The earthquake of 28 February 1969: Portugal’s strongest of the century

At 3.41 on the morning of Friday 28 February 1969 the needles of the seismographs in Lisbon and Coimbra began to swing, and soon ran off the scale. Clocks stopped. In Faro, the clock on the Arco da Vila, the gate into the old town, stopped at 3.42.3 sources3 sources

The shaking went on long enough for people to understand what it was. In Lisbon men, women and children groped their way out to the street in the dark, because the electricity had failed, many in pyjamas and nightdresses. Whole families got into their cars and drove to open ground. Many spent the rest of a cold, drizzly night outside, wrapped in blankets.2 sources2 sources

In the hospital of the Misericórdia at Alhos Vedros, across the Tagus, a 20-year-old woman called Inácia da Silva was in labour with her first child. When the building began to shake, she and the midwife ran down the corridors and did not stop until they reached the square outside. Later, when the panic had passed, they went back in and she gave birth to a girl. Both were well.1 source1 source

Far out at sea, the crews of passing ships thought they had run aground or hit something. The Esso Newcastle, a 226-metre ship of 51,429 gross tons, was about 167 kilometres from the epicentre when her captain was shaken out of bed.2 sources2 sources

Off the south-west corner of Portugal, the African and Eurasian plates converge at about 4 millimetres a year. The boundary there is not one clean fault but a band of deformation about 200 kilometres wide, with faults running from the Gorringe Bank, south-west of Cape São Vicente, towards the Strait of Gibraltar. The south of the country is where Portugal's seismographs record the most earthquakes.3 sources3 sources

The same zone is the leading suspect for the great Lisbon earthquake and tsunami of 1 November 1755. Closer to living memory, the Benavente earthquake of 23 April 1909, on land in the Tagus valley, had killed more people than any other Portuguese earthquake of the century: 46 or 47 dead in the scholarly counts, though Observador puts the toll at 42.4 sources4 sources

Portugal had only three seismograph stations, in Lisbon, Coimbra and Porto. Its anti-seismic building rules dated from 1958 and applied only to buildings put up after that year. In 2019, according to Lisbon city council figures quoted by Observador, 67 per cent of the capital's buildings were older than those rules.3 sources3 sources

The earthquake began at 02:40:32 Greenwich time under the Horseshoe Abyssal Plain, a deep stretch of ocean floor near the Gorringe Bank. Portuguese clocks read 3.40, because from 1966 to 1976 the country kept the former summer time, an hour ahead of Greenwich, all year; the waves reached the Lisbon seismograph under a minute later, at 3.41. The ISC-GEM catalogue places the epicentre 188 kilometres south-west of Sagres at a depth of about 15 kilometres; studies of the source in the 1970s and 1980s put the focus at 22 to 33 kilometres.5 sources5 sources

The first communiqué of the Serviço Meteorológico Nacional, the forerunner of today's IPMA, put the epicentre about 230 kilometres south-west of Lisbon and the magnitude at 7.3 on the Richter scale. By the afternoon the US national earthquake information centre in Washington had given 7.9. Later estimates, from the full international network, settled between 7.8 and 8.0.3 sources3 sources

It was felt up to 1,300 kilometres away, in Bordeaux and the Canary Islands, across Spain and Morocco, and in Madeira, where an empty house collapsed. The strongest shaking, intensity VIII on the Mercalli scale, was in the western Algarve; Lisbon was given intensity VI. A 2021 reassessment on the modern European scale put the Algarve's maximum a degree lower, at VII.2 sources2 sources

The aftershocks came for weeks. The largest, of magnitude 6.2, struck at 04:25 Greenwich time, less than two hours after the main shock. That morning's communiqué noted another tremor felt in Lisbon at 5.28, much weaker there, at intensity III. Observador, citing IPMA's catalogue, counts 89 aftershocks on the first day and 256 by 11 March.3 sources3 sources

In Lisbon countless chimneys came down, along with poorly built walls, cornices and balconies. In Rua de São Mamede ao Caldas, between the Costa do Castelo and Rua da Madalena, balconies fell onto the cars parked below. Parts of the city lost electricity and telephones. The 25 operators on duty at the telephone faults service in Rua da Trindade kept working through the crisis, and at 3am the next day 10,000 to 12,000 calls from Lisbon to the rest of the country were still waiting to be put through.3 sources3 sources

Vieira and Oliveira, who went through the newspapers and the fire brigade's call logs for the following month, identified 909 damaged private buildings in the city, most of them Pombaline houses built after 1755 and the taller “gaioleiro” blocks that went up from the mid-nineteenth century. At the Paços do Concelho, the city hall, the ceilings of three rooms fell in.1 source1 source

The Hospital de São José, the old central hospital, was among the worst hit. Wide cracks opened in the walls and ceilings of its surgical and orthopaedic wards, some ceilings partly fell, and patients had to be moved out. Its casualty department treated 60 people that night, for cuts from glass and stones, falls and “nervous crises”.2 sources2 sources

Panic did as much harm as masonry. In Faro, a 23-year-old bank clerk threw himself from a first-floor window and was taken to hospital with several fractures. In Nazaré a woman did the same. Across the river in Cova da Piedade, the roof of a house fell in on two families, and the ceiling boards held long enough to save them.2 sources2 sources

Damage reached far north of the capital. In Leiria chimneys fell and part of a roof collapsed. At the monastery of Batalha, pillars along the cornice of the main front shifted, the carved finials fell and the main tower cracked. In Covilhã, in the mountains of the interior, people ran into the street after a long tremor, but nobody was hurt.2 sources2 sources

The Algarve was nearest the epicentre, and the western end of it, the Barlavento, took the worst. In Bensafrim, a village north-west of Lagos, more than 20 houses fell, according to IPMA's report; Vieira and Oliveira and the Diário de Notícias put the number at about 50. In Vila do Bispo and every village of its municipality many houses fell and others were badly damaged. In Fonte de Louzeiros, a hamlet in the municipality of Silves, only one of its 16 houses was left standing, and that one badly damaged.5 sources5 sources

In Lagos many buildings had to be shored up, in Rua Afonso de Almeida and Rua Direita among others, and the upper floor of the town hall cracked and threatened to collapse. The town's Misericórdia hospital was, the Jornal do Algarve wrote, “cracked like a pomegranate”, with hardly an arch over a door or window left whole. Far to the east, near the Spanish border, the upper part of the north walls of Castro Marim's thirteenth-century castle fell.3 sources3 sources

Many of the houses that fell were built of taipa, rammed earth, or of rough stone. Vieira and Oliveira put such rural houses in the most vulnerable class of the European scale engineers use to grade damage. Heavy rain and thunderstorms followed the quake, and the water got in through the cracks and damaged roofs, putting weakened buildings further at risk. IPMA's report estimates that about 400 houses fell or were ruined; Observador gives 420.3 sources3 sources

Marcelo Caetano, five months into his premiership, visited the damaged São José hospital, as the Diário de Notícias reported on 2 March. On the afternoon of 7 March he flew to the Algarve with the Minister of Public Works, stayed the night and toured Portimão, Lagos, Bensafrim, Barão de São João, Barão de São Miguel and Vila do Bispo to see the damage for himself.3 sources3 sources

The papers on the afternoon of the 28th reported one death, near Lagos, where a house fell. IPMA's report says a person in Lagos died when a wall of the run-down house where they lived collapsed. The Jornal do Algarve reported that the ceiling of his house fell on João Gregório dos Reis, 61, of Barão de São João, a village in the municipality of Lagos.3 sources3 sources

Most of the dead were not crushed. IPMA counts 11 indirect deaths in mainland Portugal; Observador describes them as victims of the panic as people fled, and the Diário de Notícias wrote in 2019 that some had suffered “síncopes”, sudden collapses. Vieira and Oliveira count 16 deaths, 3 of them direct, and record a person killed by falling rock in Portimão. Abroad, the papers reported two dead and eight injured in Salé, in Morocco, and a man of 64 in Seville who died of a heart attack.5 sources5 sources

The ship closest to the epicentre was the Norwegian Ida Knutsen, of about 32,000 tonnes, some 19 kilometres away. At 02:45 GMT she took a violent vertical blow from below. Her navigation and radio equipment was wrecked, her hull and frames buckled, and instruments were torn off the bulkheads. She had to put into Lisbon, and she was declared a total loss. The blow reached her through the water as a pressure wave, a “seaquake”, not as a breaking sea.2 sources2 sources

On the Portuguese ship Manuel Alfredo, the third officer, João Rocha, stopped the engines because he thought they had run aground. He told the Diário de Lisboa that it felt like climbing stairs, then huge waves of thick brown water broke over the bow. For hours afterwards the ship rolled in unusual swells.1 source1 source

Tide gauges along the coasts of Portugal, Spain and Morocco recorded a small tsunami. It reached Lagos and Cascais in under 40 minutes, the Tagus at Pedrouços in 52, and the Azores in a little over two hours. In Montijo a wave was seen running along the quay, at Cabedelo beach near Viana do Castelo the sea reached the dunes, and fishermen in Matosinhos and Fuseta noticed a sudden stir in the sea.2 sources2 sources

Seismologists abroad studied the records to work out how the fault had moved: López-Arroyo and Udías and McKenzie in 1972, Yoshio Fukao in 1973, and Buforn and colleagues and Grimison and Chen in 1988. At home, Porto had since 1963 had a station of the World-Wide Standard Seismograph Network.1 source1 source

They all found the same thing: a reverse, or thrust, fault, striking north-east to south-west and dipping about 50 degrees, with a small sideways component. One block had been pushed up over the other, as compression between Africa and Eurasia predicts. Fukao titled his paper “Thrust faulting at a lithospheric plate boundary”. The Porto station recorded 47 aftershocks by 24 March, and the mechanisms of the larger ones later helped pin down the plane of the fault.2 sources2 sources

On land, the only record of the strong shaking that did not go off the scale came from an accelerograph the Laboratório Nacional de Engenharia Civil had installed in the north anchorage of the Tagus suspension bridge, today's Ponte 25 de Abril. IPMA calls it a complete record; Vieira and Oliveira say a power cut stopped it after 27 seconds.2 sources2 sources

After 1969 the national network improved considerably, in steps. Even so, the analogue network of the 1970s and 1980s could locate only 20 to 25 per cent of the local earthquakes it recorded, and only three or four stations reported to Lisbon in real time.1 source1 source

Since then, 1969 has been the yardstick for 1755. Almost every study of the great Lisbon earthquake compares it with 1969 and with a magnitude 6.0 quake of 12 February 2007 in the same area, which had the same mechanism and a similarly deep source. One 2021 study describes the 1969 tsunami as “centimetric” on the south-west coast, against waves several metres high in 1755, and cites earlier work putting the energy released in 1755 at 15 to 30 times that of 1969.1 source1 source

Which fault broke in 1969 is still argued over. Ocean-bottom seismometers laid near the Horseshoe fault in 2012 recorded quakes 40 to 60 kilometres down under the Horseshoe Abyssal Plain, in the mantle. In 2019 a team led from the University of Lisbon wrote that the 1969 earthquake's source and its capacity to generate tsunamis “remain largely unknown”, partly because no fault that could be responsible shows on the sea floor. In 2021 a team from Barcelona's Institute of Marine Sciences proposed a deep thrust under the plain as the most likely source of 1755 and of other recent large earthquakes in the area.3 sources3 sources

In 2014 IPMA described 1969 as the last great earthquake on the Portuguese mainland, and the most important of the twentieth century once both magnitude and effects are counted. When it turned fifty in 2019, it was still the last quake to cause serious damage and deaths there.2 sources2 sources

Lagos, one of the worst-hit towns, marked the fiftieth anniversary in 2019 with a public session on seismic risk. The organisers, the town council and the Portuguese societies of earthquake engineering and of meteorology and geophysics, said the collective awareness of the risk fades with time and needs to be renewed.1 source1 source

The engineers' conclusion from 1969 is plain. The buildings that failed were the old ones: rural houses of rammed earth and rubble in the Barlavento, and older masonry in downtown Lisbon and Setúbal. Vieira and Oliveira single out the same types as the most fragile if a quake of similar or greater intensity strikes, which, they write, is inevitable.1 source1 source