The analyses of trends in vegetation dynamics require a profound knowledge of its seasonality. For the determination of the seasonality conventional methods of time series analyses often use a simple averaging of measured values of the identical time in different cycle of the whole time series (e.g. bfast). Then it is assumed that the resulting seasonal portion of a time series is constant and stable for the entire time series. However, analyses of vegetation time series show that trends in vegetation dynamics do not always run steadily, but show structural breaks, especially in regions with high potential for possible landscape changes. For such conversion areas, the assumption of a constant seasonality is not always ensured. The dynamic or variability of the seasonality can have temporal effects by a shift of the start of the season (SOS) or the end of the season (EOS) and therefore also on the length of the vegetation period. To show whether seasonal dynamics can be detected in vegetation time series, two requirements must be fulfilled. (1) High-temporal resolution vegetation information provided for example as MODIS-NDVI. (2) Indicators are needed which allows the description of the variability of seasonality. As a result these metrics allow a better modeling of long-term vegetation dynamics in the trend, taking into account the variability of the seasonality. But at the same time the metrics itself serve as indicators for long term vegetation dynamics. The aim of the present study is to analyse phenological and greenness metrics for the modelling of vegetation dynamics in the nature reserve Konigsbrucker Heide. Detailed analyses of key metrics like SOS and EOS using different metric approaches and interpolation methods are applied and compared. The results show that it is difficult to determine consistent information for example for the trend of single phenology metrics.