Modeling the potential for closing quinoa yield gaps under varying water availability in the Bolivian Altiplano

In the Bolivian Altiplano, the yields of rainfed quinoa are relatively low and highly unstable. We use a validated crop water productivity model to examine the potential of closing quinoa yield gaps in this region. We simulate the expectable yields under rainfed cultivation and under different deficit irrigation (DI) strategies using the AquaCrop model for the Northern, Central and Southern Bolivian Altiplano. Simulated DI scenarios include a reference strategy avoiding stomatal closure during all sensitive growth stages and allowing drought stress during the tolerant growth stages (DI₀) and various restrictive deficit irrigation strategies (DI_i) representing cases when water resources are limited. We obtain a logistic crop water production function for quinoa by plotting the seasonal actual evapotranspiration versus total grain yield. Due to the large scatter, this function only indicatively provides expectable yields. From the scenario analysis, we derive yield probability curves for the 3 agro-climatic regions. DI, without restriction in irrigation water during the drought sensitive growth stages, is able to close the yield gaps in the Northern, Central and Southern Bolivian Altiplano, and would guarantee a high and stable level of water productivity (WP). The yields of quinoa under rainfed cultivation during dry years are only 1.1, 0.5 and 0.2 Mg ha^-1 in the Northern, Central and Southern Bolivian Altiplano, whereas under DI₀ they are 2.2, 1.6 and 1.5 Mg ha^-1, respectively. Under limited water availability for irrigation, these stable yield levels decrease, most drastically in the Southern Bolivian Altiplano. Below a minimum water availability of 600 m³ per ha and 700 m³ per ha in the Central and Southern Bolivian Altiplano, respectively, the application of DI for quinoa is not significantly effective and should be avoided to save valuable resources. The yield probability curves we derive can serve as input for stochastic economic analysis of DI of quinoa in the Bolivian Altiplano.