Bruce Lankford, Dorice Agol, Colin Steley, Philippe Floch, Tafadzwanashe Mabhaudhi and Annelieke Duker
About 20-25 years ago, an influential literature – critically reviewed elsewhere (Lankford and Agol, 2024) – considered the virtues of adding blue water to green water through irrigation (Barron et al., 1999; Rockström, 2003; Rockström et al., 2002). This well-intentioned literature attempted to address the adverse impacts of intermittent agro-meteorological drought on rainfed agriculture, and received significant traction in water policy discourse and financing. Paraphrasing, it suggested that, particularly in semi-arid conditions, 'green water' rainfed farmers and their crops would benefit from irrigating with 'blue water' to bridge breaks in the rain, when soil moisture was insufficient, as this improves crop water productivity (Steley & Makin, 2023). This literature also identified an agricultural water continuum from wholly rainfed to partially rainfed/irrigated, to fully irrigated.
However, we believe this understanding was overly simplistic. It avoids and cannot escape questions that come with irrigated agriculture in conditions of water scarcity. Critically, this oversimplification – as appealing as it looks to policymakers and financial institutions – continues to have unintended serious consequences for a blue water crisis in water-scarce catchments and how we respond to it.
To stimulate discussion, we make three observations regarding this 'add blue water to green water' literature. 1) Whilst on paper adding an irrigation to rainfed crops makes agronomic sense, it is difficult to do this economically, technically, and institutionally at different scales. 2) The latter challenge is best understood by an irrigated systems framing which includes farmers. 3) It failed to see the important category-type differences in complexity between rainfed farming and irrigated systems. In other words, this literature saw the topic of irrigation as the act of watering crops, rather than irrigation as peopled collective agricultural systems sustainably withdrawing and controlling water over large areas in water-scarce basins and on top of aquifers. In the advice to 'top up rainfed farming using supplementary irrigation', gaps include:
- It drew up policy advice by considering irrigating at the farmer/field level, as distinguished from 'multi-scale irrigated systems' managing common-pool resources and serving multiple farmers and other water sectors in basins and aquifers. Through this omission, it failed to add internal system controls and effective water governance institutions without which blue water withdrawals and consumption increase over time and space.
- The literature did not explain how small doses of irrigation, of +/- 50 mm, would be economically and efficiently controlled and applied both 'per dosage' and over a wider area and season length. For example, small bucket-drip technologies, suited to horticultural crops, cannot easily or economically be scaled out to thousands of hectares of field crops such as maize.
- The top-up irrigation argument is predicated on the agricultural water continuum that sees no systemic differences moving from rainfed to supplementary to full irrigation, or from small to large irrigation systems, or from small irrigated areas to large coalesced cumulative areas of irrigation.
- Unclear definition of 'supplementary'. As well as adding irrigation to rainfed agriculture (the Rockström definition), rainfall is supplemental to irrigation (the definition by Perrier et al. (1991)). This is not just semantics; in the former, rainfed farmers have no experience of irrigation. In the latter, irrigators have experience working with both rainfall and irrigation.
- It was of its era, i.e., it assumed that abundant water is available for irrigation and other sectors – which may no longer apply in a world facing climate change and rising demands for water. Yet we struggle to move beyond this abundance paradigm, as witnessed by no major update to the FAO methodology for irrigation planning (Allen et al., 1998) and business-as-usual irrigation planning across most investments. Also of its era was the omission of farmers with their own agency and understanding regarding soils, crops, land, water and fellow farmers.
- It keeps the emphasis on tools that attempt to fix field-level irrigating, such as wetting front detectors (Magombeyi et al., 2021). However, such tools are rarely complemented by other methods (Lankford, 2023; Renault et al. 2007) that interrogate the constraints and structures placed, or not placed, on farmers who reside within complex irrigated systems.Yet why are such field-based 'irrigating' views appealing and how do we use these views to benefit a wider discussion?
A more contemporary omission can also be identified. Did the Global Commission on the Economics of Water (GCEW, 2023) and related publications (Stewart-Koster et al., 2024), which ascribe the global blue water crisis mainly to irrigation, miss the implications of Rockström's recommendations to add blue water to rainfed green water? In other words, why hasn't the GCEW drawn the conclusion that the blue water planetary threshold is being crossed because adding 'supplementary irrigation' to rainfed agriculture encourages blue water withdrawals and depletion?
These gaps highlight that irrigating is not the same as irrigation. We are easily drawn to 'irrigating' (the visibility and tangibility of water being added to soil and crops via pipes and channels) but we miss the multi-dimensional puzzles of irrigated systems (their architecture, dimensions, ratios, densities, and operability). The latter puzzles make the designing, managing, and governing of irrigated systems to deliver water carefully across thousands of hectares in semi-arid catchments highly challenging. The former simplistic understanding feeds the (unending) illusion that irrigation investments are straightforward and ignores decades of analytical work highlighting the nature of irrigated systems.
We are not saying rainfed farmers in Sub-Saharan Africa (SSA) should not irrigate. For example, judicious supplementary irrigation may be technically feasible where there is good well-regulated access to hydrologically sustainable surface water and groundwater.Nonetheless, accepting that water is often the missing agronomic input, we ask: What is the scarce resource in semi-arid irrigated catchments? It is rational to answer 'water' when irrigating is our refracting lens. But taking a systems view, which includes farmers (Duker, 2023) and support agencies, the scarce resource is the acuity and democratic vitality of farmer-group (i.e. 'commons') knowledge and learning, with water as the communication medium. If we don't appreciate this, we are unlikely to see farmers as water puzzlers in need of support to review their own systems. Finally, by distinguishing 'irrigation proper' from 'the agricultural water continuum' to what extent should a capacity-building programme (Lankford and Mabhaudhi, 2024) on water in agriculture concentrate on postgraduate irrigation training?
Summarising, we are asking for a more critical context-specific framing of crop watering and irrigation systems. What does this more critical framing look like, especially one that accepts that irrigation in SSA may be different from irrigation in Asia? For example, would it be more cost-effective and productive to manage and intensify existing public and FLID/smallholder irrigation systems, (recognising their problematic investments) than to expand supplementary irrigation on rainfed lands? In other words, might it be more appropriate, from a crop water productivity angle (Steley and Makin, 2023) to concentrate on water management in existing irrigation systems, rather than promoting, and then managing and regulating one or two small dosages of water by many rainfed farmers?Or, how do we effectively govern supplementary irrigation on rainfed lands?And if supplementary watering is to be limited, how do we support production by vulnerable rainfed farmers (Lankford and Grasham, 2021)?
We warmly invite you to contribute to this forum on this matter.
References
Allen, R. G., Pereira, L. S., Raes, D., & Smith, M. (1998). Crop evapotranspiration. guidelines for computing crop water requirements. FAO Irrigation and Drainage Paper 56 (FAO, Rome, Issue. [1] These reports do not necessarily reflect the views of the World Bank or national governments per the stated disclaimer.