Centre for Development and Environment (CDE)

Forschung

Detecting land degradation and regeneration in Mongolia

land degradation mongolia
Illustration 1: Significant linear regression slope values for trends derived from MODIS 16-day composite NDVI observations from 2001 to 2011. Blue areas indicate positive NDVI trends indicating active vegetation increase. Red areas indicate negative NDVI trends indicating active vegetation decrease. Graphic: Sandra Eckert, CDE

Ecosystems in Mongolia are particularly fragile due to the country's relatively high altitude and its continental climate. Degradation is generally severe, causing not only a decline in herbage yields but also an overall deterioration of the ecological environment. The Mongolian government is well aware of these environmental problems. However, the country's vast size poses a major challenge. Any effort to assess and monitor the complex processes of land degradation – and its severity, extent, and spatial distribution – must make use of remotely sensed data in addition to field data.

land degradation
Illustration 2: Two selected validation points confirmed by two Landsat TM 5 scenes acquired in 2000 and 2011. a) Expansion of mining activities at the Zaamar gold mine (ID 15). b) Area deforested due to a forest fire (ID 1). Both examples are overlaid with the negative trend layer, which clearly indicates the resulting land cover change. Graphic: Sandra Eckert, CDE

MODIS NDVI time series analysis

This study is based on an analysis of Moderate Resolution Imaging Spectroradiometer (MODIS) data recorded by two sensors on board the NASA's Terra and Aqua platforms. It explored 11 years of MODIS 16-day composite Normalized Difference Vegetation Index (NDVI) time series to detect land cover change – in particular land degradation and regeneration areas in Mongolia. Regression analysis was used to generate a trend map that takes account of each time series vector's significance. In order to gain a better understanding of the derived trend areas, we also examined land cover development from 2001 to 2009, as well as trends in meteorological data from 2000 to 2009. High resolution satellite data and field photographs were used to validate small, local NDVI trends.