Premium
Ecological consideration for several methodologies to diagnose vegetation phenology
Author(s) -
Lim Chi Hong,
An Ji Hong,
Jung Song Hie,
Nam Gyung Bae,
Cho Yong Chan,
Kim Nam Shin,
Lee Chang Seok
Publication year - 2018
Publication title -
ecological research
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.628
H-Index - 68
eISSN - 1440-1703
pISSN - 0912-3814
DOI - 10.1007/s11284-017-1551-3
Subject(s) - phenology , vegetation (pathology) , enhanced vegetation index , normalized difference vegetation index , digital camera , scale (ratio) , image resolution , remote sensing , physical geography , growing season , digital imaging , environmental science , digital elevation model , digital image , ecology , cartography , climate change , geography , image processing , vegetation index , computer science , artificial intelligence , biology , image (mathematics) , medicine , pathology
A challenge of phenological research is to integrate remotely sensed observations obtained at different spatial and temporal scales to provide information that contains both a high temporal density and fine spatial scale observations. This study aims to improve the level of spatial detail and the temporal density required for efficient monitoring of vegetation phenology by applying two remote sensing techniques, MODIS and digital camera images. Based on the vegetation indices extracted from each measurement, we analyzed phenological changes of vegetation and deduced phenophases transition dates, such as start dates of green‐up and senescence by applying different two methods, the rate of change of curvature, K and HMV. The start and end of the growing season of Mongolian oak expressed in ExG–DC, which were extracted from digital camera image, were agreed well with that of visual assessment. EVI among three vegetation indices (EVI, NDVI and ExG–MI) showed the high correlation with ExG–DC and visual assessment. Based on RMSE, the transition dates assessed by visual observation were agreed better with the dates, which were estimated based on curvature K than with the dates estimated from HMV in all vegetation indices. Sap flow time‐series estimates for the phenological transition dates were closely accorded with the estimates derived from near‐surface time‐series, and coincided better with the dates estimated based on curvature K than that based on HMV. In conclusion, based on the result of this study, we suggest that it is effective to use EXG–DC obtained from digital camera and EVI from MODIS when these two instrument are integrated as the vegetation indices and that curvature K is an effective method for extracting the phenological event dates of vegetation.