Earthquakes in the Virgin Islands

How often are there earthquakes in the Virgin Islands?

Every day. The U.S. Geological Survey recorded 2,924 earthquakes under the Virgin Islands and Puerto Rico in the twelve months to October 8, 2026, about 8 a day, almost all of them too small to feel. 1,152 reached magnitude 3, the size people near one sometimes notice; 49 reached magnitude 4. The largest, magnitude 6.0, was 61 km ESE of Codrington, Antigua and Barbuda on May 16, 2026.

The map covers 15.9° to 20.9° N and 61.2° to 69.4° W: the east of Hispaniola to Antigua and Guadeloupe, the Puerto Rico Trench to the Muertos Trough. 79,662 earthquakes are on record in it since 1906 (USGS ComCat); 77% carry the Puerto Rico Seismic Network’s location. The page reads the USGS catalogue again every hour.

Earthquakes near the Virgin Islands this month

Within 160 km of St. Thomas in the past 30 days, magnitude 2.5 and up, newest first: 55 of the 57 recorded that close (all 444 in the map’s area are on the map). Times are Atlantic Standard Time; each magnitude links to its USGS page.

The biggest earthquakes near the Virgin Islands

Every earthquake of magnitude 6 or more in the map’s area since 1906, largest first (USGS ComCat; the early ones are the ISC-GEM relocations). 4 of them were within about 150 km of the U.S. Virgin Islands; the most recent, magnitude 6.1 in 2008, was 91 km N of Charlotte Amalie, U.S. Virgin Islands.

The October 11, 1918 earthquake (magnitude 7.1 in the ISC-GEM relocation; 7.3 to 7.5 in older catalogues) broke under the Mona Passage north-west of Aguadilla at 10:14 in the morning. Its tsunami reached 6.1 m (NOAA NCEI); the Puerto Rico Seismic Network gives 20 ft at Punta Agujereada, where 8 people drowned, and 12 ft at Aguadilla, where 32 did. The official toll is 116 dead, the 40 tsunami victims among them; NOAA's database counts 144 and other studies more. A seafloor survey ruled out the submarine landslide once blamed for the tsunami (ten Brink and others, 2023). The largest in this map's box is older and further west: magnitude 7.8 on August 4, 1946, 16 km NNW of Miches, Dominican Republic, for which NOAA's database counts 1,790 dead, its tsunami included (the USGS catalogue's older note says 100).

At 4:24 in the morning of January 7, 2020 a magnitude 6.4 earthquake broke 6 km down, just off Puerto Rico's south coast, 4 km south-south-east of Indios. It was the largest of a sequence that began on December 28, 2019, and in January 2020 the USGS forecast aftershocks for years to decades: this map's record jumps from 6,242 earthquakes in 2019 to 12,967 in 2020, and a machine-learning catalogue finds about 180,000 between December 28, 2019 and January 1, 2023 (Yoon and others, 2023). The USGS counts one death and $3.1 billion of damage (NOAA's database counts four deaths, three of them heart attacks); its tsunami was 6.3 cm on the tide gauges at La Parguera and Vieques.

Before instruments: NOAA's Significant Earthquake Database lists 18 earthquakes in this box before 1906, back to 1690. Those it places in the Virgin Islands: August 2, 1837 (Virgin Islands); April 16, 1844 (Puerto Rico: San Juan; St. Thomas); November 18, 1867 (Virgin Islands); December 1, 1867 (Virgin Islands; Puerto Rico); March 17, 1868 (Virgin Islands; Puerto Rico: San Juan); August 21, 1871 (Virgin Islands: St. Thomas); September 1871 (British Virgin Islands: Tortola Island); March 11, 1874 (Virgin Islands). One old giant has shrunk: the May 2, 1787 earthquake, long given as magnitude 8 to 8.25 in the Puerto Rico Trench, was probably about 6.9 and could have been north, west or south-west of Puerto Rico (ten Brink, Bakun and Flores, 2011).

The 1867 earthquake and tsunami

On the afternoon of Monday, November 18, 1867, two violent shocks ten to fifteen minutes apart shook the Danish West Indies, and each raised a sea wave (NOAA NCEI, after Reid and Taber, 1920). NOAA's catalogue puts the first at 18:45 UTC: 2:45 pm by today's Atlantic clocks, about 2:25 pm by the local mean time of St. Thomas; an officer at Frederiksted wrote "three o'clock in the afternoon". Magnitude estimates run from about 7.2 (Barkan and ten Brink, 2010) through 7.3 (USGS) to 7.5 (NOAA's surface-wave value).

Where it broke is still argued. Tsunami models by Barkan and ten Brink (2010) put the source on the north wall of the Virgin Islands basin, the slope south of St. Thomas that drops to more than 4,000 m, or on the platform north of it, on a fault about 50 km long striking north-west to south-east with mixed normal and sideways slip; a scarp found by multibeam survey across that wall is a candidate. The film draws one such fault (the dashed line). NOAA's account places the shocks 15 to 20 km south-west of St. Thomas, along the north scarp of the Anegada Trough.

At Frederiksted the U.S. Navy's screw sloop Monongahela lay at anchor. "Full five minutes had elapsed since the shock" when her anchor chain went "as taut as a harp string"; the sea filled the town to "a depth of twenty-four feet" along the beach street, drew back until the roadstead lay "nearly bare, for a distance of half a mile", then came in as "a high threatening wall of green water" that carried her broadside ashore, "landing on the edge of the street in a cradle of rocks" (Lt. Louis van Housel, Scribner's Monthly, 1878). The Secretary of the Navy's report of December 1868 says she was carried over the warehouses into a street, came back with the returning sea and was left on a coral reef at the water's edge, five of her crew lost; she was refloated on May 10, 1868. (The Navy's ship-history entry has her "almost a mile inland" with no crew lost; the 1868 report and the eyewitness both disagree.)

NOAA counts 24 dead: 12 at Charlotte Amalie, 5 more on St. Thomas, 5 at Frederiksted and 2 at Friedensthal on St. Croix. The account in the USGS catalogue says forty, counting 23 at Deshaies on Guadeloupe. Aftershocks went on for five to seven months; one, on March 17, 1868, raised the sea four feet at Frederiksted (USGS).

Where the waves were seen

Every observation of the 1867 tsunami in this map’s area in the NOAA NCEI Global Historical Tsunami Database, beside the first arrival this map computes from the fault drawn on the film (the range covers the other placements of the fault that Barkan and ten Brink (2010) allow). Heights are how high the water reached on land; “sea fell” means only the withdrawal was recorded.

How the arrival times were computed. The front of the first wave was solved with fast marching (scikit-fmm) at the long-wave speed √(g × depth) over the NOAA NCEI seabed at 0.0025° (about 270 m), starting from a 10 km band along the dashed fault. Barkan and ten Brink (2010) publish where and how the fault lay (the north wall, strike 120° to 135°, 50 km) but their abstract gives no coordinates, so the line on the film is our placement within those limits; the range beside each time covers the other placements they allow (turned to 120° or 135°, moved 10 km either way along the wall, or on the platform north of the basin). This is the first movement of the water, not the biggest wave, which in 1867 came later, and it says nothing about height. Two observations check it. At Frederiksted it computes 5 minutes (5–6 min); van Housel's chain went taut after "full five minutes" (NOAA's database gives 12 minutes there without saying which wave). At St. Thomas it computes 18 minutes in Charlotte Amalie (14–19 min); the USGS account says the waves came "about 15 minutes after the earthquake". The shallow bank south of St. Thomas, 20 to 60 m deep, is what slows the wave there. NOAA says some of the runups, the highest among them (15.2 m at Little Saba), came from local tsunamis raised by landslides the earthquake set off.

How fast would a tsunami reach the beaches today?

From the same 1867 source, over today’s seabed, the first wave reaches every beach and harbour on VIslandGuide’s maps in 2 to 35 minutes: soonest Cane Bay, latest Setting Point. This is one source; an earthquake elsewhere (the Puerto Rico Trench, the Anegada Passage) would give other times. The lesson does not change: the warning is the shaking.

Why the ground moves here

The Caribbean plate moves east past the North American plate at about 2 cm a year (USGS). North of the islands the North American plate goes down under the Caribbean plate at the Puerto Rico Trench, at a sharp slant to the trench. GPS measures the two plates moving past each other at about 17 mm a year at western Puerto Rico, and nearly 85% of that motion is taken up offshore to the north (Jansma and Mattioli, 2005).

The deep earthquakes outline the sinking plate. The USGS Slab2 model puts its top 43 km under Anegada, 64 km under Tortola, 77 km under St. Thomas and 135 km under St. Croix.

Between the islands the crust is being stretched. The Anegada Passage, the chain of deep basins from the Virgin Islands basin past Virgin Gorda to the Sombrero basin, was probably formed by a mix of normal and sideways faulting (ten Brink and others, 2004), and GPS measures it opening by about 2 mm a year (Jansma and Mattioli, 2005). The 1867 earthquake broke at its western end, on the north wall of the Virgin Islands basin. South of Puerto Rico the Caribbean plate's edge goes down under the island at the Muertos Trough.

That is why the tsunami risk here is about minutes, not hours. The faults are close to the coasts: for an earthquake north of Puerto Rico the warning centre's guide gives travel times that "can be less than 10 minutes". And the steep north edge of the platform has shed great landslides into the trench, scars as much as 50 km across; the USGS judges the tsunami hazard from such slope failures to the north coasts of Puerto Rico and the Virgin Islands to be high.

Tsunami warnings in the Virgin Islands

Since April 3, 2017 tsunami warnings for Puerto Rico and the U.S. and British Virgin Islands come from NOAA's Pacific Tsunami Warning Center in Honolulu. For an earthquake within 300 km and shallower than 100 km it issues a warning at magnitude 7.1 or more and an advisory at 6.5 to 7.0. The National Weather Service office in San Juan passes its messages on in Puerto Rico and the USVI, on NOAA Weather Radio, local radio and television, and as Wireless Emergency Alerts to phones.

In the U.S. Virgin Islands, VITEMA orders evacuations and sends alerts through Alert VI (sign up at vitema.vi.gov). Outdoor sirens on St. Thomas, St. John, St. Croix and Water Island are meant for people outside, not inside buildings. VITEMA's tsunami maps (2013–14) draw the evacuation zone, land below 82 ft (25 m) within 2 miles of the coast, and the assembly points: on St. Thomas the University of the Virgin Islands, Lindbergh Bay Hill and the Paradise Point parking lot; on St. John the Crest Hill and SDA Church parking in Cruz Bay, the Johnny Horn Trail and Sea Grape Hill in Coral Bay; on St. Croix the Claude O. Markoe School, the Educational Complex and UVI, and the Seventh Day Adventist School.

In the British Virgin Islands the Royal Virgin Islands Police Force control room is the 24-hour tsunami warning point, and the Department of Disaster Management (bviddm.com) runs the warning systems and publishes evacuation maps for Road Town, Tortola, Virgin Gorda, Jost Van Dyke and Anegada: yellow is the inundation zone, green is safe.

St. Thomas, St. John and St. Croix are TsunamiReady communities (National Weather Service), and every March the region rehearses a tsunami in the CARIBE WAVE exercise; the 2026 one was on March 19. No warning can outrun a wave from a fault this close, which is why every agency's first rule is the earthquake itself.

How this map is made

Earthquakes: every event the USGS ComCat holds in the box (all magnitudes, 1906 to the build day), plotted at its catalogued latitude, longitude and depth; the early ones (52, to 1971) are the ISC-GEM relocations. 3,072 events carry a depth the network did not solve for (a default such as 10 or 33 km with no depth error); they are dimmed in the 3D views. Magnitudes are as catalogued (for small events mostly the Puerto Rico Seismic Network's duration magnitude), so small events from different years are not on exactly one scale.

Seabed: the NOAA NCEI DEM Global Mosaic, read at 0.0025°, shaded from the north-west; vertical scale 1:1 in the 3D views. The mosaic blends surveys of different ages: along the trench it reads as deep as 8,622 m in places, about 200 m deeper than the deepest point measured by sonar (8,408 m), so read the trench floor as approximate.

The plate: the top of the subducting slab from USGS Slab2 (Hayes and others, 2018), the Caribbean (car) and Muertos (mue) models, contoured every 20 km.

The 1867 tsunami: see the note under the table above. Heights are NOAA NCEI's observed water heights for event 966; deaths as NCEI lists them.

1. USGS ANSS Comprehensive Earthquake Catalog (ComCat) — every event in the box, all magnitudes, 1906 to today
2. Puerto Rico Seismic Network (University of Puerto Rico at Mayagüez)
3. ISC-GEM Global Instrumental Earthquake Catalogue
4. USGS real-time earthquake feed
5. NOAA NCEI Global Historical Tsunami Database (event 966, 1867)
6. NOAA NCEI Significant Earthquake Database
7. NOAA NCEI DEM Global Mosaic
8. USGS Slab2 subduction zone geometry (Hayes and others, 2018)
9. Barkan and ten Brink (2010), Tsunami simulations of the 1867 Virgin Island earthquake, BSSA 100(3)
10. Louis van Housel, An Earthquake Experience, Scribner's Monthly (1878), transcribed by the St. John Historical Society
11. Report of the Secretary of the Navy, December 1868 (Naval History and Heritage Command)
12. Jansma and Mattioli (2005), GPS results from Puerto Rico and the Virgin Islands, GSA Special Paper 385
13. ten Brink and others (2004), New seafloor map of the Puerto Rico Trench, Eos 85(37)
14. Stewart and Jamieson (2019), The five deeps, Earth-Science Reviews 197
15. ten Brink, Bakun and Flores (2011), Historical perspective on seismic hazard to Hispaniola and the northeast Caribbean, JGR 116
16. ten Brink and others (2023), Seafloor observations eliminate a landslide as the source of the 1918 Puerto Rico tsunami, BSSA
17. Yoon and others (2023), the 2019–2023 Puerto Rico sequence by deep learning, BSSA 113
18. NOAA Pacific Tsunami Warning Center, Puerto Rico and Virgin Islands user's guide (2024)
19. National Weather Service, Tsunami safety
20. VITEMA, tsunami evacuation maps and Alert VI
21. BVI Department of Disaster Management, tsunami evacuation maps
22. NWS TsunamiReady, Puerto Rico and the U.S. Virgin Islands