Technological developments from last decades offer unprecedented opportunities to monitor the Earth system. International research projects like BACI are joint efforts to provide free-of-charge, unified and high quality Earth Observations and the development of tools to analyze them. The ability to detect and monitor anomalous behaviour in multivariate environmental time series is crucial. These events are signals of changes in the underlying dynamical system and their detection can be used as an early-warning system for land ecosystems. In this study we present a methodology to detect these anomalies in biosphere data by a combination of a multivariate autoregressive model together with a distance measure. This work is framed within the EU-funded project BACI 'Detecting changes in essential ecosystem and biodiversity properties - towards a Biosphere Atmosphere Change Index'.