Win-win: Improved irrigation management saves water and increases yield for robusta coffee farms in Vietnam

2020 
Abstract Robusta coffee is critically important for the economy and farmers of Vietnam, but also requires substantial irrigation leading to dwindling water resources. Developing clear recommendations for improved irrigation water management, while maintaining or increasing yield is therefore a key knowledge need for the coffee industry. We analyse 10-cropping-year data (2008/2009–2017/2018) of 558 farms across four major coffee-producing provinces in Vietnam’s Central Highlands using CROPWAT and hierarchical Bayesian modelling to (1) identify irrigation requirements under different climatic conditions, and (2) investigate the potential for improved irrigation management strategies. In average rainfall years the majority of farmers in Dak Nong and Lam Dong supplied an equivalent of 455–909 L tree−1 (assuming 1100 plants ha−1) with corresponding average yields ranging from 2149 to 3177 kg ha−1. In Dak Lak and Gia Lai the predominant range was equivalent to 1364–1818 L tree-1 (corresponding average yields: 2190 to 3203 kg ha−1). In dry years more water was supplied through irrigation at various levels depending on the province: varying between 1364–1818 L tree−1 in Dak Lak and Gia Lai, and 909–1364 L tree−1 in Dak Nong and Lam Dong. Our study also shows that irrigation water can be reduced by 273–536 L tree−1 (300–590 m3 ha−1) annually from the current levels in average rainfall years while still achieving average yield levels greater than 3000 kg ha−1. In dry years reductions of 27–218 L tree−1 (30–240 m3 ha−1) are possible. With adequate management of the key crop practices affecting coffee yields, substantial water savings at the provincial scale could be achieved. Thus, our findings could serve as a basis for province-specific irrigation water management in robusta coffee farms that will not only reduce overall water use, but also potentially maintain satisfactory yield levels.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    25
    References
    6
    Citations
    NaN
    KQI
    []