Identification of Years with Extreme Vegetation State in Central Europe Based on Remote Sensing and Meteorological Data

Abstract

Background and Purpose: Determination of an extreme year from the aspect of the vegetation activity using only meteorological data might be ambiguous and not adequate. Furthermore, in some ecosystems, e.g. forests, the response is not instantly visible, but the effects of the meteorological anomaly can be seen in the following year. The aim of the present paper is to select and characterize typical and anomalous years using satellite-based remote sensing data and meteorological observations during the recent years of 2000-2014 for Central Europe, based on the response of the vegetation. Materials and Methods: In the present study vegetation characteristics were described using remotely sensed official products of the MODerate resolution Imaging Spectroradiometer (MODIS), namely NDVI, EVI, FPAR, LAI, GPP, and NPP, with 8-day temporal and 500 meter spatial resolution for the period of 2000-2014. The corresponding mean temperature and precipitation data (on the same grid) were derived from the Open Database for Climate Change Related Impact Studies in Central Europe (FORESEE) daily meteorological dataset. Land cover specific anomalies of the meteorological and vegetation characteristics were created and averaged on a country-scale, where the distinction between the main land cover types was based on the synergetic use of MODIS land cover and Coordination of Information on the Environment (CORINE) Land Cover 2012 datasets. Results: It has been demonstrated that the anomaly detection based solely on basic meteorological variables is ambiguous since the strength of the anomaly depends on the selected integration time period. In contrast, the effect-based approach exploiting the available, state-of-the-art remote sensing based vegetation indices is a promising tool for the characterization of the anomalous behaviour of the different land cover types. The selection of extreme years was performed in an explicit way using percentile analysis on pixel level. Conclusions: Plant status in terms of both positive and negative anomalies shows strong land cover dependency in Central Europe. This is most likely due to the differences in heat and drought resistance of the vegetation, and species composition. The selection of country-specific extreme years can serve as a basis for forthcoming research.