Blue‒green water demand, water footprint, and productivity: an experimental study for effective agricultural water management in a semi-arid area of southern Ethiopia

Water scarcity constrains agricultural production in semi-arid regions and particularly affects water-sensitive crops such as haricot bean ( Phaseolus vulgaris L.). Although supplemental irrigation is widely promoted in these environments, the partitioning of haricot bean water use into green (rainfall) and blue (irrigation) components has not been quantified for southern Ethiopia, and this partition determines how far the crop draws on contested dry-season water sources. A field experiment was conducted at Ambara, Arba Minch Zuria Woreda, in a randomized complete block design with six stage-specific supplemental irrigation treatments and three replications. Seasonal crop water requirement under non-limiting moisture was 309 mm, of which effective rainfall supplied 221 mm; supplemental irrigation ranged from 0 to 88 mm, so blue water accounted for at most 28.5% of seasonal crop water use. Full supplemental irrigation (T2) gave the highest grain yield, 3095 kg ha −1 or 36.9% above the rainfed control, and the highest blue water footprint (284 m 3 Mg −1 ). Irrigation restricted to the vegetative, flowering and pod formation stages (T3) yielded 2685 kg ha −1 , 18.8% above the control, while using 49.2% less irrigation water and reducing the blue water footprint to 166 m 3 Mg −1 . T3 was not significantly separated from either full irrigation or the rainfed control at the 5% level. T2 and T3 converted supplemental irrigation into grain at nearly the same marginal efficiency, about 0.95 kg m −3 , whereas treatments omitting a reproductive stage returned about 0.6 kg m −3 . Stage-specific supplemental irrigation at the vegetative, flowering and pod formation stages is therefore a promising blue-water-saving strategy for the study area, pending confirmation across seasons and sites.