Monitoring ecological trends and conditions receive a growing interest. Satellite imagery is very suitable to monitor the ecological conditions as it shows well the vegetation properties, provides objective information on a regular basis and has a complete coverage. In this project we developed a method to quantify plant phenology by a time series analysis of satellite images using the HANTS algorithm. This algorithm considers only the most significant frequencies expected in the time profiles, and applies a least squares curve fitting procedure based on harmonic components (cosines). For each frequency the amplitude and phase of the cosine function is determined in an iterative procedure. The resulting amplitudes and phases describe the plant phenology.
The next step was to subtract the amplitude and phase values from the two years time series. Thus, among other things, hot spots (areas with extreme deviations) become apparent. With e.g. aerial photographs these hot spots were further investigated. Human interventions, such as forest cutting, reclaiming agricultural lands for nature, or removal of shrubs, explained most of the hot spots.
With this method we can quantify ecological changes within large areas. It has the potential to monitor land surface with its vegetation dynamics in an operational way.
