Coastal flooding in Bonabéri: drain, plinths or no-build zone?
The Douala IV council must decide how to spend a 1.2 billion FCFA flood budget for Bonabéri. The district sits on the left bank of the Wouri estuary, 1–3 m above mean sea level, and receives roughly 3,500 mm of rain a year, most of it between June and October. At spring tides the river stands 1.8–2.2 m above chart datum, and residents report that water now rises through the ground as well as falling from the sky. Flooding occurs 8–10 times a year, typically 30–60 cm deep, lasting two to four days. A 4.5 km main drain built in 2008 is now about 60% silted and choked with refuse; meanwhile 40 ha of reclaimed sand-fill plots have been sold and built on since 2015. Three proposals are on the table. (A) Rebuild the main drain. (B) Raise 300 houses on 1 m plinths and install a 400 m³/h pump station. (C) Declare the 40 ha a no-build zone, clear it and compensate the owners. The council can afford one proposal plus a small contingency.
1. Explain why flooding has become more frequent and deeper since 2008 even though the main drain was built then.
Read the worked answer
Separate three mechanisms before blaming rainfall.
Outfall, not capacity alone. A drain only removes water when its outfall is above the receiving water. With spring tides at 1.8–2.2 m and the district at 1–3 m, the outfall is submerged for hours each day during the wettest months. In those hours the drain cannot discharge and may admit river water - gravity drainage fails exactly when it is needed.
Lost capacity. At 60% silt and refuse, the channel carries perhaps 40% of design flow, so a storm it once handled now backs up.
Sealed catchment. The 40 ha of sand fill placed since 2015 removes infiltration and substitutes a hard surface, so a larger share of each storm becomes runoff and reaches the drain faster. Filling also loads and compacts the soft estuarine mud, so neighbouring plots settle relative to the new ground.
Net result: the volume arriving per storm has risen while the drain's capacity and its discharge window have both fallen. The structure was overtaken by land-use change rather than by a change in rainfall alone - which is why a 2008 drain is no answer in 2024.
2. Which proposal should the council fund, and what should be done about the houses already standing on the 40 ha?
Read the worked answer
Neither A nor C alone is defensible; a package led by B, with a regulatory element of C, is the better judgement call.
Why not A alone: because the outfall is tide-locked, gravity drainage will still fail during the high-tide storms that cause most of the 8–10 floods a year. Rebuilding it would absorb the budget without removing the peak events.
Why B: plinths give immediate, measurable protection to the 300 most exposed households at a known unit cost, and the pump station adds discharge capacity precisely when the outfall is drowned. Its weakness is that it protects property while leaving the cause - continued building on fill - untouched.
Why C is needed but cannot be crude: simply clearing 40 ha and paying cash compensation is unaffordable and politically explosive, and it punishes owners who bought plots that were officially sold. A by-law freezing new construction, combined with a transferable building right to a serviced plot elsewhere, stops the problem spreading at a fraction of the cost.
So: pump station plus targeted plinth-raising from the main budget; rebuild the drain as routine maintenance from the contingency; freeze new building on the fill. Before sign-off, check cost per protected household against the transfer scheme.
3. What additional evidence would most change your recommendation?
Read the worked answer
The recommendation rests on three assumptions, and each can be tested.
That the outfall is genuinely tide-locked during flood events. A two-month record pairing river level, tide and rainfall with drain discharge would confirm or refute this. If it showed the outfall was usually free, rebuilding the drain (A) would become far more attractive and the pump station unnecessary.
That the 40 ha really is the main source of extra runoff. A simple before-and-after comparison of runoff from filled and unfilled plots, or a hydraulic model of the catchment, would show how much of the peak the fill generates. If it is small, the no-build zone loses its priority.
That households will accept relocation. A cadastral map of the 40 ha, showing how many owners are absentee, how many live on their plots, and what they would accept in exchange, determines whether a transferable building right is workable or whether compensation would be contested in court.
Two further checks matter: a costed comparison of plinth-raising per household against a serviced plot elsewhere, and a count of households whose ground level is already below the 2.2 m spring tide, since those cannot be protected by plinths at all.