Vegetation phenology in global change studies

Abstract

Changes in the character of the vegetated land surface are frequently expressed in terms of temporal trends in the Normalized Difference Vegetation Index (NDVI) retrieved from spaceborne sensors. In the past these change studies were typically based upon AVHRR data. By the end of 2011, we acquired 11 full years of NASA MODIS data which is a greatly improved dataset compared to extant AVHRR datasets. In this chapter, we present a change analysis based on a global NASA MODIS product (MCD43C4) at a 0.05° (~5.6 km) spatial resolution and a 16-day temporal resolution from 2001 through 2011. This new change map based on 11 years of data presents statistically significant positive and negative changes resulting from both direct and indirect impacts of climatic variability and change, disturbances, and human activity. We found significant negative changes in 8.7 % of the global land area (or 11.8 × 10 6 km 2), with hotspots in Canada, southeastern USA, Kazakhstan, and Argentina. Significant positive changes appeared in 6.0 % of the global land area (8.0 × 10 6 km 2) with hotspots in Turkey, China and Western Africa. Attribution is the key challenge in any change analysis. We provide several examples attributable to major modes of change, focusing both on natural disturbances arising from climatic variability and change, and also on changes arising directly from human actions.