Monitoring vegetation dynamics by coupling linear trend analysis with change vector analysis: a case study in the Xilingol steppe in northern China

Abstract

Timely and accurate monitoring of grassland vegetation dynamics is essential for sustainable grassland management in China. We coupled linear trend analysis (LTA) with change vector analysis (CVA) to improve the effectiveness of grassland monitoring. LTA was used to detect continuous inter-annual vegetation trends to identify significant change trend regions (SCTRs) in location and significant change trend periods (SCTPs) in time. Then CVA was used to depict intra-annual change intensities in SCTRs for a SCTP. The Xilingol steppe in northern China was selected to evaluate the method's performance. Digital images of degraded grasslands derived by the proposed method using data from the VEGETATION instrument on board the Systme Probatoire d’Observation de la Terre (SPOT/VGT) were compared to those derived from Landsat images of the same area. Linear regression analysis comparing degraded grassland areas from the two imagery sources showed good correspondence. An overall accuracy of 85.33% and a kappa coefficient of 0.66 were obtained through error matrix analysis. The results showed a general grassland degradation trend from 1998 to 2007. The SCTRs were mostly distributed in the north, and the grassland degradation trend in SCTRs was more significant from 1998 to 2001 and from 2003 to 2007 during the study period. About 19% of the vegetated area was composed of degraded steppe grassland for the two time periods.

Acknowledgement

The research was supported by the National Basic Research Programme of China (grant no. 2010CB950901) and the State Key Laboratory of Earth Surface Processes and Resource Ecology, Beijing Normal University, China (grant no. 2010-KF-02).

Notes