Manufactured vulnerability in the Ganges–Brahmaputra–Meghna delta: engineered drivers and coastal displacement in south-west Bangladesh

Arghya Protik Chowdhury 

Displacement from the Ganges–Brahmaputra–Meghna delta in south-west Bangladesh is usually attributed to sea-level rise, cyclones and salinity. The evidence assembled here points to a jointly produced process, in which hazards act together with engineered landscape change, governance arrangements and the interval households spend unable to return. Peer-reviewed and grey literature for Khulna, Satkhira and Bagerhat is sorted into four categories: physical transformation, ecological degradation, human impact and governance process. The strongest evidence concerns polder-related elevation loss: land inside embankments has lost 1.0–1.5 m of relative elevation against the adjacent Sundarbans and natural riverbank terraces, a differential of roughly 20–30 mm yr⁻¹ over five decades and some seven to ten times the regional water-level trend of about 3 mm yr⁻¹. Chronic waterlogging affects approximately 128,000 ha for six to nine months annually. Modelled dry-season sodium intake from drinking water in one Khulna upazila reaches 2.5 to 8 times the WHO recommended maximum for adults, although the measured urinary biomarker in the same study is roughly a third of the modelled upper bound. Attribution is far weaker: the contributions of aquaculture, upstream freshwater reduction, subsidence and climate forcing to salinisation remain unresolved. A second finding is methodological. Vertical land motion estimates for this delta diverge by up to an order of magnitude, and the divergence is a category error rather than an empirical dispute, because gauge, InSAR and stratigraphic methods sample different depth intervals and timescales. Across three cyclones spanning thirteen years, national mortality fell 131-fold while embankment breaches and livelihood disruption persisted for months to years afterwards. Recovery latency, the interval between hazard-related loss of return conditions and their restoration, is therefore proposed as a complementary measure for climate-displacement research. Restoration intervals are reported only anecdotally and at landscape scale in the reviewed literature, never at household level against a defined restoration criterion, so the quantity is right-censored and needs survival-analytic estimation from panel data that do not yet exist.