Quick Facts
| Primary Authority | NHC — nhc.noaa.gov — updated every 6 hours for active storms |
| Most Useful Product | Wind Probability Product — not the track cone |
| Reliable Forecast Window | Days 1–2; day 3–5 requires treating as a range, not a track |
| Best Long-Range Model | ECMWF (Euro) generally outperforms GFS at 5–7 days |
| Preparation Trigger | 5-day 20% wind probability in your area — begin preparation |
| VHF Forecast | NOAA Weather Radio WX1–WX7 — Caribbean-specific advisories |
The cruising community has access to better hurricane tracking tools than any generation of sailors before it. The National Hurricane Center publishes forecast products updated every six hours; satellite imagery is available in near-real time; multiple numerical weather prediction models run twice daily at global resolution; and visualisation tools built for sailors put all of it on a map at a glance. The challenge is not access to information — it is knowing which products to use, what each one actually shows, and when the information is reliable enough to act on versus when it requires a wider margin of uncertainty.
The National Hurricane Center: Your Primary Source
The National Hurricane Center (NHC), part of NOAA's National Weather Service, is the authoritative source for Atlantic tropical weather. Its forecasts, advisories, and products are updated every six hours for active tropical cyclones — at 5 AM, 11 AM, 5 PM, and 11 PM Eastern Time, with intermediate updates when conditions change rapidly. For any system that has developed to at least a tropical depression, NHC is the first place to check and the last word on official track and intensity forecasts.
NHC publishes several distinct products, and they serve different purposes. Understanding which product answers which question is the starting point for effective monitoring.
The Track Cone: What It Actually Shows
The track cone — the familiar cone-shaped graphic that shows the storm's projected path — is the most widely shared and most widely misunderstood NHC product. The cone shows the probable track of the storm's centre over the forecast period, with the cone widening to reflect the increasing uncertainty of the track at days 3, 4, and 5 compared to days 1 and 2.
What the cone does not show is the storm's size. This is the critical misreading that has led sailors to under-prepare because their location was outside the cone. The cone represents the area within which the storm's centre is likely to pass — it says nothing about the radius of tropical storm force winds or hurricane force winds extending outward from that centre. A major hurricane can have tropical storm force winds extending 200–300 miles from the centre. A location 250 miles from the cone's outer edge and "outside the cone" can still experience 50-knot winds and 10-foot seas as the outer bands pass through.
The practical rule: do not use the track cone as your primary decision-making tool for storm preparation. Use it to understand the general track. Use a different product to understand your specific exposure.
The Wind Probability Product: The Useful One
The NHC Wind Probability Product is the tool that answers the question sailors actually need answered: what is the probability that tropical storm force winds (34+ knots) or hurricane force winds (64+ knots) will pass through a specific location in the forecast period?
The product displays as a colour-coded map of the Atlantic basin, with each grid cell showing the cumulative probability of wind speeds at or above a specified threshold over the next 120 hours (5 days). A grid cell showing 40% probability for tropical storm force winds means there is a 4-in-10 chance that location will see at least 34 knots during the 5-day forecast window. This accounts for both track uncertainty and the storm's wind field radius — it is inherently a more conservative and more useful metric for preparation decisions than the track cone alone.
The 20% threshold at the 5-day range is a widely used preparation trigger. If the wind probability product shows your location with 20% or higher probability for tropical storm force winds within 5 days, begin preparation. Do not wait for a higher probability or a shorter time window. By the time probability rises to 50% at 3 days, the best anchorage positions are taken, the boatyards are full, and the conditions that make preparation safe are deteriorating. The 20% at 5 days trigger preserves adequate time to do the preparation correctly.
NHC Forecast Confidence: Days 1–2 vs. Days 3–5
NHC track forecasts are significantly more reliable at short range than long range. At 24–48 hours (days 1–2), track errors average 50–75 miles — meaningful but manageable for preparation decisions. At 72–120 hours (days 3–5), track error cones expand substantially. A storm forecast to pass 400 miles east of your location at day 5 has a realistic range that could put it 200 miles west of you, depending on model performance, upper-level steering pattern changes, and intensity fluctuations that alter forward speed.
The practical implication is that days 3–5 forecasts should be treated as an indication of general threat, not precise track information. Use them to decide whether preparation should begin; use days 1–2 forecasts to manage the details. Never delay preparation because "the track might shift away" — storm tracks shift toward boats as often as away from them.
The Model Debate: GFS vs. ECMWF
Two primary global numerical weather prediction models are available to cruisers monitoring Atlantic systems: the Global Forecast System (GFS), run by NOAA four times daily, and the European Centre for Medium-Range Weather Forecasts model (ECMWF, commonly called "the Euro"), run twice daily by the European agency.
The long-standing consensus among meteorologists is that the ECMWF generally outperforms the GFS at forecast ranges beyond 4–5 days, particularly for track prediction of Atlantic tropical systems. The famous example is Hurricane Sandy in 2012, where the ECMWF correctly forecast a northwestward landfall on New Jersey more than a week out while the GFS continued to show the storm moving harmlessly out to sea. The Euro's advantage comes from its data assimilation techniques and its handling of the upper-level atmospheric patterns that steer tropical systems.
In practice, neither model is consistently correct, and the models frequently diverge on track beyond day 4. When GFS and ECMWF agree on a track, confidence is higher. When they significantly disagree — placing the storm's projected position 200+ miles apart at day 5 — that disagreement itself is the forecast: high uncertainty, wide possible outcomes, more conservative preparation warranted.
The regional models (NAM, HWRF) focus on shorter-range, higher-resolution forecasts for the 0–72 hour window and are useful for local detail once the track is relatively well established. They are not the tools for 5-day planning.
Ensemble Spaghetti Plots
Ensemble models run the same numerical model dozens of times with slightly different initial conditions, producing a spread of possible outcomes rather than a single deterministic forecast. The tracks from each ensemble member, when plotted on a map, produce the "spaghetti plot" — a visual representation of forecast uncertainty that is directly usable by sailors.
A tight cluster of spaghetti lines at day 5 means the ensemble members are largely agreeing on the track — relatively high confidence. A wide scatter of tracks, some showing the storm moving northwest and others northeast, means the models are highly uncertain about the steering pattern — low confidence, wide possible outcomes. The spread of the spaghetti is a direct visual measure of how much uncertainty to build into preparation decisions.
The ECMWF ensemble (ENS) and the NOAA Global Ensemble Forecast System (GEFS) are both publicly accessible. Most cruiser-oriented weather platforms (Passage Weather, Windy.com, PredictWind) display ensemble tracks overlaid on the base forecast.
NOAA Weather Radio and VHF Monitoring
NOAA Weather Radio broadcasts continuous weather information on VHF frequencies WX1 through WX7 (162.400 to 162.550 MHz). Caribbean-specific advisories from NHC, tropical weather outlooks, and basin-wide tropical storm statements are broadcast on these channels and updated with each NHC advisory. For boats underway or at anchor without reliable internet access, VHF weather radio is the primary real-time source for official NHC product text.
The practice of monitoring VHF weather radio from June 1 should be standard for any Caribbean cruiser. Most modern VHF radios have a weather channel scan function; set it to scan all WX channels and alert on any broadcast. The 5 AM and 5 PM advisories (Eastern Time) are the most complete; the 11 AM and 11 PM advisories are full updates as well. During rapid development or near-active storms, intermediate advisories (Special Advisories) are issued as conditions warrant.
Hurricane Hunter Aircraft
Once a tropical cyclone reaches tropical storm strength, NOAA's Aircraft Operations Center and the US Air Force Reserve's 53rd Weather Reconnaissance Squadron begin flying reconnaissance missions into the storm. These "hurricane hunter" flights are the most accurate method of determining storm intensity — satellite estimates of maximum sustained wind are less reliable than direct aircraft measurement of pressure and wind speed.
Hurricane hunter data is incorporated into NHC advisories and is what allows NHC to issue precise intensity figures rather than ranges. Flights operate every 6 hours when a storm is within range of the aircraft's operational envelope. Each flight produces a reconnaissance report with measured minimum pressure, maximum wind speed, and storm motion data. This information feeds directly into the intensity forecasts that determine warning classifications. When reading NHC advisories and wondering how they know the storm is "80 knots," the answer is usually that a WC-130J or a P-3 Orion flew through it within the previous six hours and measured it.
Sailor-Oriented Visualisation Tools
Passage Weather (passageweather.com) and Windy.com are the two most widely used weather visualisation tools among offshore sailors monitoring tropical systems. Both display GFS and ECMWF model data on zoomable maps, with the ability to overlay storm tracks, model comparisons, and wind fields. Windy.com specifically shows tropical storm tracks from multiple model runs with a clean overlay that makes the model spread immediately visible. PredictWind offers similar functionality with GEFS ensemble integration and a sailor-specific interface.
These tools are supplements to NHC products, not replacements. They provide visual context and model comparison that NHC's text-heavy products don't offer in the same format. Use Windy.com or Passage Weather to understand the big picture and the model spread; use nhc.noaa.gov for the authoritative track, intensity, and probability products.
When to Abandon the Wait-and-See Approach
The most dangerous error in hurricane preparation is waiting too long to act because the forecast might change. Every experienced cruiser who has been through a major Caribbean storm has seen boats lost that were lost entirely because their owners waited one more advisory before beginning preparation — and that advisory came when conditions were already deteriorating, anchorages were already full, and the time available for safe preparation had closed.
The 5-day 20% wind probability threshold is a reliable operational trigger. When the NHC Wind Probability Product shows a 20% or higher probability of tropical storm force winds at your location within 120 hours, begin preparation. This is not an overreaction — at 20% probability, you are responding to a real and non-trivial threat with sufficient time to do the work properly. If the storm shifts away and your preparation turns out to be unnecessary, you have done good maintenance and have a well-prepared boat. If it doesn't shift away, you are ready.
The alternative — waiting for 50% probability or a 3-day window — leaves preparation under time pressure, in a crowded hole, with deteriorating weather, and with no margin for the unexpected problems that always arise when preparation is rushed. The history of Caribbean hurricane boat losses is not a history of over-preparation. It is a history of waiting too long.
Complete the Hurricane Preparation Series
Season Basics & Risk Gradient → | Hurricane Holes → | 72-Hour Preparation Checklist →
← Hurricane Season Guide | Season Basics →
Frequently Asked Questions
What is the track cone and why should sailors not use it as their primary decision-making tool? The NHC track cone shows the probable track of the storm's centre, widening to reflect increasing uncertainty at days 3–5. The critical misreading is that the cone represents where the centre may pass, not the extent of dangerous winds. A major hurricane can have tropical storm force winds extending 200–300 miles from the centre — a location 250 miles from the cone's outer edge and 'outside the cone' can still experience 50-knot winds. The Wind Probability Product, not the track cone, is the tool that answers the question sailors actually need answered.
What is the NHC Wind Probability Product and what preparation threshold should sailors use? The Wind Probability Product displays a color-coded map showing the cumulative probability of tropical storm force winds (34+ knots) or hurricane force winds (64+ knots) passing through a specific location over the next 120 hours. A grid cell showing 40% probability for tropical storm force winds means a 4-in-10 chance of at least 34 knots during the 5-day window. The 20% threshold at the 5-day range is the widely used preparation trigger — begin preparation when the product shows 20% or higher probability of tropical storm force winds within 5 days.
Why does ECMWF generally outperform GFS for Atlantic hurricane track forecasting? The ECMWF (the Euro model) generally outperforms the GFS at forecast ranges beyond 4–5 days, particularly for track prediction of Atlantic tropical systems, due to superior data assimilation techniques and handling of upper-level atmospheric patterns that steer tropical systems. The classic example is Hurricane Sandy in 2012, where the ECMWF correctly forecast the northwestward New Jersey landfall more than a week out while GFS showed the storm moving out to sea. When GFS and ECMWF agree, confidence is higher; when they diverge 200+ miles at day 5, that disagreement itself is the forecast — high uncertainty, more conservative preparation warranted.
What do ensemble spaghetti plots show and how should sailors interpret them? Ensemble models run the same numerical model dozens of times with slightly different initial conditions, producing a spread of possible tracks. A tight cluster of spaghetti lines at day 5 indicates relatively high forecast confidence; a wide scatter with some tracks showing the storm northwest and others northeast indicates low confidence and wide possible outcomes. Sailors should use the spread of the spaghetti as a direct visual measure of how much uncertainty to build into preparation decisions. The ECMWF ensemble (ENS) and NOAA's GEFS are both displayed on Windy.com and Passage Weather.
At what forecast lead time do NHC track forecasts become reliably actionable? NHC track forecasts are significantly more reliable at 24–48 hours, where track errors average 50–75 miles. At 72–120 hours, error cones expand substantially — a storm forecast 400 miles east of your location at day 5 has a realistic range that could put it 200 miles west of you. Days 3–5 forecasts should be treated as an indication of general threat used to decide whether preparation should begin; days 1–2 forecasts are used to manage the details. Never delay preparation because 'the track might shift away' — storm tracks shift toward boats as often as away from them.