Vegetation Growth Abnormality by NDVI’s Response to Precipitation in Mongolia and its Grassland, 2001 to 2020

YAN Guanyu*, TAKEUCHI Wataru

Abstract

The sustainability of the natural environment is vital to Mongolia due to pastoralism. It is easy to observe the phenology of vegetation growth represented by NDVI but challenging to tell vegetation changes due to human interventions from rainfall fluctuations. This can be solved through NDVI's response to precipitation changes. We calculated the temporal Pearson Correlation Value between NDVI and Precipitation data in Mongolia from 2001 to 2020, emphasizing grassland changes with MODIS LULC. In this period, the mean and median values of NDVI-Precipitation Correlation Values are 0.327 and 0.331, with a standard deviation of 0.204. Grassland has grown
4.69%, and about 51% of grassland is deemed abnormal in vegetation growth by the NDVI-Precipitation correlation. There are substantial portions of grassland in Mongolia under heavy stress, although the overall NDVI of grassland is increasing.
Model verification shows NDVI-Precipitation Correlation Value = 0.330-0.008 × Population Density, which links low correlation value to higher population density.

Keywords: NDVI, GSMaP, Pearson Correlation, Time series data, Arid lands, Biomass

Full Text: