Air
26 °C
dew point 26 °C
measured at the airport
METAR MHRO · September 23, 22:00
A forecast tells you it will rain. It does not tell you whether that rain lasts twenty minutes or ruins the day, which months the sea is flat enough for a boat to go out, or which month I would talk a traveller out of. That is what this page is for. I have lived on Roatán since 2014 and I take people out on the water for a living.
These readings are more than 12 hours old.
Air
26 °C
dew point 26 °C
measured at the airport
METAR MHRO · September 23, 22:00
Wind on the island
3.7 km/h
variable
measured at the airport
METAR MHRO · September 23, 22:00
Swell
1.22 m
4.9 s · E
open sea, not the West End dock
WAVEWATCH III · September 24, 01:00
Wind offshore
17 km/h
ESE
Rain : 1.25 mm/h
NCEP GFS · September 24, 00:00
Today's tides
NOAA CO-OPS TEC5655 · Port Royal, Isla de Roatan · September 24, 2026
Swell over the next days, 6 am to 6 pm
WAVEWATCH III · forecast pulled September 24, 00:55
Open-sea forecast on a 55 km grid, not the wave at the dock. The numbers only: whether a boat goes out is a call made on the day.
The satellite view below shows what is actually moving over the island right now. It is the honest answer to “is something coming”, and it beats any number I could write here and leave to go stale.
85.6 °F
Our September normal is 85.5 °F.
Computed from 2,890 satellite readings, 2003 to 2026. NOAA publishes its own anomaly of 1.7 °F against a different baseline, and the two are not the same figure.
Watch
Level 1 of 4 on NOAA's scale. The east side of the island is one step higher.
| Area | Degree heating weeks | Status |
|---|---|---|
| West End | 0.17 | Watch |
| Coxen Hole | 0.61 | Watch |
| Mary's Place | 0.60 | Watch |
| Port Royal | 0.89 | Warning |
| Guanaja | 1.84 | Warning |
NOAA scale
NOAA's scale continues above what is shown here: levels 3, 4 and 5 cover multi-species mortality, then more than 50 %, then more than 80 % of corals. The 5 km grid genuinely tells one side of the island from the other, but it does not resolve a single dive site.
Reading of September 19, 2026, NOAA Coral Reef Watch, 5 km grid.
Map by Windy
Map by Windy
The wet months are August, September, October, November, December. October counts the most rainy days of the year, 15 on average, and November receives the most water, 207.3 mm. Across a full year the island gets 1059.2 mm.
And yes, Roatán has a dry season, whatever some travel pages say. April averages 2 rainy days and March 3, against 15 in October. That is not a small difference, it is about 8 times more rainy days.
The part a rain total hides: tropical rain here usually arrives as a shower and moves on. A month with 15 rainy days out of 31 is not a month of grey skies, it is a month where you plan around the afternoon.
Why those months and not others? Because the planet's rain belt — the intertropical convergence zone — follows the sun, a month or two behind. When the sun passes straight overhead at Roatán in early May (see how high the sun gets, below), it drags that belt north: rainfall climbs from 35.2 mm in May to 86.9 mm in July. When the sun goes back down, so does the belt.
⚖️ It also explains what surprises visitors: the wettest month is not midsummer, it is November. The rain belt crosses the island again on its way south, and hurricane season adds its own moisture. August takes a breath — 55.8 mm against 86.9 mm in July — what Hondurans call the canícula. One dip in a monthly average is thin evidence, and we are not going to promise it.
On weather alone, April and March are the driest months of the year — 2 and 3 rainy days on average, with daytime temperatures around 28 °C / 82 °F.
Temperature is not a deciding factor here. The warmest month averages 29 °C (84 °F) and the coolest 25 °C (77 °F). The entire year fits inside about four degrees. Nobody picks a month on that. Rain, wind and the state of the sea are what change a trip.
Two things this page deliberately does not decide for you. Seaweed runs on its own calendar, and it has its own page: where to go to avoid sargassum. And hurricane season is a question of record, not of forecast — see below.
Roatán is a cruise port, and the number of passengers landing on an average day goes from 3,879 to 8,383 depending on the month. No weather page tells you that.
| Month | Rainy days | Passengers per day | Days with a ship |
|---|---|---|---|
| January | 8 | 7,994 | 77 % |
| February | 5 | 8,066 | 75 % |
| March | 3 | 7,520 | 77 % |
| April | 2 | 6,361 | 67 % |
| May | 4 | 4,660 | 45 % |
| June | 8 | 4,736 | 53 % |
| July | 7 | 5,311 | 52 % |
| August | 9 | 3,949 | 48 % |
| September | 12 | 4,221 | 60 % |
| October | 15 | 3,879 | 58 % |
| November | 13 | 6,946 | 80 % |
| December | 9 | 8,383 | 81 % |
passengers per day, 2026
December is the busiest month of the year, with 259,877 cruise passengers landing over the month. May is almost as dry and lands 44% fewer. And October is both the wettest month and the quietest: 15 rainy days, and fewer passengers than any other month.
This is the published schedule for 2026, not a historical record. Cruise lines move calls. Only a year whose every month is actually filled is counted: schedules are published 18 to 24 months ahead, and a half empty month would drag its average down.
Daytime and night air temperatures, the temperature of the sea, and the average number of rainy days.
🟢 The sea column is ours. We computed it from 2,890 daily satellite measurements taken off Roatán between 2003 and 2026, credit NOAA Coral Reef Watch. Air and rain figures: World Weather Online, read 19 September 2026.
| Month | Day | Night | Sea | Sunshine | Rainy days |
|---|---|---|---|---|---|
| January | 26 °C / 79 °F | 24 °C / 75 °F | 26.9 °C / 80 °F | 165.3 h23 days | 8 |
| February | 26 °C / 79 °F | 24 °C / 75 °F | 26.7 °C / 80 °F | 197.8 h24 days | 5 |
| March | 27 °C / 81 °F | 25 °C / 77 °F | 27 °C / 81 °F | 256.5 h28 days | 3 |
| April | 28 °C / 82 °F | 26 °C / 79 °F | 27.6 °C / 82 °F | 287.8 h28 days | 2 |
| May | 28 °C / 82 °F | 27 °C / 81 °F | 28.3 °C / 83 °F | 271.4 h27 days | 4 |
| June | 29 °C / 84 °F | 27 °C / 81 °F | 28.8 °C / 84 °F | 214.2 h22 days | 8 |
| July | 29 °C / 84 °F | 27 °C / 81 °F | 28.7 °C / 84 °F | 235.7 h24 days | 7 |
| August | 29 °C / 84 °F | 27 °C / 81 °F | 29.2 °C / 85 °F | 237.1 h22 days | 9 |
| September | 29 °C / 84 °F | 27 °C / 81 °F | 29.7 °C / 85 °F | 221.5 h18 days | 12 |
| October | 28 °C / 82 °F | 27 °C / 81 °F | 29.5 °C / 85 °F | 144.1 h16 days | 15 |
| November | 27 °C / 81 °F | 25 °C / 77 °F | 28.4 °C / 83 °F | 117.8 h17 days | 13 |
| December | 26 °C / 79 °F | 25 °C / 77 °F | 27.5 °C / 82 °F | 157.5 h22 days | 9 |
Sunshine is the figure that separates the months: 287.8 hours in april against 117.8 in november — 2.4 times more.
The UV index sits between 6 and 7 all year long — high enough to burn in twenty minutes at midday, in December as much as in July. There is no good month to skip sunscreen here.
The arrows move the sea column only. It is the one we computed ourselves, from 2,890 daily NOAA satellite measurements. Air and rain are published averages with no year-by-year detail, so they stay put — we will not pretend otherwise.
Rainy days rather than millimetres on purpose: our two published sources disagree by 73% on annual rainfall, and by almost 200% on September. They do agree on which months are wet, so that is what we show.
Coming from December, the sea loses 0.6 °C, and there is 1 fewer rainy day.
No storm on record since 1892.
February has the coolest sea of the year — and “coolest” means 26.7 °C (80 °F).
No storm on record since 1892.
March is the busiest month for cruise calls: about 7,432 passengers land on the island on an average day.
No storm on record since 1892.
April is the driest month of the year here — 2 rainy days on average — and the sunniest, with 287.8 hours.
No storm on record since 1892.
Coming from April, the sea gains 0.7 °C, sunshine drops by 16 hours, and there are 2 more rainy days.
No storm on record since 1892.
June is the windiest month of the year, averaging 45 km/h — the month when a boat trip is most likely to be called off.
4 storms recorded near the island in june since 1892.
Coming from June, sunshine rises by 22 hours, and there is 1 fewer rainy day.
4 storms recorded near the island in july since 1892.
Coming from July, the sea gains 0.5 °C, and there are 2 more rainy days.
3 storms recorded near the island in august since 1892.
September carries more storms than any other month: 15 of the 43 recorded near Roatán since 1892 came in september.
October is the wettest month of the year on Roatán: 15 rainy days on average, and the lowest sunshine after november.
14 storms recorded near the island in october since 1892.
November gets the least sunshine of the year: 117.8 hours, against 287.8 in april.
3 storms recorded near the island in november since 1892.
Coming from November, the sea loses 0.9 °C, sunshine rises by 40 hours, and there are 4 fewer rainy days.
No storm on record since 1892.
Wind, not rain, is what stops a day on the water. Rain passes. A sea that is up stays up, and it decides whether a boat goes out, whether snorkelling visibility is worth the trip, and how comfortable the crossing will be.
The windiest month of the year on Roatán is June, and two independent sources agree on that: our own thirty-year read of the ERA5 reanalysis, and timeanddate.com.
On how strong, they part ways: ERA5 gives a median of 26.8 km/h in June, and 36.3 km/h one day in ten; timeanddate reports 45 km/h (read 19 September 2026). The factor of 1.7 between them comes from measuring different things — timeanddate does not publish its method, so we stand on the one we know. The map below shows live gusts.
Gusts, not average wind, are what a skipper looks at. Load this the morning of your trip: it is the one number that changes hour to hour.
Loads on demand.
Each rose shows the 16 directions the wind blew FROM, over the 30 years of the climate normal. The longer the petal, the more often it blew from there.
Median significant wave height per month, and the height reached one day in ten. Offshore: this is the swell that arrives, not the one that breaks.
| Month | From | Steadiness | Wind | Swell |
|---|---|---|---|---|
| January | ENE | 0.62 | 21.4 km/h · 33 km/h one day in ten | 1.11 m · ENE |
| February | ENE | 0.73 | 23.2 km/h · 34.5 km/h one day in ten | 1.12 m · ENE |
| March | ENE | 0.75 | 24.3 km/h · 36.5 km/h one day in ten | 1.13 m · ENE |
| April | E | 0.82 | 25 km/h · 37 km/h one day in ten | 1.11 m · ENE |
| May | E | 0.78 | 24.2 km/h · 35.8 km/h one day in ten | 1.02 m · ENE |
| June | E | 0.83 | 26.8 km/h · 36.3 km/h one day in ten | 1.13 m · E |
| July | E | 0.93 | 25.7 km/h · 34.9 km/h one day in ten | 1.02 m · E |
| August | E | 0.83 | 22.3 km/h · 33 km/h one day in ten | 0.88 m · ENE |
| September | E | 0.65 | 17.3 km/h · 29.8 km/h one day in ten | 0.76 m · ENE |
| October | no settled direction | 0.34 | 16.3 km/h · 30.4 km/h one day in ten | 0.8 m · NE |
| November | no settled direction | 0.49 | 17.7 km/h · 30.9 km/h one day in ten | 0.98 m · NE |
| December | ENE | 0.57 | 17.9 km/h · 30 km/h one day in ten | 1.02 m · NE |
ERA5 does not see the island. Its grid cell is about 31 km across and Roatán sits inside one: the model sees open sea there. Read this as an OFFSHORE climate — it is not the wind on the dock at West End.
Genere a partir de la reanalyse ERA5 du Copernicus Climate Change Service (C3S) de l'ECMWF. Ni la Commission europeenne ni l'ECMWF ne repondent de l'usage qui est fait de cette donnee.
Each bar covers the middle 80 % of the 30 years: the bad years are at the top, the calm ones at the bottom, the line is the typical year. This is the question you actually have — what if I get a bad one.
Thirty years let you compute a slope. They do not let you prove one: around here the year-to-year swing is driven largely by El Niño and by the Atlantic multidecadal oscillation, whose cycles are as long as the window itself.
And that is measured, not asserted: across the twelve months, the gap between a calm year and a rough one is 6.7 times the whole thirty-year drift for wind, and 4.2 times for swell. The drift is buried in the noise.
| Month | Calm year | Typical | Rough year | Span |
|---|---|---|---|---|
| January | 19 km/h | 22 km/h | 24 km/h | 5 km/h |
| February | 21 km/h | 23 km/h | 26 km/h | 5 km/h |
| March | 21 km/h | 25 km/h | 27 km/h | 5 km/h |
| April | 21 km/h | 25 km/h | 28 km/h | 7 km/h |
| May | 17 km/h | 25 km/h | 29 km/h | 12 km/h |
| June | 21 km/h | 26 km/h | 30 km/h | 9 km/h |
| July | 22 km/h | 26 km/h | 28 km/h | 6 km/h |
| August | 18 km/h | 22 km/h | 26 km/h | 8 km/h |
| September | 13 km/h | 17 km/h | 22 km/h | 9 km/h |
| October | 13 km/h | 16 km/h | 21 km/h | 8 km/h |
| November | 14 km/h | 17 km/h | 21 km/h | 7 km/h |
| December | 14 km/h | 18 km/h | 21 km/h | 7 km/h |
Roatán sits at 16.3° north — inside the tropics. Twice a year the sun passes exactly overhead at noon, and a post stops casting a shadow. In 2026 that falls around May 5 and August 8.
The rest of the year it never gets low. At the December solstice the midday sun still stands 50.4° above the horizon here, against 25.9° in New York. That is the whole difference between a winter you endure and a winter you swim in.
And the day barely changes length: 116 minutes between the longest and the shortest of the year, 11.0 h against 13.0 h. Sunset never comes before 17:08 nor after 18:19. There is no “the days are drawing in” here.
Near the equator the sun drops almost vertically, so it crosses the horizon band fast. Civil twilight — the useful light after sunset — lasts 22 to 24 minutes here, against 33 minutes in New York at midsummer. Between “still light” and “can’t see a thing” you have about a third less time than you are used to. Plan the way back before it matters.
The sun does not set “in the west”. Over the year the point swings by about 49°, from WSW (245.7°) in December to WNW (294.4°) in June. That decides which beach gets the sunset, and when — worth knowing before you pick a room.
Honduras never changes its clocks: UTC−6 all year. New York does. So the gap between you and the island is one hour in winter and two in summer — it shifts under your feet twice a year while your destination has done nothing.
This is the one figure here that is not computed: the UV index comes from World Weather Online. It runs from 6 to 7 — twelve months, one point of spread. No winter dip, because the sun never goes low enough to give you one.
⚠️ But read it as a daily figure, not as noon. A sun standing 87° up cannot produce an index of 6 under a clear sky — at that height the midday value is far above it. Treat 7 as a floor, not a ceiling, and cover up in the middle of the day whatever the month says.
| Month | Sun at noon | UV | Day length | Sunrise | Sunset | Twilight | Sets at |
|---|---|---|---|---|---|---|---|
| January | 52.6° | 6 | 11.1 h | 06:21 | 17:30 | 23 min | WSW |
| February | 61.1° | 6 | 11.5 h | 06:15 | 17:45 | 22 min | WSW |
| March | 71.7° | 7 | 11.9 h | 05:57 | 17:53 | 22 min | W |
| April | 83.6° | 7 | 12.4 h | 05:34 | 17:58 | 22 min | W |
| May | 87.4° | 6 | 12.8 h | 05:19 | 18:06 | 23 min | WNW |
| June | 83.0° | 7 | 13.0 h | 05:18 | 18:16 | 24 min | WNW |
| July | 84.8° | 6 | 12.9 h | 05:26 | 18:19 | 23 min | WNW |
| August | 87.7° | 6 | 12.6 h | 05:34 | 18:07 | 22 min | WNW |
| September | 76.6° | 6 | 12.1 h | 05:38 | 17:45 | 22 min | W |
| October | 65.1° | 6 | 11.7 h | 05:42 | 17:22 | 22 min | W |
| November | 55.1° | 6 | 11.2 h | 05:53 | 17:08 | 23 min | WSW |
| December | 50.4° | 6 | 11.0 h | 06:10 | 17:12 | 23 min | WSW |
Computed from the date and latitude alone, with the NOAA solar equations — no outside source, nothing that goes stale. Sunrise and sunset are geometric (disc centre on the horizon, no refraction, no terrain): the real sunrise is a few minutes earlier. Honduras never changes its clocks — UTC−6 all year.
Three things on Roatán run on the moon rather than on the calendar, and none of them is folklore.
Computed with the Meeus lunar algorithm, checked against a published new moon and correct to the minute. Times are Roatán local (UTC−6, no daylight saving). ⚠️ Computed on September 28, 2026, and the moon does not repeat yearly the way the sun does: if you are reading this well after that date, the first lines may already have passed.
A shower is not a lost day here. It is a reason to move the day around rather than cancel it. What still works when the sky opens:
The water never gets cold here. It runs from 26.7 °C (80 °F) in february to 29.7 °C (85 °F) in september — a swing of only 3.0 °C across the whole year. Nobody skips a snorkelling day here because of water temperature.
Between 2003 and 2025, the sea off Roatán warmed by about 0.7 °C: 27.99 °C on average over 2003-2005, 28.69 °C over 2023-2025. That is 0.031 °C a year, measured from satellite, not projected. 23 years is not a climate in the WMO sense, which takes thirty: this is a bounded, dated measurement, not a forecast.
2003: 28.07 °C on average, −0.13 °C against 2003-2025. Warmest in September (29.6 °C), coolest in February (26.3 °C).
2004: 27.83 °C on average, −0.37 °C against 2003-2025. Warmest in September (29.5 °C), coolest in January (26.3 °C).
2005: 28.08 °C on average, −0.11 °C against 2003-2025. Warmest in September (29.8 °C), coolest in February (26.0 °C).
2006: 27.99 °C on average, −0.21 °C against 2003-2025. Warmest in September (29.7 °C), coolest in February (26.2 °C).
2007: 28.18 °C on average, −0.01 °C against 2003-2025. Warmest in October (29.4 °C), coolest in March (26.9 °C).
2008: 27.85 °C on average, −0.35 °C against 2003-2025. Warmest in September (29.9 °C), coolest in December (26.6 °C).
2009: 27.96 °C on average, −0.24 °C against 2003-2025. Warmest in September (29.7 °C), coolest in February (25.9 °C).
2010: 28.15 °C on average, −0.05 °C against 2003-2025. Warmest in September (29.9 °C), coolest in March (26.6 °C).
2011: 27.92 °C on average, −0.28 °C against 2003-2025. Warmest in September (29.8 °C), coolest in January (26.3 °C).
2012: 28.08 °C on average, −0.11 °C against 2003-2025. Warmest in September (29.5 °C), coolest in January (26.5 °C).
2013: 28.12 °C on average, −0.08 °C against 2003-2025. Warmest in October (29.6 °C), coolest in March (26.8 °C).
2014: 28.04 °C on average, −0.16 °C against 2003-2025. Warmest in September (29.3 °C), coolest in January (26.9 °C).
2015: 28.08 °C on average, −0.12 °C against 2003-2025. Warmest in October (29.6 °C), coolest in February (26.5 °C).
2016: 28.42 °C on average, +0.23 °C against 2003-2025. Warmest in September (29.7 °C), coolest in February (26.8 °C).
2017: 28.29 °C on average, +0.09 °C against 2003-2025. Warmest in September (30.1 °C), coolest in February (27.0 °C).
2018: 27.97 °C on average, −0.23 °C against 2003-2025. Warmest in September (29.4 °C), coolest in January (26.3 °C).
2019: 28.37 °C on average, +0.17 °C against 2003-2025. Warmest in October (30.0 °C), coolest in January (26.9 °C).
2020: 28.49 °C on average, +0.29 °C against 2003-2025. Warmest in September (29.9 °C), coolest in March (27.2 °C).
2021: 28.28 °C on average, +0.09 °C against 2003-2025. Warmest in October (29.7 °C), coolest in January (27.1 °C).
2022: 28.28 °C on average, +0.08 °C against 2003-2025. Warmest in September (29.5 °C), coolest in February (27.0 °C).
2023: 28.65 °C on average, +0.45 °C against 2003-2025. Warmest in September (30.3 °C), coolest in February (27.1 °C).
2024: 28.86 °C on average, +0.66 °C against 2003-2025. Warmest in September (30.6 °C), coolest in February (27.2 °C). The warmest year of the series.
2025: 28.58 °C on average, +0.38 °C against 2003-2025. Warmest in September (30.0 °C), coolest in January (27.4 °C).
Our own calculation from daily NOAA satellite measurements at 16.325 N / 86.525 W, one day in three. Only complete years are shown.
The flat-sea months line up with the dry months rather than with the warm ones: February, March, April, May combine the fewest rainy days of the year with the calmest conditions I see on the water.
But an average month does not tell you about tomorrow. Wave height is the single number that decides whether a boat leaves the dock, and the map below is the one I check before a departure.
Significant wave height, updated continuously. Under about half a metre the crossing is comfortable for everyone; as it builds, the ride gets wet and some stops come off the itinerary. Switch to swell inside the map to see where the sea is coming from.
Loads on demand.
Measured against reality: on 19 September 2026 the NOAA buoys in the western Caribbean read 0.8 m and 1.5 m, and the wave models used by these maps were within 10 cm of both. Sources: NOAA NDBC buoys 42056 and 42057.
Across the 360 months from 1991 to 2020, the dominant swell always came from somewhere between north and east: 351 months from northeast to east, 9 from north or north-northeast. Not a single month from the south or the west (0 out of 360). The north shore and the East End take it head on; the south shore and the West End sit out of the direct swell.
One month stands apart from all the others: October 1998, hurricane Mitch, with a rough-day swell (90th percentile) of 3.50 m. The next highest month tops out at 2.22 m (October 2005).
Offshore swell, on a 55 km grid (0.5°): the model sees neither the reef nor the shelter of the coast. It says where the sea comes from, not how it breaks on a given beach. "Out of the direct swell" is geometry, not a measurement.
One square per month, darker the more the swell came from the east. The circled square is the roughest month of the thirty years.
NNNENEENEE
| Month | Months per direction (30 years) | Typical | Rough day | Worst year |
|---|---|---|---|---|
| January | ENE 18 · NE 11 · E 1 | 1.11 m | 1.74 m | 2003: 2.15 m |
| February | ENE 26 · NE 3 · E 1 | 1.12 m | 1.62 m | 2016: 1.92 m |
| March | ENE 25 · E 3 · NE 2 | 1.13 m | 1.74 m | 1998: 2.05 m |
| April | ENE 23 · E 7 | 1.11 m | 1.68 m | 1998: 1.97 m |
| May | ENE 17 · E 13 | 1.02 m | 1.55 m | 2003: 1.84 m |
| June | E 27 · ENE 3 | 1.13 m | 1.57 m | 2019: 1.74 m |
| July | E 23 · ENE 7 | 1.02 m | 1.42 m | 1992: 1.67 m |
| August | E 15 · ENE 15 | 0.88 m | 1.32 m | 2007: 1.56 m |
| September | E 15 · ENE 14 · NE 1 | 0.76 m | 1.20 m | 2010: 1.37 m |
| October | ENE 15 · NE 10 · NNE 4 · N 1 | 0.80 m | 1.36 m | 1998: 3.50 m |
| November | NE 20 · ENE 7 · NNE 2 · N 1 | 0.98 m | 1.65 m | 1996: 2.19 m |
| December | NE 19 · ENE 10 · NNE 1 | 1.02 m | 1.56 m | 2018: 1.99 m |
Genere a partir de la reanalyse ERA5 du Copernicus Climate Change Service (C3S) de l'ECMWF. Ni la Commission europeenne ni l'ECMWF ne repondent de l'usage qui est fait de cette donnee. Coastline: © OpenStreetMap contributors.
Roatán has a small tide. Across a full year of NOAA predictions, the daily range averages 49 cm, and the month barely matters: from 46 cm in the quietest month to 52 cm in the strongest. Six centimetres apart.
What does change, and by a lot, is the day. The range runs from 16 cm to 80 cm depending on the moon — a factor of five. So there is no useful “best month for tides” on this island. There are good days and big days.
Half a metre decides nothing for a boat. It decides a great deal over the shallow reef: on a big low tide, the water above the coral gets thin, and a snorkeler who stands up or kicks down does real damage. If you are snorkelling the shallows, go on the rising tide, keep horizontal, and never stand on the reef.
Tide predictions for your dates, from the NOAA station on the island
Short version: Roatán sits at the southern edge of the Caribbean hurricane belt and is hit far less often than people assume — but “less often” is not “never”, and 1998 is the reason nobody here jokes about it.
That question deserves numbers, not reassurance, and it has its own page with every storm recorded since 1892: hurricanes in Roatán, what 132 years of records actually say.
And here are the tracks themselves. Every line is a real storm recorded by NOAA; the list on the left tells you which. By default: the ones that came within 300 km of the island since 1966 — the year satellites began seeing them all.
Three questions this tool answers: has one already come close? in which month? and that one storm — where did it go?
Matching storms
MITCH was a category 5 — but a category 2 when it passed Roatán. This chooses which of the two every count uses.
El Niño years are quieter here: 0.40 storms a year against 2.14 in La Niña, over the satellite era.
Before 1966, a storm that crossed no ship went unrecorded. The older the window, the more it under-counts.
Loading 2,004 storm tracks…
Old counts are incomplete. Before satellites, a storm that crossed no ship went unrecorded, so the decades from 1851 to 1950 are under-counted. Comparing 1870 with 2020 on this map would compare the record, not the weather.
Open the tool full page — all 2,004 Atlantic storms since 1851, with every filter and the CSV export.
NOAA’s record holds tropical cyclones only — no tropical waves, no cold fronts. Yet from November to March, cold North American air spills across the Gulf of Mexico and reaches this far south. Hondurans call it a norte: a steady northerly that blows for two to four days and builds a swell on the island’s north shore.
Our own numbers carry its fingerprint. January is both the coolest month of the year (25 °C average) and by far the wettest of the dry half: 144.7 mm across 8 rainy days, against 45.4 mm across 3 days in March. That is not the rainy season — that is the nortes.
So “not one storm from December to May” is accurate, and it does not mean “dead calm”. It means no tropical cyclone. Winter here runs on a wind regime of its own, one that carries no hurricane name and appears on no map on this page. The practical consequence follows by itself: a trip called off in January cannot be down to a hurricane — there has never been a single one — but a norte will do it.
In 43 recorded approaches since 1892, not one came in January, February, March, April, May and December.
storms recorded within 100 km of the island, by month
September and October account for 29 of the 43, or 67%.
A month with no storm on record since 1892 is not a month where nothing can happen. It is a month where nothing has. We keep the distinction, because a record is not a guarantee.
Every storm that has passed within 100 km of Roatán since 1950, sorted by the ENSO phase of its hurricane season. We computed this ourselves from two public NOAA datasets.
Right now: El Niño ONI index +1.8 (JJA 2026)
El Niño is a periodic warming of the central Pacific Ocean. NOAA measures it with the ONI — the Oceanic Niño Index — which is simply how far that patch of ocean sits above or below its own long-term average, in degrees Celsius, over a three-month window. “JJA” means June-July-August.
The official threshold is +0.5 °C for El Niño and −0.5 °C for La Niña. Anything in between is called neutral. Today's reading of +1.8 is 3.6 times that threshold.
A warm Pacific changes the winds high above the Atlantic, increasing the shear that tears young storms apart before they organise. That is the accepted explanation for why El Niño years are quieter here. Our own count, above, is a correlation measured on 75 years of records — it agrees with the explanation, it does not prove it.
storms per year
In 20 El Niño years, one single storm came near the island. In 18 La Niña years, twelve did.
Honest caveat: 25 events is a small sample. But the result holds up to testing — we expected 6.6 storms in El Niño years and counted 1 (probability 0.0105), and the neutral phase lands exactly on expectation (0.519), which is the control. This is a strong indication, not a forecast for any given year.
Sources: NOAA Climate Prediction Center (ONI) and NOAA HURDAT2. Read 2026-09-19.
The same questions come back every week, so here are the short answers with the figures attached. Ask me anything that is not here — I answer these myself.
Scroll or use + and − to zoom, drag to move, Esc to close.