This overview of weather information is an extract from the Royal Institute of Navigation booklet Electronic Navigation Systems Guidance for safe use on leisure vessels .

GRIB

GRidded Information in Binary (GRIB) is a data format that allows large amounts of historic and forecast weather data to be stored and transferred in relatively small files. It is used by most meteorological centres worldwide to generate their weather prediction models.

Weather prediction models

Global forecast data models include the Global Forecast System (GFS) model, produced by the US National Oceanic and Atmospheric Administration (NOAA), and the European Centre for Medium Range Weather Forecasting (ECMWF) model. There are other global models as well as many regional models. Data points in the global models are relatively spread apart, typically 9km or more between data points. Regional models reprocess the global data on a much finer grid of closer to 1km between data points, factoring in a more detailed representation of land features.

Weather apps

Weather forecasts and weather routeing capabilities may now be packaged together in a single app. The variety of such apps has markedly improved the access to, and use of, digital weather information. All of the various weather apps source their data from one or more of the data models. Some things to be aware of

• A relatively low resolution model, with data points spread 9km or more apart, is usually adequate offshore, away from the effects of land
• Regional interpolations are better for coastal sailing, but meaningful data can still only be produced for land features that are bigger than the data grid. So, even at the much higher regional resolution any localised coastal effects, such as acceleration zones off high peaks, around headlands or up and down river valleys, are unlikely to be factored in with any accuracy
• Forecast timing is important. Some apps do not use the most recent data set and, depending on the frequency at which source data is published, there may be up to a twelve hour time lag in what we see in an app
• High resolution interpolations tend to give unstable results after about 36-48 hours, so using them as the weather input for a routeing app passage plan for longer passages is usually counterproductive.

The best apps use recent, original data, at the best available resolution, from a reputable source. It is easy to get sucked into the comfort of professional displays and easy-to-use routeing programs without understanding the data behind them, but it is still just as important as ever to understand and monitor basic weather patterns in order to judge how accurately the screen predictions match reality. We should monitor a variety of forecasts for several days before a passage and then continue to monitor them, as much as possible, during the passage itself.

When weather systems are more settled the forecasts will tend to align. When systems are more chaotic the different models tend to interpolate differently and there may be diverse presentations of forecast weather. In itself this is a good indicator of how trustworthy, or not, any of the predictions might be.

Weather Sources

The list below can be downloaded for offline use as a Progressive Web App using this link.