Open AccessMonotonic Vegetation Trend Detection Based on Ensemble Empirical Mode Decomposition in Ningxia, China
Author(s) -
Liang Tang,
Haijun Yang
Publication year - 2020
Publication title -
iop conference series. earth and environmental science
Language(s) - English
Resource type - Journals
eISSN - 1755-1307
pISSN - 1755-1315
DOI - 10.1088/1755-1315/502/1/012042
Subject(s) - desertification , normalized difference vegetation index , vegetation (pathology) , urbanization , environmental science , trend analysis , china , loess plateau , physical geography , threatened species , climate change , hilbert–huang transform , geography , soil science , mathematics , statistics , geology , ecology , medicine , oceanography , white noise , archaeology , pathology , habitat , biology
China and its local governments have made a variety of efforts in vegetation protection and restoration to ensure sustainability especially in the area threatened by severe desertification, such as Ningxia. Studying the vegetation trend and the underlying driving factors is integral to the assessment of the effect of those efforts. This paper applied Ensemble Empirical Mode Decomposition (EEMD) in combination with Mann-Kendall (M-K) significance test to detect the monotonic trend of vegetation in Ningxia based on the 16-day MODIS NDVI time series spanning the period from 2006 to 2015. The result showed that the vegetation in Ningxia exhibited an upward trend in general. The area with upward and downward trend accounted for 80.94% and 14.87%, respectively, and the area without significant change accounted for 4.46%. A rapid increase was found in the central part of the Loess Plateau, whereas a non-increase occurred in the urban expansion area. The possible major driving factors for the detected trend include the implementation of the “Grain for Green” program and urbanization.