The Yearly Land Cover Dynamics (YLCD) method: An analysis of global vegetation from NDVI and LST parameters Yves Julien ⁎, José A. Sobrino Global Change Unit, Imaging Processing Laboratory (IPL), Universitat de València, Poligono La Coma s/n - 46980 Paterna, Spain abstract article info Article history: Received 9 May 2008 Received in revised form 4 September 2008 Accepted 29 September 2008 Keywords: NDVI LST Vegetation monitoring NDVI (Normalized Difference Vegetation Index) has been widely used to monitor vegetation changes since the early eighties. On the other hand, little use has been made of land surface temperatures (LST), due to their sensitivity to the orbital drift which affects the NOAA (National Oceanic and Atmospheric Administration) platforms flying AVHRR sensor. This study presents a new method for monitoring vegetation by using NDVI and LST data, based on an orbital drift corrected dataset derived from data provided by the GIMMS (Global Inventory Modeling and Mapping Studies) group. This method, named Yearly Land Cover Dynamics (YLCD), characterizes NDVI and LST behavior on a yearly basis, through the retrieval of 3 parameters obtained by linear regression between NDVI and normalized LST data. These 3 parameters are the angle between regression line and abscissa axis, the extent of the data projected on the regression line, and the regression coefficient. Such parameters characterize respectively the vegetation type, the annual vegetation cycle length and the difference between real vegetation and ideal cases. Worldwide repartition of these three parameters is shown, and a map integrating these 3 parameters is presented. This map differentiates vegetation in function of climatic constraints, and shows that the presented method has good potential for vegetation monitoring, under the condition of a good filtering of the outliers in the data. © 2008 Elsevier Inc. All rights reserved. 1. Introduction Traditionally, vegetation has been monitored by remote sensing through vegetation indices, among which the NDVI (Normalized Difference Vegetation Index) is by far the most widely used. However, NDVI has been showed to be responding primarily to the highly absorbing red reflective band, thus mimicking red reflectance and saturating over forested regions, while being sensitive to canopy background variation in arid and semi-arid areas (Huete et al., 1997). Therefore, additional information is needed to complete NDVI information and palliate these drawbacks. Few attempts have been made by the scientific community to integrate additional information to vegetation monitoring, mainly through the analysis of at sensor brightness temperatures. However, since the remotely sensed data with the longest time extent is derived from the AVHRR (Advanced Very High Resolution Radiometer) sensor aboard NOAA (National Oceanic and Atmospheric Administration) satellites, and since these data are contaminated by orbital drift of the NOAA platforms, observation of vegetation index and temperature relationships have been limited to short time series (Nemani & Running, 1997), for which orbital drift can be neglected. To this date, the relations between NDVI and land surface temperature (LST) have been studied in two different ways. The first one is related to their spatial variation, when the purpose is the determination of land surface parameters such as surface moisture or evapotranspiration; the second one is related to their temporal variation, to characterize vegetation changes. However, some of the studies carried out used LST retrieved only from sensor brightness temperatures (BT). In those cases, LST is replaced by BT in the following paragraphs. Nemani and Running (1989) studied temporal variations of NDVI and BT in Montana, showing that this relation evolved in time. The slope between BT and NDVI was sensitive to changes in canopy resistance, identifying this slope as a useful parameter for evapotranspiration estimation. Ehrlich and Lambin (1996) built a land cover classification of Africa through principal component analysis of BT/NDVI slopes over a year of monthly data. This classification compared well with a previous classification. Schultz and Halpert (1995) studied the correlations between NDVI, BT and precipitation over the globe, evidencing a generally positive correlation, especially in the high and middle latitudes, with some subtropical areas presenting a negative correlation. They also found low correlations between NDVI and BT anomaly. Lambin and Ehrlich (1996) reviewed extensively the drivers between NDVI and BT, and described a general spatial pattern of relationships between NDVI and BT, related to land cover. They concluded that BT/NDVI slope could be used to classify land cover, and monitor land cover changes over time, when associated to seasonality information, retrieved from NDVI Remote Sensing of Environment 113 (2009) 329–334 ⁎ Corresponding author. Tel./fax: +34 96 354 3115. E-mail addresses: yves.julien@uv.es (Y. Julien), sobrino@uv.es (J.A. Sobrino). 0034-4257/$ – see front matter © 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.rse.2008.09.016 Contents lists available at ScienceDirect Remote Sensing of Environment journal homepage: www.elsevier.com/locate/rse