Bangladesh's Hidden Earthquake Threat: The Dhaka Megacity Risk

On June 12, 1897, the ground beneath the Shillong Plateau in what is now the Indian state of Meghalaya lurched violently for approximately three minutes. The earthquake — estimated at M8.0–8.1, now known as the Great Shillong earthquake — was the largest instrumentally recorded earthquake in the Indian subcontinent's modern history at the time of its occurrence. Church spires fell in Calcutta, 200 kilometers to the south. Railway lines buckled. Bridges collapsed. Landslides blocked roads throughout the Assam hills. The shaking was felt across an area of more than 3 million square kilometers — from Rangoon to Delhi, from Lhasa to Dacca. In the epicentral region of the Shillong Plateau, the ground ruptured along the Dauki fault — the tectonic structure that had been accumulating strain against the advancing front of the Indian plate — in a surface rupture extending for tens of kilometers. The death toll was approximately 1,500 people — relatively modest for an M8.0, primarily because the epicentral region in 1897 was lightly populated hill country and because the population of the Bengal lowlands to the south, where shaking was still severe, lived predominantly in bamboo and timber construction that survived the shaking better than masonry would have.

Today, that same Bengal lowland — the Ganges-Brahmaputra delta that stretches from the Shillong Plateau south to the Bay of Bengal — is one of the most densely populated landscapes on Earth. Dhaka, the capital of Bangladesh, has grown from a city of perhaps 500,000 in 1950 to a megacity of approximately 22 million people — one of the ten largest cities in the world and one of the most densely packed, with population densities in the central city reaching 44,000 people per square kilometer in some wards. The building stock of this megacity — built in one of the world's poorest countries under minimal regulatory oversight, predominantly in non-engineered reinforced concrete frames that have never experienced a major earthquake — sits on river delta sediments that seismologists have shown will amplify any incoming ground motion by factors of 5 to 10 and liquefy across enormous areas of the city when the next great earthquake arrives.

The next great earthquake on the Dauki fault or on the active fault systems beneath Dhaka itself is not a speculative possibility — it is the expected outcome of a tectonic system that is accumulating strain at measurable rates, that has produced M8.0+ earthquakes in the historical record, and that the paleoseismic evidence indicates has a recurrence interval for great earthquakes shorter than the 129 years that have passed since 1897. Bangladesh's seismic risk — concentrated in Dhaka and in the vulnerable cities of Sylhet, Chittagong, and Mymensingh — is one of the most consequential under-discussed earthquake hazards in the world. This post covers the complete science: the tectonic setting, the fault systems, the historical earthquakes, the amplification geology, the building stock vulnerability, and what a major earthquake in this most densely populated of all delta environments would mean.

The Tectonic Setting: Three Plates at the Bengal Corner

Bangladesh occupies the extreme eastern corner of the Indian subcontinent — the point where the Indian plate, the Eurasian plate, and the Burma microplate all converge within a remarkably compact geographic area. This three-plate convergence is the source of Bangladesh's seismic hazard and the reason the country faces threat from multiple distinct fault systems rather than a single dominant source.

The India-Eurasia Collision Front

The same India-Eurasia collision that builds the Himalayas and loads the Main Himalayan Thrust beneath Nepal (covered in the previous post) extends eastward across the Shillong Plateau and into northeastern India as the northeast-directed compression of the Indian plate against the Eurasian-Himalayan orogen. In this eastern section, the primary surface manifestation of the collision is the Dauki fault — a major east-west trending reverse fault that separates the relatively rigid Shillong Plateau (an uplifted basement block of the Indian plate) from the Bengal Basin to the south.

The Shillong Plateau is one of the few places in the Himalayan foreland where ancient Precambrian basement rocks are exposed at the surface — a geological anomaly explained by the Dauki fault's reverse movement pushing the plateau upward relative to the subsiding Bengal Basin. GPS measurements indicate that the Dauki fault is currently locked — accumulating elastic strain — at a convergence rate of approximately 5–10 mm per year. This slip rate, applied to the fault's estimated seismogenic area, implies a recurrence interval for M8.0 events of approximately 300–800 years — and the 1897 earthquake occurred approximately 129 years ago, placing the fault somewhere within the early to mid portion of its estimated interseismic interval depending on which end of the recurrence range applies.

The Burma Plate and Eastern Bangladesh

To the east of Bangladesh, the situation is different but equally hazardous. The Burma microplate — a small tectonic plate that occupies the Irrawaddy basin of Myanmar and extends westward beneath the Chittagong Hill Tracts — is converging with and subducting beneath the Indian plate along the east-west-trending Indo-Burman Ranges. This subduction generates a northwest-verging fold-and-thrust belt extending from Myanmar westward through Chittagong into the Bay of Bengal — a sequence of active thrust faults that are responsible for the seismicity of eastern Bangladesh, the Chittagong Hill Tracts, and the offshore Bengal basin.

The Indo-Burman subduction zone is capable of generating its own large earthquakes — the 2023 M6.0 earthquake in Chittagong and multiple M5–6 events in the historical record from this zone demonstrate its ongoing seismic productivity. More significantly, geodetic measurements indicate that the Indo-Burman subduction interface may be partially locked over a broad area, with some researchers suggesting the potential for a future M8.5+ megathrust event on this system — an event that would devastate Chittagong (population 5 million, Bangladesh's second-largest city and primary port) and generate a significant tsunami in the Bay of Bengal affecting Bangladesh's and Myanmar's coastal populations.

The Madhupur Tract: A Fault Beneath Dhaka Itself

Directly beneath Dhaka lies a geological feature that has become increasingly important to Bangladesh's seismic hazard assessment: the Madhupur Tract — a slightly elevated alluvial terrace that represents the surface expression of the Madhupur fault system, a northwest-dipping blind thrust fault (one with no surface trace) running beneath the city itself. The Madhupur fault accommodates compression between the relatively elevated Shillong Plateau to the north and the subsiding Bengal Basin to the south, and it is independently capable of generating M7.0–7.5 earthquakes at shallow depth directly beneath Bangladesh's capital.

The Madhupur fault's blind character — it does not reach the surface — makes it invisible in conventional geological mapping and means that its seismic hazard was not recognized in Bangladesh's early building code hazard zonation. It came to scientific attention primarily through the work of Michael Steckler, Syed Humayun Akhter, and colleagues at Columbia University and BUET (Bangladesh University of Engineering and Technology) who used GPS, seismic reflection profiles, and careful analysis of the Dhaka area's geological structure to identify the fault and characterize its geometry, slip rate, and seismic potential. The recognition of a M7.0–7.5 capable blind thrust fault running beneath one of the world's most densely populated cities represents one of the most significant recent advances in Bangladesh's seismic hazard understanding — and one of the most concerning.

The 1897 Shillong Earthquake: What It Tells Us

The 1897 Great Shillong earthquake — occurring just 12 years after the invention of the seismograph and in a region with growing British colonial scientific infrastructure — is one of the best-documented pre-instrumental great earthquakes in the world. The British geologist Richard Dixon Oldham conducted one of the most thorough field investigations of any historical earthquake, publishing a landmark 1899 monograph that described the pattern of intensity observations, surface rupture features, and ground deformation in extraordinary detail and established the scientific framework for intensity-based earthquake magnitude estimation that influenced seismology for decades.

Oldham's observations are scientifically valuable precisely because they describe the effects of an M8.0 Dauki fault earthquake on the Bengal delta environment in rich qualitative and quantitative detail. The intensity distribution he mapped shows that the delta region — including what is now Dhaka, which was then a modest provincial city — experienced Modified Mercalli Intensity VII–VIII (very strong to severe shaking), despite being 150–200 km from the epicenter. This distant intense shaking reflects exactly the amplification properties of the Bengal Basin sediments: the deep, soft sediment column transmitted and amplified the waves from a distant source to produce ground motions comparable in amplitude to much closer, smaller earthquakes.

The 1897 observations of surface effects in the delta provide the most direct empirical evidence available for understanding what a future great earthquake will do. Oldham documented widespread sand boil eruptions across the delta plains, indicating extensive liquefaction in the river sediments. He recorded the cracking and partial collapse of brick masonry buildings throughout the delta region while noting that bamboo and timber structures survived largely intact — the material-selective damage pattern that defines seismic vulnerability in poorly developed building stocks. Converted to modern population and building stock terms, the 1897 pattern projects a disaster many orders of magnitude larger in the Dhaka of today.

Dhaka's Building Stock: Twenty Million People in Unengineered Concrete

No honest assessment of Dhaka's earthquake risk can avoid confronting the character of its building stock. Dhaka has experienced one of the fastest urbanization rates of any city in the world — growing from roughly 2 million people in 1980 to 22 million today, a more than tenfold increase in 45 years. This growth has been accommodated almost entirely through informal and semi-formal construction: reinforced concrete frame buildings of 3–10 stories, built by local contractors without structural engineering input, to designs that do not meet Bangladesh's National Building Code seismic provisions, using rebar quantities, cement mixes, and construction practices that reflect the economic constraints of construction in one of the world's poorest countries rather than the technical requirements of seismic resilience.

The Rana Plaza Warning

The April 24, 2013 collapse of the Rana Plaza building complex in Savar on the outskirts of Dhaka — in which 1,134 garment workers were killed when a structurally deficient eight-story building collapsed under its own weight without any earthquake trigger — provided the world with a direct demonstration of the structural quality of Dhaka's informal reinforced concrete construction. Rana Plaza was not an outlier in Dhaka's building inventory: it was a representative example of the non-engineered RC frame construction that houses millions of people and thousands of businesses throughout the city. Investigators found unauthorized additional floors, inadequate column reinforcing, poor concrete quality, and column-beam connection details that bore no resemblance to the ductile seismic detailing required by the building code. The building collapsed under its own static weight. In an earthquake, the lateral forces would be additional — superimposed on the gravity loads that already exceeded the structure's capacity.

The Rana Plaza disaster triggered regulatory reforms in Bangladesh's garment industry building safety standards and generated international pressure for building inspection improvements. But the structural quality problem it revealed extends far beyond garment factories — it pervades the residential, commercial, and institutional building stock of the entire metropolitan area. The difference between Rana Plaza and the average Dhaka apartment building is one of degree, not kind: both are products of the same informal construction system, the same limited regulatory oversight, and the same economic constraints.

The High-Rise Problem

In addition to the dominant 3–10 story informal construction, Dhaka has in recent decades added a significant inventory of taller structures — 15 to 30+ story residential towers, office buildings, and hotels built to serve the growing middle class. Many of these buildings were designed by structural engineers and nominally meet the Bangladesh National Building Code (BNBC) provisions for seismic loading. However, the soft soil conditions beneath most of Dhaka mean that the site amplification for tall buildings — which respond at longer periods (2–5 seconds) than low-rise construction — may substantially exceed the amplification factors assumed in code-compliant design. Strong motion recordings from moderate distant earthquakes felt in Dhaka show that the long-period spectral accelerations at soft soil sites in the city substantially exceed the BNBC code design values — suggesting that even code-compliant high-rises in Dhaka may be underdesigned for the ground motions they will actually experience in a major near-field earthquake.

The Amplification and Liquefaction Hazard in Detail

The Bengal Basin's seismic amplification hazard is understood with increasing precision thanks to a decade of intensive geotechnical and geophysical investigation by Bangladeshi researchers at BUET, the Bangladesh Meteorological Department, and international collaborators. The primary investigation tools — seismic refraction surveys, multichannel analysis of surface waves (MASW), borehole shear wave velocity measurements, and analysis of ambient noise recordings from the growing seismograph network — have progressively mapped the three-dimensional structure of the velocity field beneath Dhaka with sufficient resolution to generate site-specific amplification estimates for different parts of the city.

The picture that emerges from this work is consistently concerning. The deepest, softest sediments underlie the central business districts, old Dhaka, and the low-lying areas adjacent to the Buriganga River — precisely the areas of highest population density and most important economic activity. Amplification factors of 5–10 at 0.5–2 second periods are the consensus estimate for these zones. The slightly higher, older alluvial terraces of the Madhupur Tract — where the geological basement is somewhat shallower — show lower amplification (3–5 times) but remain highly vulnerable compared to any bedrock reference site.

Liquefaction susceptibility in Dhaka is very high across most of the lower-lying areas and the river-adjacent districts. The combination of saturated, loose Holocene river sands and silts at shallow depth (0–20 meters), a high water table (typically within 1–3 meters of the surface in much of the city), and ground motion levels that would significantly exceed the liquefaction trigger threshold at these soil conditions means that a major earthquake would produce widespread liquefaction-induced ground failure throughout large parts of the city — disrupting buried utilities, damaging road surfaces, undermining shallow building foundations, and producing lateral spreading along river banks and drainage channels.

Sylhet and Chittagong: Bangladesh's Other Vulnerable Cities

While Dhaka's combination of scale and vulnerability makes it the dominant scenario concern, two other major Bangladeshi cities face seismic exposures that, in relative terms, may actually be more severe than Dhaka's.

Sylhet: Closest to the Dauki Fault

Sylhet — Bangladesh's fourth-largest city with approximately 500,000 people in the urban core and a metropolitan area approaching 1 million — is located approximately 80 km southeast of the 1897 Shillong earthquake epicenter and directly in the near-field of the Dauki fault system. The city experienced strong shaking in 1897 — accounts describe extensive damage to brick buildings and widespread ground deformation — and its position closer to the fault than Dhaka means it would experience substantially higher ground motions in a repeat event. Sylhet is built primarily on river terrace sediments with amplification properties intermediate between Dhaka's deep delta and the Shillong bedrock outcrops, but the near-field source proximity more than compensates for any soil improvement. BUET scenario analyses project that Sylhet would experience ground motions of 0.4–0.8g in a Dauki fault M8.0 event — sufficient to collapse the majority of unreinforced masonry and non-engineered RC frame construction in the city.

Chittagong: Port City at the Burma Plate Boundary

Chittagong — Bangladesh's second city (population approximately 5 million in the metropolitan area) and the country's primary commercial port — sits at the western edge of the Indo-Burman fold-thrust belt, in the near-field of the active thrust faults that constitute the Burma plate boundary in eastern Bangladesh. Chittagong experiences elevated background seismicity compared to the rest of Bangladesh — M3–4 events occur several times per year — and the city's position directly adjacent to the thrust fault system means it would be the closest major population center to the rupture zone of any future Indo-Burman megathrust earthquake.

Chittagong's port infrastructure represents a specific economic and logistical vulnerability: Bangladesh's entire import and export economy passes through Chittagong port, and any major earthquake that damages port facilities, disrupts the road and rail network connecting the port to the national supply chain, or generates a Bay of Bengal tsunami affecting port operations would have cascading economic consequences extending far beyond the immediate physical damage. The 2004 Indian Ocean tsunami — which caused minor wave effects in the Bay of Bengal but provided a direct demonstration that distant great earthquakes can affect Bangladesh's coastal infrastructure — is a calibration for the potential tsunami exposure from a future Indo-Burman megathrust event.

The Preparedness Landscape: What Exists and What Doesn't

Bangladesh's earthquake preparedness infrastructure has grown substantially since the early 2000s, driven in large part by the work of BUET engineers, the Comprehensive Disaster Management Programme (CDMP) funded by the United Nations and the UK government, and international research partnerships that produced the first modern seismic hazard assessment for Bangladesh. The country now has a functioning seismograph network operated by the Bangladesh Meteorological Department (BMD), a national seismic hazard map that classifies most of northern and eastern Bangladesh in the highest hazard zone, and a Bangladesh National Building Code (BNBC) that has been revised (most recently in 2020) to incorporate modern seismic design provisions including site amplification factors for the Bengal Basin sediments.

The gap between this technical infrastructure and actual practice in the built environment is, however, enormous. Building permit enforcement in Dhaka is notoriously weak — a 2010 survey by BUET found that a large fraction of recently constructed buildings in the city had been built without required engineering certification or had deviated substantially from approved plans. The engineering profession in Bangladesh, while growing in capacity, cannot supply sufficient structural engineers to oversee the volume of construction occurring in Dhaka's rapidly expanding metropolitan area. And the economic pressure to minimize construction cost — particularly for the informal housing that provides shelter for the majority of Dhaka's urban poor — consistently overrides the technical requirements of seismic design in a country where most people have never felt a damaging earthquake and cannot viscerally imagine why seismic provisions matter.

The Multi-Hazard Compound: Earthquake in a Flood-Prone Delta

Bangladesh is already the most flood-prone country on Earth — approximately one-third of the country floods annually during monsoon season, and the major river systems that drain the Himalayas through Bangladesh regularly produce catastrophic floods that displace millions of people. Any major earthquake in Bangladesh will almost certainly occur in this context: a country whose emergency response infrastructure is simultaneously managing flood response, whose key infrastructure corridors run through flood-prone lowlands, and whose population has limited capacity to absorb additional economic shock.

The specific compound hazard interaction between earthquakes and flooding in Bangladesh is particularly concerning. Bangladesh's embankment and flood control infrastructure — thousands of kilometers of earthen embankments built to protect agricultural land and communities from river flooding — is built on exactly the liquefiable river sediments that would fail in a major earthquake. Liquefaction-induced embankment failure during or after a major earthquake would expose previously protected areas to river flooding simultaneously with the earthquake damage response — a compound event analogous to, but potentially much larger than, the flood-earthquake interaction scenarios that affect other delta environments.

Conclusion

Bangladesh's earthquake risk is, by any objective metric, among the highest-consequence under-discussed natural hazard risks in the world. The combination of a tectonic setting that has produced M8.0+ earthquakes in the historical record, fault systems including a blind thrust directly beneath the capital, deep Bengal Basin sediments that amplify ground motion by factors of 5–10, a building stock of 22 million people in a megacity built predominantly without seismic engineering, and a background of poverty and infrastructure fragility that limits every aspect of preparedness and response — creates a scenario potential that scenario analyses project at 85,000 to 200,000 deaths in a single event.

This is not inevitable. The technical knowledge to reduce this risk — through building codes, engineering education, retrofit programs, and community preparedness — exists and is being developed by Bangladeshi researchers and institutions with international support. The gap between the knowledge and the practice is the gap between what is technically required and what a country with Bangladesh's income level and governance capacity can implement in the time available before the next great earthquake. That gap is the defining challenge — and it is a challenge that, unlike the fault beneath Dhaka, is not beyond human control. The Dauki fault will rupture. The Madhupur thrust will slip. What Bangladesh will look like when they do depends on decisions being made now, in building permit offices, engineering schools, government ministries, and international aid programs, about how seriously to treat a risk that no living Bangladeshi has personally experienced but that the geological record makes unambiguous.