Fathym Forecaster
Harnessing the power of machine learning to create safer roads.
The Fathym Forecaster combines the world’s best weather forecasts with statistics-based, machine-learning techniques to tackle the largest datasets, including road weather.
Transforming prediction
The Fathym Forecaster is a robust, feature-rich API that offers a powerful suite of weather forecasting, as well as open-source data visualization tools.
The Fathym Forecaster offers developers comprehensive weather forecasting capabilities over freely chosen locations and routes across the globe, with any combination of variables.
In addition to the wide array of weather conditions, the API delivers a unique suite of highly specialized forecast variables derived through statistically based machine learning models. These derived variables include road temperature, road state, road condition, and a delay risk factor that computes the weather’s impact on destination arrival estimates.
The Fathym Forecaster provides forecasts out to 120 hours globally. The API also offers 90-day historical queries.
All forecasts are time-interpolated to the nearest second, rather than the lag of hourly updates offered by other forecast APIs. Additionally, map tile output is compatible with all industry standard map overlay APIs such as Azure Maps, Google Maps and OpenStreetMap.
The Fathym Forecaster is designed to ingest any spatial temporal datasets, including additional weather models to provide enhanced forecast granularity as needed.
See the Fathym Forecaster in Action
This data application was built with the Fathym Framework and serves as an example of how the Fathym Forecaster API can pull weather and road condition forecasts for routes globally. All data visualizations used in this application are open source and easily customizable via Fathym’s low-code tools.
Scalable Forecasting
Potential use cases include, but are not limited to, individual point forecasts through time, as well as individual forecasts for large sets of locations, such as the locations of all electric charging stations in the United States.
Dynamic routing features can extend forecast capabilities to user-selected transportation routes, supporting many potential applications targeting commercial transportation and travel where road and surface condition predictions impact planning and logistics.
You can get started for free with our Hobby plan. Fathym also offers Starter, Pro and Enterprise tiers.
Weather and Road Weather Variables
Weather Variables
- Ambient temperature
- Precipitation amount
- Precipitation type
- Precipitation rate
- Wind direction
- Wind speed
- Wind gust
- Barometric pressure
- Relative humidity
- Dewpoint
- Irradiance
- Snow depth
Geospatial Capabilities
- Point forecast
- Multiple point forecast
- Route forecast
- Alternative routes forecast
Derived Variables
- Surface/Road conditions
- Road temperature
- Delay risk / ETA
Forecast Coverage
- Global
Trusted Data
Fathym’s surface weather variables are each carefully and continuously validated via third party and independent sources, such as roadside weather stations with infrared cameras pointed at the road and live webcam images of roads in differing weather conditions to verify predictions. This validation constantly approves the quality of the model, leading to ever increasing accuracy and granularity of data.
Scalable Architecture
Fathym uses a containerized architecture to orchestrate a constellation of programs that work together to collect and aggregate all incoming data, generate forecast outputs and perform constant online validation. The Forecaster’s simple, elegant design ensures top-line performance and reliability.


Surface Weather Forecasts
Unlike typical weather forecasts, surface weather forecasts offer information about how precipitation is going to affect pavement surfaces, roads and soil.
Augment Solutions with Sensors
Environmental monitoring stations can provide real-time observational insights to complement weather forecasts.