Time senes of vegetation index data provide useful information of vegetation activity. These curves can be used for retrieving key phenological stages such as the day of season onset, peak, and end. These parameters allow for better understanding the landscape and, if conducted for several years, its natural variability or human-induced change. This study employs eight years of MODIS vegetation index time series to retrieve above-mentioned phenological parameters. It was found that the EVI shows earlier phenological dates than the NDVI, possibly linked to its susceptibility to subtle changes. A spatial analysis for ecosystems indicated the shifted phasing of the Mediterranean California region with winter ram and suggested a general North-South gradient with increasing growing season length. Only small differences were found for different curve-fitting algorithms if the same parameters were used. However, user-defined parameter changes have a high impact on the retrieval of phenological dates.