Studying interactions between climate variability and vegetation dynamic using a phenology based approach

In this paper we investigated if and how a signature of climate control on vegetation growth can be individualized at regional scale using time series of SPOT-VEGETATION NDVI and ECMWF meteorological data. Twelve regions characterized by dominant and stable cropland or grassland covers were selected in Europe and Africa. Our results show that the relationship between NDVI and meteorological parameters is highly complex and significantly vary trough the phenological cycle of the plants. Hence, interactions between vegetation dynamics and climate variability must be studied at a smaller time scale in order to identify properly the limiting factors to vegetation growth. Using NDVI metrics, vegetative phases (from green-up to maximum NDVI) and reproductive phases (from maximum NDVI to maturity) were identified for each region. Cross-correlation analysis revealed that, in most of the cases, the best scores of Pearson's r are obtained when we considered the vegetative phase (from green-up to maximum of NDVI) and the reproductive phase (from maximum of NDVI to maturity) separately. We also showed that climatic constraints identified using yearly proxies of climate and vegetation do not depict correctly or completely the climate control on vegetation development. In that sense the complexity of the climate-vegetation relationship, which is spatially and temporally variable, is well underlined in this study.

► We developed a method to apprehend more thoroughly the climate control on vegetation growth.
► Various time lags (climate vs. NDVI) and accumulation periods (climate) were considered.
► Interactions between climate and NDVI vary depending on the region and the vegetation type.
► ‘Climate – NDVI’ correlations improved significantly when considering phenological phases separately.
► As climate constrains on vegetation are not constant through the year, they are inaccurate when derived from yearly metrics.