mediating relationships and ways of knowing in groundwater irrigation
By Tavengwa Chitata , Jeltsje Kemerink-Seyoum  and Frances Cleaver 
In this vignette we discuss based on empirical data collected in the Rufaro Irrigation scheme in Zimbabwe how infrastructure matters in shaping the relationship between farmers and aquifers. In particular, we will show how the form and materiality of infrastructure shapes knowledge on the groundwater sources and how it mediates relations among irrigators. On this basis, we will argue that changes in infrastructural arrangements therefore often also changes ways of knowing water as well as social relations of power among water users. For this research we use socio-technical tinkering (Kemerink-Seyoum et al., 2019) as an entry point to study the mundane everyday practice of the farmers and engineers in their interaction with infrastructure to abstract, direct, share and use groundwater. In this approach, we pay attention to acts of tinkering which produce deviations from the original plans and designs of irrigation infrastructure in terms of the form, materiality and functioning of the infrastructure as well as the changes it provokes in social relations and institutional arrangements.
2. Knowing groundwater with infrastructure
In the Rufaro community, groundwater is the only source of water for irrigated agriculture, animal watering and domestic water use. Since the rehabilitation in 2016 by an FAO programme funded by the SWISS Cooperation the irrigation scheme is dependent on seven boreholes, each 60 metres deep. They all operate on electricity and are equipped with submersible pumps. Before the boreholes broke down in 2012 the irrigation scheme relied on six boreholes, equipped with mono-pumps and operated on electricity, plus another borehole, which was equipped with a mono-pump and operated on diesel. The boreholes are connected to a mainline pipe which supplies water to the water storage tank. From this tank water is piped into the irrigation scheme. The domestic water needs are supplied from a 60-metre deep hand-pumped borehole, a Zimbabwean bush pump (B-type).
These different pumping technologies used to abstract groundwater are essentially the connection mediating the interactions between the aquifers and the people. Most male farmers, who are predominantly involved in irrigation activities, are not worried about the groundwater levels. According to these farmers the groundwater level does not change. They say it is still the same for as long as they can remember and they do not anticipate water shortages in the near future. However, farmers who used to be responsible for operating the diesel borehole acknowledge that, although they never faced shortages of water, there were significant fluctuations in the level of the water when the irrigation scheme was in full operation. This diesel pump for several years (2010 – 2012) was the only functioning pump in the irrigation scheme. The farmers involved in operating this pump say that they used to observe the fluctuation in water level through the sound of the former diesel engine. They complained that the current system of submersible pumps operated on electricity does not allow for this embodied way of knowing the condition of the groundwater source. One of the former operators of the diesel engine corroborated the idea that technology matters to how people relate to, know and feel about the groundwater:
“With the diesel engine, it was easier to know if the water levels were getting low. The sound of the diesel engine would change when the groundwater levels were low, and the time it would take for water to get to the tank was dependent on the groundwater level. With these electricity-powered submersible pumps, you can only tell if it is pumping or not when you are within a metre of radius to the borehole, and they cease in silence".
The sound of the diesel operated borehole served to warn the farmers if there was a technical problem with the borehole. This allowed them to stop the engine and repair the pump before it was completely broken. However, with the submersible borehole the sound does not change even when it is not pumping water. The warning sign of the diesel mono-pump before breaking down saved the farmers from losing their crops as they would plan for the repairs without affecting their irrigation schedules. Moreover, the farmers indicated that this former diesel pump was easier for them to repair because it resembles a car engine – something they have more experience with – and could be maintained with locally available spare (car engine) parts.
Women and children, who predominantly participate in the provision for domestic water using the hand-pump, argue that the groundwater levels are decreasing. They say the decrease in water levels is more noticeable when all the boreholes are in operation to supply water for irrigation. Interestingly, they share the aural perception to knowing about groundwater levels, and the hand-pumped borehole is said to produce a defined squeaking sound when the water level is very low; a result of more pipes being above the water table. Moreover, women say the water levels are decreasing drastically as they need to put in more and more efforts to pump the water out of the borehole. One woman explains the differences in the way men and women know about the aquifer to the kind of infrastructure they use to obtain their groundwater:
‘’Those whom you asked [interviewed] do not know that the groundwater levels are decreasing. How can they know when water comes to them or is pumped to the tank after one person presses on the [electricity] switch? We know the water is decreasing because of the physical efforts we put into getting water out of the borehole. When the water level is close to the surface, we use less effort, but when the water level is too low we need to use more effort. These days we really work hard to get the water out and it’s a pity they tell you the water level is not decreasing.’’
Whether these are seasonal fluctuations or a declining trend of the groundwater levels remains unclear. Also not all of the ways of knowing the quantity and quality of groundwater in the Rufaro community can be (yet) explained and ambiguity remains as some of the women could not explain how they knew that the groundwater level was decreasing. They, however, stressed that they know the aquifer because “because we come and fetch the water every day”, while the men only irrigate once in a while.
3. Changing the downstream and upstream relations
Most empirical studies on irrigation schemes document the privileged access to water for farmers located in the upstream part of the system (Komakech et al. 2011). This was also the case in the Rufaro irrigation scheme for three decades until they changed the water infrastructure from open canal irrigation to a piped surface irrigation system in 2016. The farmers did not support this modification of the infrastructure as they believed the pipe system would be less flexible to irrigate with than the open canals. However, the engineers argued it would increase the water efficiency of the system and on that basis the pipes were installed. As one of the engineers recalled:
‘’We have changed the water infrastructure from open canal irrigation system to pressurised pipe surface irrigation system. With this change, we have managed to improve the overall efficiency of this irrigation system to sixty-five percent from forty-five percent. This is a commendable accomplishment and will lead to the sustainability of this irrigation scheme”.
However, what none of the actors had foreseen was that by replacing the infrastructure the upstream-downstream relationship between the irrigators was also drastically modified. In the open canal irrigation system, upstream irrigators had better access to water and would irrigate first as they could easily direct the water to their fields and the downstream irrigators had to negotiate with the upstream irrigators for the release of water. The upstream irrigators would irrigate as often as they deemed fit for their crops and as long the water was flowing in the canals. On the other hand, the downstream farmers only irrigated when the upstream farmers were not irrigating or when they formed teams to guard the canals to secure water flow to the downstream irrigation plots. Although there are no historical records to compare the yields in the upstream and downstream plots, the farmers who are located in the downstream end regularly stopped irrigating their plots and could not harvest any crops (see also Figure 1).
Figure 1: Location of irrigation plots which were abandoned during the time of the open canal irrigation system.
The new piped system benefits the downstream farmers because pressure within the pipes needs to be built up from downstream: water flows from the tank into the pipes and when a pipe fills up to the first opening it encounters, it will flow into the field by gravity. If the downstream farmers do not close their hydrants, the water will continue to flow out of the pipes into their fields, reducing the pressure in the system and leaving the upstream irrigators with no water in their part of the pipes.
The change in the form and materiality of the infrastructure has thus restructured- or even reversed - the relationships between the upstream and downstream irrigators with actual implications for the livelihoods of the different farmers. As one upstream farmer explained:
“I used to irrigate at any time as I wished because I am in the upstream, but not anymore. Today I came to irrigate but until those in the downstream have finished, I cannot irrigate. I will go home and come back later”. Simultaneously, an owner of a plot downstream welcomed the pressurized pipe system and the associated change in power hierachachy when it comes to accessing water. He commented that “I once gave up on irrigating my plot in the downstream because I could not get water, but now that problem is over”.
In this vignette we documented acts of tinkering with water infrastructure to show how the form and materiality of infrastructure matters in mediating socio-nature relationships. The example of the changes in the pumping technology used to access groundwater showed how the type of infrastructure shapes how people know, and thus relate to and engage with, aquifers. The gendered water practices, and the different infrastructures used for irrigation and domestic water use, make that knowledge on the aquifer is highly gendered. The second example of the change in the infrastructure to distribute water to the fields shows how infrastructure matters in shaping the relationships between irrigators. It unpacks how engineering solutions to techno-managerial objectives – such as water efficiency – can lead to (unitended) consequences in terms of access to water and as such alter social relations of power.
Picture 1: The boreholes which supply water for domestic uses
Picture 2: A new borehole system equipped with submersible pump and fixed onto the position of the old mono pump system.
Picture 3: Remains of the old diesel operated borehole system
Kemerink-Seyoum, Jeltsje Sanne, Tavengwa Chitata, Carolina Domínguez Guzmán, Luis Miguel Novoa-Sanchez, and Margreet Z. Zwarteveen. 2019. “Attention to Sociotechnical Tinkering with Irrigation Infrastructure as a Way to Rethink Water Governance.” Water (Switzerland). https://doi.org/10.3390/w11081670.
Komakech, Hans Charles, Marloes L. Mul, Pieter Van Der Zaag, and Filbert B.R. Rwehumbiza. 2011. “Water Allocation and Management in an Emerging Spate Irrigation System in Makanya Catchment, Tanzania.” Agricultural Water Management. https://doi.org/10.1016/j.agwat.2010.07.017.
 PhD Candidate, University of Sheffield and a member of the T2GS team
 IHE Delft, Department of Water Governance. University of Amsterdam, Governance and Inclusive Development
 Chair in Political Ecology, Lancaster Environment Centre, Lancaster University