AbstractArid and semiarid environments are susceptible to environmental degradation and desertification. Modelling net primary productivity (NPP) and analysis of spatio-temporal patterns help to understand ecological functioning especially in these areas. In this study, we apply the Biosphere Energy Transfer Hydrology Model (BETHY/DLR) to derive NPP for Kazakhstan for 2003–2011. Results are analyzed regarding spatial, monthly, and inter-annual variations. Mean annual NPP for Kazakhstan is 143 g C m−2 and maximum productivity is reached in June. Most monthly NPP anomalies occur in semiarid North of Kazakhstan. These regions seem to be most strongly affected by changes in meteorology and are likely to be vulnerable to changing climate. Arid ecosystems show lower inter-annual NPP variability than semiarid lands. Correlations between NPP and meteorological parameters reveal variable influence of temperature, PAR, and precipitation on vegetation productivity during the year. Reaction of vegetation growth to precipitation is delayed 1–2 months. Temperature is most critical in spring and precipitation in summer affects NPP in August–October. The results presented in this study help to identify regions that are vulnerable to global change. They allow predictions on possible effects of expected future climate change on vegetation productivity in arid and semiarid Kazakhstan and support sustainable land management.