These are the forecast models available in PredictWind.

Our goal is to give you access to the most accurate forecast data available. No single forecast model will be the most accurate in every situation, so comparing multiple leading forecast models can give you greater confidence in your forecast and decision-making.

PredictWind has been the market leader for accurate forecasts in the recreational market since 2008. Using the CSIRO CCAM model with 450 high-resolution domains around the world, PredictWind covers most popular recreational marine users in the world.

PWG is PredictWind's proprietary weather model that uses NCEP global initial conditions, processed through the CSIRO CCAM model to generate the PWG forecast.

PWE is PredictWind's proprietary weather model that uses ECMWF global initial conditions, processed through the CSIRO CCAM model to generate the PWE forecast.

The ECMWF and NCEP initial conditions provide a 'photographic snapshot' of the current state of the Earth's atmosphere. We use these initial conditions to run our own proprietary weather models worldwide, using technology that is unique to PredictWind in the private weather forecasting sector.

PWAi is PredictWind's proprietary AI-powered global forecast model. It combines information from ECMWF, AIFS, Fengwu and GraphCast to generate a single optimized forecast covering the globe.

PWAi provides 1-hour time steps for the first three days and performs particularly well in validation studies over short- to medium-term forecast periods and larger-scale areas.

You can view this evaluation which represents the first phase of its validation. The results are very encouraging, PredictWind will continue refining the model through real world testing and user feedback.

ECMWF stands for the European Centre for Medium-Range Weather Forecasts. Its global forecast model has a 9 km resolution and is highly regarded by meteorologists and professional navigators around the world. It consistently performs strongly in global forecast verification and is the only global forecast model available at this high resolution.

GFS stands for Global Forecast System and is produced by the US National Centers for Environmental Prediction (NCEP). GFS is a global numerical weather prediction model and is widely used by weather services, websites and applications around the world. We display the GFS-FV3 model when you see the GFS label in PredictWind.

GFS-FV3 is able to simulate vertical movements such as updrafts, a key component of severe weather, at very high resolution. Tests suggest that the FV3 model has more accurate five-day forecasts, as well as better predictions of hurricane tracks and intensification. Although the new FV3 core has shown improvements over GFS it remains ranked 3rd for accuracy behind ECMWF(1st) and UKMO(2nd).

UKMO refers to the global forecast model produced by the UK Met Office. Also known as the Unified Model, it has a long-standing reputation as one of the world's leading global weather models.

ICON is Germany's global numerical weather prediction (NWP) model, developed by the German Weather Service (DWD) in collaboration with the Max Planck Institute for Meteorology (MPI-M).

ICON uses an icosahedral grid that provides approximately even coverage across the globe. This helps avoid some of the distortion and computational inefficiencies that can occur near the poles with traditional latitude-longitude grids.

AIFS is an AI-powered global weather forecasting system developed by the European Centre for Medium-Range Weather Forecasts (ECMWF).

It uses artificial intelligence and machine learning to generate global forecasts and performs particularly well offshore, capturing large-scale weather patterns over the medium to longer term.

HRRR stands for High-Resolution Rapid Refresh. It is NOAA's 3-km high-resolution weather model covering the contiguous United States and is updated hourly to provide detailed short-range forecasts.

HRRR incorporates radar observations and is designed to capture small-scale and rapidly developing weather systems. For more information please see this video.

NAM stands for North American Mesoscale Forecast System and is one of NOAA's regional weather models covering North America.

As a mesoscale model, NAM can represent land and other geographical features at a higher resolution than many global models, providing more detailed regional forecasts.

AROME is a high-resolution regional numerical weather prediction model developed by Météo-France. It is designed for detailed short-range forecasting and is particularly useful for forecasting smaller-scale weather features and severe weather events such as intense Mediterranean precipitations (Cévenole events).

Comparing forecasts from several models can help you assess forecast confidence. When different models show similar conditions, confidence in the forecast is generally higher. When they differ significantly, it is worth monitoring subsequent forecast updates and considering the range of possible conditions.

When viewing the PredictWind forecast tables, the menu containing the different forecast variables collapses as you scroll through the forecast days. This allows more forecast time steps to be displayed on the screen.

When the menu is collapsed, the forecast model names are shortened:

To display the full model names again, swipe all the way back to the left.