Designing for Water: Lessons from Nyundo
Introduction
Every water project must achieve two things at once: it must be technically sound and financially viable. Technical soundness means water that is safe, sufficient, and reliably delivered. Financial viability means a system that generates enough income for the CBWSO to cover its costs and for DMDO to sustain its cost-recovery model.
In Nyundo, a system was built to bring water to more than 6,000 people. Yet within two years, the water proved too salty to drink. This article explores how the original decision was made, what changed, and how DMDO is responding. The aim is not to justify but to reflect: to show that decisions were made with the best information available, to acknowledge what has gone wrong, and to learn openly from the experience.
The First Intervention (2023–2024)
The Nyundo Water System cost $180,000 and was designed to serve 6,000 people in Nyundo 1 and Nambahu villages.
Infrastructure included:
210m depth borehole with 6,100 litres per hour capacity
75m3 litre, raised (6m) water storage tank
13 automated distribution points
6,000 l/s Grundfos pump at Head of 204 meters
15.9KW solar array (30 solar panels (530W/42V)
The design met the three core objectives of technical soundness:
What Changed
In February 2025, we learned from prepaid metre data that consumption was very low.
Anecdotal evidence told us that this was because the water tasted salty.
“Most of community members goes to Chiuta for fresh water” said Said Juma, one of the elders, Nyundo I village.
“Cooking beans or cowpeas is so difficult nowadays (it takes much time) due to the high level of saltiness” said Mwajuma Bakari, a mother of two children from Nambahu village.
Our survey data then confirmed this issue.
In June 2025, we undertook new salinity tests.
Salinity Tests
The graph below shows how the Nyundo Water System was already very close to the Tanzanian Standards for tests relating to salinity before construction began and how salinity has become significantly worse since then:
Remedies Underway
DMDO, together with government partners, is now pursuing a blending solution.
It will do this by bringing in a second source. A borehole at Mikongi, which has been drilled by government, produces abundant (6000 litres per hour), good quality water. The plan is to connect this source to Nyundo so that the two supplies can be blended, reducing salinity to safe levels.
From Mikongi, which is 9 km away, a new pipeline will be laid to carry water directly into the Nyundo system. At the Mikongi site, a dedicated pump house will manage the inflow, housed within a secure fenced compound. The pump will be powered by a new solar array, complete with panels, control units, and accessories, ensuring that the system can operate reliably and without dependency on grid electricity.
Once pumped into the Nyundo storage tank, the Mikongi water will mix with the existing borehole water. The ratio will be carefully managed so that the final supply delivered through the kiosks is both safe and palatable. From there, the blended water will flow through the same network of pipelines and kiosks already built, restoring confidence in the system and providing households with the clean, affordable service they were promised.
The yield for the Mikongi borehole is enough to cover Nyundo blending and nearby villages.
Decision Analysis
The graph above is damning on our decision-making. Not only were the salinity tests for the Nyundo system very close to the Tanzanian standard when they were taken before constructions but they were much higher than the salinity measurements for our other water systems.
Why did we still go ahead?
Nyundo 1 and Nambahu had been included in our Phase 2 projects for some time and the villages were particularly exposed to water shortage. There was therefore significant pressure on the DMDO to provide water.
Was it worth the risk? Should we have thought about a mitigation at the time?
There is a well-trodden ‘standard’ decision-making process when deciding whether to develop a borehole to supply water to a community. This relies on (a) a hydrogeological survey (for water yield) and (b) a water quality test.
The results from the water quality test are returned with Pass, for tests with results within the Standard, or Fail for results beyond the Standard.
If the hydrogeological survey indicates that there is sufficient yield and the water quality test are all ‘Passed’, this acts as a green light to go ahead and develop the borehole. If some water quality tests ‘Fail’ , water quality treatments are applied.
This is the process we followed, and we developed the Nyundo Water System borehole based on this ‘green light’ hydrogeological survey and water quality test results (which all showed a Pass).
Clearly this process has been exposed here.
Lessons
Our lesson here, and one we will share with our partners and with RUWASA, is that hydrological surveys and water quality test results need more scrutiny if water systems are going to function well over the long term. Abnormally high levels of salinity (or any other water quality issue) should be flagged and further testing should be undertaken. In addition, a regimen of continual testing should be employed by all system operators as levels can naturally change over time.
In this case, the Mikongi solution would have been a good mitigation strategy and arguably it would still have been ‘worth the risk’ to go ahead with the Nyundo borehole alone at the outset. Particularly given the high need. And then we could always have fallen back on the Mikongi solution. The economics would also have justified this: the Nyundo borehole cost $30 per person and the Nyundo + Mikongi borehole will be $45 per person.
We cannot pretend that we thought like this at the time but it is some reassurance that we have a good mitigation strategy.
In future, we will review the underlying results of hydrogeological and water quality test reports and make decisions based on clear-eyed ideas about what could go wrong and the cost and practicality of a mitigation strategy.
Conclusion: Transparency in an Imperfect World
Rural water supply will always involve uncertainty. Each project is a balance between need, cost, and the limits of what can be known at the time. Nyundo has exposed a weakness in the way we interpret “passing” results and in how pressure to deliver can narrow the field of vision. The outcome is not what anyone hoped for, but it is part of the work of building something durable in a complex environment. What matters now is that the response is practical, proportionate, and rooted in learning. The experience at Nyundo will not be the last time judgement is tested, but it will help ensure that next time the questions we ask – and the evidence we insist upon – are stronger.
