The Sinking Delta City: Subsidence, Levees, and Land-Use Tradeoffs
A coastal city of 1.2 million people sits on a river delta of soft alluvial sediment. Groundwater pumping for industry and households has lowered the water table, and measured subsidence averages 2 cm per year in built-up districts, versus 0.3 cm per year in the surrounding undeveloped marsh. The city's flood defenses are earthen levees rated for a 1-in-50-year storm. Regional sea level has risen about 3 mm per year over the same period. A planner proposes three options: (A) raise all levees by 1 meter, (B) halt groundwater pumping and switch to a distant river intake plus conservation, (C) convert the lowest-lying 8% of the city to managed wetland and relocate residents inland. Budget allows only one option this decade.
1. Using the subsidence and sea-level figures, estimate the combined effective rate of relative sea-level rise in built-up districts and explain why this figure, not the global sea-level number alone, should frame the decision.
Read the worked answer
Relative sea-level rise = global/regional sea-level rise + local subsidence. In built-up districts: 3 mm/yr + 20 mm/yr = 23 mm/yr, roughly 23 cm per decade. In undeveloped marsh: 3 mm/yr + 3 mm/yr = 6 mm/yr.
Why it matters: flood risk depends on the relative water level against the land, not on the ocean alone. A levee rated for a 1-in-50-year storm was designed against a historical baseline; at 23 cm/decade the same storm surge overtops sooner, and the design return period silently shrinks. The 1-meter levee raise (Option A) buys time but is being consumed from below at ~23 cm/decade in the districts that need it most. The planner should frame the decision around relative rise and note that subsidence is the dominant term - and unlike sea level, it is locally controllable.
2. Compare options A, B, and C in terms of what each actually addresses, and recommend one, justifying the recommendation with the data given.
Read the worked answer
A (raise levees 1 m): addresses the symptom (overtopping height) but not the cause. At 23 cm/decade the added meter is eroded in ~40 years, and higher levees can worsen consequences when they fail. It also does nothing for the marsh, which is the natural buffer.
B (stop pumping): attacks the dominant term. If subsidence in built-up areas falls toward the natural 0.3 cm/yr, relative rise drops from ~23 mm/yr toward ~6 mm/yr - a large reduction. It is slow (compaction lags) and has political cost, but it is the only option that changes the trajectory rather than the defense.
C (managed retreat of 8%): reduces exposure permanently in the lowest land and restores the marsh buffer, but relocates people and is irreversible.
Recommendation: B as the primary action, ideally paired later with C for the lowest land. The data show subsidence is ~87% of relative rise in built-up districts (20 of 23 mm/yr); a policy that ignores the largest term while spending the whole budget on levees is treating a moving baseline as fixed. A defensible answer may favor B+C, but must justify it by the subsidence arithmetic, not by preference.