduate School of Env.Science / Faculty of Env.Earth Science Hokkaido University, 3.International Arctic Research Center, University of Alaska Fairbanks, 4.Hydrospheric Atmospheric Research Center, Nagoya University)

Keywords:phenology, remote sensing, terrestrial ecosystem, Siberia, Alaska, ground-truthing

To accurately evaluate the spatio-temporal variability of ecosystem functions and service in sub- and Arctic regions under rapid meteorological and climate changes, global, long-term, and comprehensive phenological observations are required. Towards this aim, satellite remote-sensing is useful to detect the spatio-temporal variability of plant phenology such as the timing of start (SGS) and end of growing season (EGS). However, from the in situ ecological research viewpoint, the satellite remote-sensing has not been sufficiently tested and validated by ground-truthing. Here, (1) we performed daily field observations with time-lapse digital cameras in boreal forests in Alaska and Siberia; (2) we examined the relationship between satellite-observed vegetation indices and plant phenology; and (3) we evaluated the spatio-temporal variability of the timing of SGS and EGS in sub- and Arctic regions by using MODIS Terra and Aqua-observed green-red vegetation index (GRVI). We found that (1) satellite-observed vegetation indices (i.e. NDVI, EVI, and GRVI) mainly detected the plant phenology of forest floor in sparse forests; (2) large year-to-year variability of the timing of SGS was detected in eastern Siberia and western Ural Mountains, while that of EGS was not clearly detected; and (3) in contrast, large year-to-year variability of the timing of EGS was detected in western Alaska, which is mainly covered by tundra vegetation, while that of SGS was not clearly detected.