Where weather leaves a footprint

Weather Footprints is an automated, probabilistic weather-impact assessment system written primarily in Ada. Its purpose is not simply to forecast meteorological conditions, but to identify where potentially hazardous weather is most likely to have meaningful consequences for people, transport and infrastructure across the UK and Ireland.

The system downloads forecast data from NOAA's Global Ensemble Forecast System (GEFS), processing 31 ensemble members over an eight-day forecast period. It contains its own GRIB2 decoding and data-processing infrastructure, converting the model output into probabilities for six weather hazards: snow, heavy rain, high winds, thunderstorms, frost and ice, and fog.

Each hazard is assessed against three configurable severity thresholds. The proportion of ensemble members exceeding each threshold provides an initial probability estimate. Frost, fog and thunderstorms additionally use statistical calibration models fitted against historical forecasts and airport observations, intended to correct systematic forecast biases and improve probability reliability.

Weather Footprints then incorporates geographical exposure. It maintains spatial datasets describing population, major roads, elevated roads, railways, electricity infrastructure, hospitals, schools, care homes, airports and areas susceptible to surface-water flooding. These are combined using hazard-specific weightings, recognising that different weather events affect different infrastructure in different ways.

For each location, the system calculates an exposure adjustment, which modifies the underlying meteorological impact associated with each severity threshold. Impact and probability are then combined into a single risk score.

The maximum score across the three severity thresholds produces a risk index on a scale from 0 to 10. This is an indicator of potential weather impact, not a prediction of specific incidents or a calibrated probability of disruption.

The geographical risk fields are interpolated onto a finer mapping grid, smoothed and analysed to identify contiguous areas where the risk exceeds a defined significance threshold. Small or insignificant areas are removed. The remaining regions form the weather footprints: geographical areas where a particular hazard has sufficient probability, severity and exposure to warrant attention.

The system automatically generates daily SVG maps showing both the combined risk from all six hazards and the individual footprints for each hazard. Footprints are numbered, classified by risk, and accompanied by information about their likelihood, impact and severity.

A separate historical-data pipeline downloads archived GEFS forecasts and meteorological observations for calibration and verification. This provides the means to investigate model performance, assess alternative predictors and refine the system using observed outcomes rather than assumptions alone.

The complete production pipeline is automated, from acquiring the latest available forecast through risk calculation, geographic footprint extraction, map generation and static website publication. The public website uses HTML, CSS and SVG, without requiring client-side JavaScript or external application services.

The defining principle of Weather Footprints is that meteorological conditions alone do not determine their significance. The same weather event can have very different consequences depending on where it occurs and what lies in its path.

Rather than presenting conventional weather measurements as its principal output, Weather Footprints attempts to answer a more practical question:

Where is the weather likely to leave a footprint on everyday life, how significant might that footprint be, and how likely is it to occur?

Weather Footprints is not a weather warning service. Footprints show where weather could cause problems, so that you can plan ahead; they are not official warnings and should not be used in place of them. Always follow the warnings of your national weather service: the Met Office in the UK and Met Éireann in Ireland.