Maga dam: releasing water for the floodplain in a dry year
The Maga dam in the Far North stores Logone water for a 5,500 ha rice scheme run by a parastatal. When the reservoir spills, the same water inundates the downstream floodplain, where about 20,000 people in several villages graze cattle on dry-season bourgou pasture and fish in residual pools. Since the dam closed, rice output has risen but fish catches and pasture have fallen, and a sluice is now opened each dry season to restore some flooding. This year the reservoir stands at 42% of capacity in early December because the August–September rains failed. The rice scheme says it needs about 90 million m³ to bring the full area to harvest. Opening the sluice for three weeks would release about 25 million m³ and flood roughly 60 km² of floodplain. Scheme managers argue that rice means cash and calories; herders and fishers reply that they have nothing else. The governor wants a written recommendation before the sluice decision on 20 December.
1. Explain how opening the sluice could be justified even though it reduces the rice harvest.
Read the worked answer
The comparison is not rice versus nothing. It is the value of the last few tonnes of paddy versus the value of the whole dry-season economy of 20,000 people.
First, the 25 million m³ needed for the full 5,500 ha is not evenly productive. The area the scheme would have to abandon is its least productive ground, so the rice actually forgone is a margin, not the average. If the reservoir at 42% cannot support 5,500 ha in any case, releasing water may cost less paddy than managers claim, particularly if the release is timed to spare the best fields.
Second, floodplain water does several jobs at once. It produces bourgou pasture that carries cattle through the months when there is no grass - cattle being the herders' main asset, so loss of condition means low sale prices and deaths. It recharges pools where fish breed and survive, supplying cheap protein to towns. It lifts the water table, benefiting wells and recession gardens far from the river. These benefits are produced by the same cubic metre that would grow perhaps a few hundred kilograms of paddy.
Third, a decision that visibly sacrifices 20,000 people's dry-season livelihood to protect a parastatal's harvest invites conflict and illegal breaching, which is a worse outcome than a managed release.
2. What information and thresholds would you require before the 20 December decision?
Read the worked answer
The decision needs four sets of numbers, each attached to a threshold.
Water accounting. Reservoir volume now, expected inflow, evaporation losses and the scheme's real crop water requirement per hectare. Paddy typically needs 10,000–15,000 m³/ha over a season, so the stated 90 million m³ implies the area that can actually be finished. Threshold: if the scheme cannot finish 5,500 ha even with the full allocation, it cannot claim to lose them.
Value per cubic metre. Compare the value of the marginal paddy grown with 25 million m³ against the value of the meat, milk and fish produced by flooding 60 km². Threshold: release if floodplain value per m³ exceeds marginal rice value per m³.
Who bears the loss. Herd numbers, body condition at the end of the dry season, number of fishing households, and whether stock would have to be sold. Threshold: if cattle deaths or distress sales are likely without flooding, the release becomes a welfare decision, not an economic one.
Timing and forecast. When the next rains are expected and how fast the reservoir refills, since a release in December costs more than one in February.
Without at least the water accounting and value-per-m³ figures, a recommendation would be advocacy dressed as analysis.
3. The governor asks how to stop the same fight recurring every year. What institutional arrangement would you propose?
Read the worked answer
The fight recurs because there is no agreed rule, so every year becomes a negotiation of strength between a parastatal and unorganised users.
Create a basin committee with a legal basis, seating the rice scheme, representatives of herder and fisher groups from the affected villages, the divisional officer and the technical services. Publish its minutes.
Agree a rule in advance, with triggers. For example: the sluice opens for a set period once the reservoir reaches a stated level on a stated date, and the release is reduced or skipped only below a lower threshold. Because the rule is set in a wet year, it is not renegotiated under pressure in a dry one.
Share the benefit, not just the water. Use part of the scheme's revenue to fund cattle feed from rice bran and husk in years when flooding fails, and to maintain water points on the floodplain. Compensation in kind makes the rule acceptable to the scheme's managers.
Monitor and publish. Reservoir level, area flooded, fish catch and cattle condition should be recorded each year, so that the next decision rests on evidence rather than assertion.
Provide an appeal route to the regional authority when the rule is disputed, so that conflict is channelled rather than expressed by breaching the embankment.