The paper proposes a technology to detect anomalies of net primary production of vegetation dynamics and the authors have done an analysis of the data. The authors are based on ecological theory for conceptual technology that allows you to monitor an extended territory, using estimates NPP (MODIS NPP) as an indicator of the conditions of vegetation. For the detection of areas with anomalies NPP dynamics was used STL method (Seasonal-Trend Decomposition Procedure Based on Loess). The study was conducted in the south of the Krasnoyarsk region, which has a great variety of natural landscapes from the steppes to the high alpine meadows of the SayanMountains. For analysis of the dynamics of anomalies of NPP used satellite data in temperature and humidity.