Tomorrow.io's Q3 release lets Gale answer questions about your own assets

Gale answers about your assets

According to Tomorrow.io, Gale now answers questions about a customer's own assets and about the events in the company's feeds, such as a named warehouse, a whole fleet or a specific storm. A user can ask which sites to watch this week. The answer comes back with the named assets underlined, each one linking to everything Gale knows about it.

Because Gale understands asset attributes, answers can branch by industry, asset type and where the exposure sits. Tomorrow.io lists runways and freighter routes, substations and pipelines, berths and gantry cranes, and rail yards and signal rooms as examples. The results are collected in an insights view of the user's own operation, using the same severity colors as the dashboard.

Alerts that see asset attributes, and reports you can share

Alerts can now trigger on an asset's own attributes, not only on the weather around it. Tomorrow.io gives the example of a dispatcher writing "alert me on flights with under 60 minutes of fuel heading into thunderstorms", a rule the company says was not possible before. The same engine covers trucks in high winds and vessels in heavy seas.

Each event carries the asset and its attributes, the window, the location, the rule that triggered it, how likely it is, the actions it calls for, who is notified and who owns it. A new insight operator catches a threshold reached within a duration, and geospatial isolines now carry altitude as well as the surface. A contextual Gale button on a flight, airport, location or storm opens the impact analysis without typing.

Events are designed to be passed on. With the MCP server and Custom Vector Events from the July release, a team's workflows and agents can take hold of an event, follow it as it changes and do the work it calls for.

Users can also start a report from any Gale answer. The insights view becomes the report's first block, and Gale's answer routes into the blocks around it. Once saved and shared, the recipient opens the same report the author built.

Models, data and map additions

The release lists several other changes:

• FOCUS: the regional decision model now runs in Mexico, the Amazon, Chile and Argentina, West Africa and Indonesia. Each region gets a new run every 15 minutes, a 48-hour ensemble and AI calibrated to Tomorrow.io's constellation.
• Bulk Timeline API: timelines for thousands of locations can be requested in a single order and delivered as files to the customer's own bucket or Tomorrow.io's, once or on a schedule.
• Tsunami alerts: alerts from GDACS, the Global Disaster Alert and Coordination System, can be shown as a map overlay.
• Languages: translation now includes Spanish, Thai and Japanese.
• Mobile app: severe weather advisories from around the world appear on the map, with details a tap away.
• Lite Map: saved locations and assets appear on the faster map, along with Tomorrow.io's proprietary layers, not just public models.
• Weather layers: new probability layers, new ECMWF altitudes, layers for expert use and skew-T display, density altitude at the surface and new satellite layers.

Tomorrow.io also describes blocks as small applications that combine a data source, decision logic and a view for one job, such as airport conditions, a lightning countdown or a diversion assessment. Teams describe what they need in AI Studio and Gale assembles it.

The quarter in orbit and in models

Between July 1 and September 30, Tomorrow.io says its microwave sounders recorded 28,437,148 scans of the atmosphere. On October 1, after the quarter closed, two more sounders, Tomorrow-S12 and S13, launched aboard SpaceX's Transporter-18.

The company says its atmospheric state model Sky 1, introduced on August 5, draws a fresh global picture of the atmosphere every hour, compared with four times a day from the global centers. RE-FOCUS, introduced on September 28, is an AI model that rebuilds past weather at storm scale. For US storms, it redraws a single event on a three-kilometre grid, hour by hour, anchored to station readings and storm reports, and runs it as 20 versions of the storm to show the most likely gust and the spread around it. Tomorrow.io says it does not replace anyone's risk model but gives that model a sharper history.

What it means for users

For operations teams, the release moves Tomorrow.io's platform from general weather data toward answers and triggers tied to specific sites, vehicles and flights. Attribute-based alerts and events that agents can pick up are aimed at getting a decision made in time, while shareable reports let the results reach colleagues. Tomorrow.io says a separate update for its aviation product, Altitude, is coming next.