Global-scale patterns and determinants of cropping frequency in irrigation dam command areas

Abstract

Abstract A growing human population and shifting consumption patterns increase the pressure on agricultural production systems. Reservoir-based irrigation has boosted agricultural production through higher cropping frequencies, whereas the magnitude of this effect varies significantly across the globe. Technological, biophysical and socio-economic constraints often limit cropping frequency in the command areas of irrigation dams, yet the relationships with these factors remain poorly understood at the global scale. Here, we first determined the size and location of 1288 command areas of irrigation dams commissioned since 1985. Within these areas, we studied cropping frequency during the period of 2001–2012 using a global time series of land cover information. We further investigated potential biophysical, socio-economic and technological constraints for intensive cropping using a Boosted Regression Trees modeling framework. Our results showed that the largest extent of reservoir-irrigated croplands are located in India (5.9 Mha), Indonesia (1.5 Mha), China (1.4 Mha), Vietnam (0.9 Mha), Turkey (0.7 Mha), Iran (0.7 Mha), and Thailand (0.6 Mha). Globally, cropping frequencies in irrigation dam command areas were on average 16% higher compared to rainfed control areas, yet pronounced differences in the strength and direction of this effect were apparent across world regions. Technological properties of dams and irrigation systems were amongst the most important variables for explaining global-scale variation in cropping frequency. Specifically, we observed low cropping frequencies in smaller command areas ( 20 km). The command areas of small reservoirs (storage capacity