The common agricultural policy (CAP) is one of the oldest policies of the European Union, however, the latest reform will reshape the agriculture land use control processes from the selected risk fields-based into an all-inclusive one, fostering the use of Sentinel-2 data. Detection of land use violations from the declared annual land use (e.g. overgrowth or tillage of permanent meadows, long-lasting quiescence of the crop fields) will turn the land use control process into massive, thus effective and customised monitoring capabilities are needed. Detecting change within the trend and seasonal components of time series can enable the detection of these violations. In this paper, we investigate the suitability of the Breaks For Additive Season and Trend Monitor (BFAST Monitor) method in combination with the analysis and customisation of segment-based temporal (NDVI) profiles, to automatically detect and verify changes on the segments that correspond to the same agricultural landscape units (permanent meadows and crop fields) using Sentinel-2 images. We have obtained and atmospherically corrected all available data over the study area (north-eastern part of Slovenia). We used cloud-free Sentinel-2 satellite images from which we derived biophysical indicators (i.e. NDVI). The history period is defined from first available images of Sentinel-2A (end of June) to end of 2016, and the monitoring period from the beginning of 2017 onwards, including Sentinel-2B images. The results obtained with BFAST method are compared to segment-based temporal NDVI profiles (customised as referential phenological profiles), and detected changes on the agricultural segments are automatically confirmed or rejected. The validation of the obtained results (agriculture land use anomalies) was performed with the information on declared land annual use and field controls, obtained in the framework of subsidy granting. Despite very short Sentinel-2 data time series, results indicate that the combination of both time series methods is a very useful approach for automatic detection of land use anomalies on agricultural areas that undergo similar phenological behaviour. As such the approach can be introduced to help the process of agriculture land use control within CAP activities.