Why the Same Earthquake Kills 10 People or 10,000

Direct answer: disaster risk is usually described as three things multiplied together — the hazard (how strong the shaking, the flood, the wind), the exposure (how many people and how much property sit where it lands), and the vulnerability (how badly those people and buildings cope when it arrives). The hazard is the only one of the three that nobody chose. The other two are the accumulated result of decisions about where people live and what they live in, and they are where almost all of the difference in outcomes comes from.

Three things, and only one of them is geology

Two earthquakes of the same magnitude, at the same depth, under the same kind of ground, can produce entirely different days. One is remembered as the morning the crockery fell off a shelf. The other is remembered for a generation.

Nothing about the physics differed. What differed is that the second one happened under a place where a great many people were, in buildings that could not take it.

This is why “how dangerous is this hazard” is not a question that has an answer on its own. A magnitude is a property of an earthquake. A death toll is a property of an earthquake and a place and a moment.

Exposure: how much is in the way

Exposure is the simplest of the three and the one that has changed fastest. More people live in more places than at any point in history, and a great deal of that growth has been in exactly the locations hazards concentrate — river deltas, coastal plains, steep valley sides, the flat ground beside a fault.

That is not carelessness. Those places are where the water, the flat land, the port and the work are. Deltas are fertile because rivers flood them. Coastal cities exist because coasts are where trade happens. The same feature that makes a place worth living in is frequently the feature that makes it hazardous, and that trade has been made deliberately, everywhere, for as long as there have been cities.

The consequence for the numbers is direct: a hazard of unchanged strength arriving at a place with more people in it produces a larger disaster, and none of that increase is a change in the weather or the geology. This is one of the reasons a rising count of disaster impacts cannot be read as a rising count of hazards.

Vulnerability is built, and mostly not by the people who bear it

Vulnerability is the hardest of the three to talk about honestly, and the easiest to get badly wrong.

It is tempting to compress it into a sentence about how some places build badly. That sentence is one short step from blaming people for their own deaths, and it is false in the way that matters most: almost nobody chooses their own vulnerability. It is produced by decisions made elsewhere and earlier.

• Where housing is affordable is where housing is exposed. Land on a floodplain or a steep slope is cheap precisely because it is dangerous. A household with no other option is not making a risk assessment; it is making the only choice available.
• A building code is not the same as an enforced building code. Most places have one. Whether it is inspected, whether an inspector can be overruled, and whether an existing building is ever brought up to it are separate questions with different answers.
• Old buildings dominate the stock. A code improved twenty years ago governs what was built since. It does not move the buildings that were already standing, and those are where most people live.
• Recovery capacity is part of vulnerability. Two households lose the same roof. One has savings and insurance and is dry within a fortnight; the other borrows at a rate it cannot service and is still worse off in five years. The hazard was identical. The outcome is not remotely comparable.
• Who is at home matters. The very old, the very young, anyone who cannot move quickly or receive a warning in a form they can use — these are the same people at higher risk from every hazard, and it has nothing to do with their choices either.

So vulnerability is best read as an inventory of decisions, most of them administrative, most of them made years earlier by people who will not be present when the ground moves.

Why this is the most hopeful part of the subject

It would be easy to read all of this as bleak. It is the opposite, and this is the single most useful thing to take from it.

Nobody can reduce the hazard. No policy makes an earthquake smaller or a cyclone weaker. But exposure and vulnerability are both made of choices, which means both can be changed — and they are the two terms that dominate the outcome.

That is why the work that actually saves lives is so unglamorous. Enforcing a code. Retrofitting a school. Not permitting the new housing on that particular slope. Clearing the drain before the season rather than after. None of it produces a moment of drama, none of it is ever visibly the reason nothing happened, and it is the reason nothing happened.

The same equation explains the reporting

Once the three terms are separate, several confusing things about disaster news stop being confusing.

• A big hazard with a small toll is not a lucky escape or an exaggeration. It is usually low exposure — it landed where few people were — or low vulnerability, which means somebody did the unglamorous work.
• A modest hazard with a terrible toll is not a mystery. It found high exposure and high vulnerability at the same time.
• Comparing two countries by death toll compares all three terms at once, and almost never says what the reader thinks it says.
• “People affected” is largely an exposure measurement. It counts who was inside the area, which is a looser number than it appears.

What this site does, and the line it does not cross

Country pages here carry a risk status built from what monitoring agencies are currently publishing — what is happening now, and what has happened before. That is a hazard-and-history picture.

What they deliberately do not carry is a vulnerability score of our own. Rating how well a country would cope requires building-stock surveys, enforcement records and household financial data that we do not hold and could not verify, and publishing a number that looks authoritative and is actually a guess would be worse than publishing nothing. Where a hazard picture is what we have, a hazard picture is what we say. For the vulnerability half, the agencies named on every page are the ones with standing, and their originals are linked.

The honest summary

The question “how dangerous is this hazard” has no answer. The question “how dangerous is this hazard, here, to these people, in these buildings” has a very precise one — and two of the three things it depends on are decisions rather than nature. When an identical event produces ten deaths in one place and thousands in another, the difference is not what happened. It is what was already there.

This article describes how disaster risk is generally framed. It is not an assessment of any country, and this site publishes no vulnerability ratings of its own.

Frequently asked questions

What is the difference between a hazard and a disaster?

A hazard is the physical event — the shaking, the water, the wind. A disaster is what happens when that hazard meets people and property that cannot absorb it. An identical hazard in an empty place is not a disaster at all, which is why hazard strength alone never predicts an outcome.

What are exposure and vulnerability?

Exposure is how many people, buildings and assets sit where the hazard lands. Vulnerability is how badly they fare when it does — construction quality, age and health of the population, whether a warning reaches people, and whether a household can recover afterwards. Exposure is about quantity; vulnerability is about capacity.

Why do similar earthquakes cause such different death tolls?

Because only one of the three terms was similar. The same magnitude under a densely populated area with older unreinforced buildings produces an outcome that is not comparable to the same magnitude under a sparsely populated area with enforced modern construction. The physics matched; the exposure and the vulnerability did not.

Is vulnerability just about being poor?

Income is strongly involved but it is not the whole of it, and framing it as a national characteristic is misleading. Vulnerability is produced by specific decisions — which land is affordable, whether a code is enforced, how old the building stock is, whether recovery finance exists. Wealthy places have highly vulnerable pockets, and specific interventions reduce vulnerability without changing income at all.

Can disaster risk actually be reduced?

The hazard cannot. Exposure and vulnerability both can, and they are the terms that dominate the outcome. Enforcing construction standards, retrofitting critical buildings, controlling what is permitted on dangerous ground and maintaining drainage all reduce risk measurably. The work is slow and never visibly the reason nothing happened.

Why are more disasters recorded now if hazards have not increased proportionally?

Partly better reporting, and partly that exposure has grown — more people and more property sit where hazards land than at any point in history. A hazard of unchanged strength arriving at a more crowded place produces a larger impact, and none of that increase is a change in the hazard itself.

Does this site publish a vulnerability score for each country?

No. Country pages carry a risk status built from what monitoring agencies are currently publishing — a hazard and history picture. Rating how well a country would cope needs building-stock surveys, enforcement records and household financial data we do not hold and could not verify, and an authoritative-looking guess would be worse than saying nothing.

Sources

• UNDRR — Hazard, exposure and vulnerability in the Sendai Framework terminology
• UNDRR — Global Assessment Report on Disaster Risk Reduction
• USGS — Earthquake hazard, shaking intensity and building performance
• WMO — Disaster risk reduction and multi-hazard early warning
• World Bank — Disaster risk management, resilient infrastructure and risk financing
• IFRC — World Disasters Report and community vulnerability assessment
• UN OCHA — Humanitarian needs assessment and affected-population figures

Found an error? See our corrections policy.