Tomorrow.io has revealed the planned instrument architecture for DeepSky, its next-generation weather satellite constellation, highlighting the role of Earth observation in the development of commercial forecasting services.

The suite combines precipitation radar, hyperspectral microwave sounding, infrared and visible imaging, GNSS radio occultation and GNSS reflectometry. The company intends to bring these complementary measurements together to improve the atmospheric data available to numerical and AI weather models.

The announcement addresses a significant challenge: forecasting systems depend on the observations used to describe current conditions. Tomorrow.io is targeting gaps over oceans and other areas poorly served by ground-based weather radar with the aim of supporting earlier warnings and better decisions.

Read Tomorrow.io’s announcement.

Connecting observations to commercial decisions

For the remote sensing sector, DeepSky offers a useful example of the relationship between instrument design and the services built from its data. The choice of what to measure needs to follow the information customers require and the decisions they need to make.

An agricultural adviser needs information that supports decisions about field operations. A maritime operator needs forecasts that can inform routing and scheduling. Infrastructure owners need evidence they can use to prepare for disruptive weather. Each application places different demands on coverage, timeliness, reliability and the communication of uncertainty.

This creates potential roles for businesses across the EO supply chain. Alongside the collection of observations there is work in checking data quality, integrating measurements into models and translating forecasts into information that fits a customer’s existing systems.

Measuring the value delivered

DeepSky’s instrument announcement describes a planned capability. It does not establish that the constellation is operational or that its proposed benefits have been demonstrated in customer use.

The next questions are therefore practical: how well will the observations perform, how quickly will they reach users and what improvement will they deliver when added to existing forecasting systems? Independent validation and evidence from operational use will be important in answering them.

The broader lesson for commercial EO is clear. Investment in sensors and investment in analytics need to develop together with performance assessed against the decisions the resulting service is intended to support.

Further information: DeepSky programme.