TRITON: listening to navigation signals reflected by the ocean

Taiwan’s TRITON satellite, listed as FORMOSAT-7R in the archive, launched in October 2023 with a domestically developed Global Navigation Satellite System reflectometry — open definition receiver. Its main observing idea differs from the atmospheric occultation used by COSMIC. TRITON receives navigation signals after they reflect from Earth’s surface, with ocean observations used to investigate sea-surface winds. The transmitting navigation system supplies the illumination; TRITON is not sending its own radar pulse toward the water.

A smooth surface and a rough surface scatter incoming radio energy differently. Wind helps shape small waves and ocean roughness, so the reflected signal contains information relevant to estimating wind speed. The receiver does not contain a tiny anemometer touching the ocean. A retrieval — open definition method connects the measured signal with a physical quantity using geometry, calibration and assumptions about the surface.

In radio occultation — open definition, navigation signals pass through atmospheric layers near Earth’s apparent edge and their propagation changes provide profile information. In ocean reflectometry, the useful path includes a bounce from the surface. Both methods reuse navigation transmissions, but they are not interchangeable. The same source of radio waves can support different experiments when the receiving geometry and analysis change.

TRITON’s development also included environmental and functional tests and rehearsals for ground operations. NIAR’s 2022 account describes these preparations before launch. Passing a ground test establishes performance under the tested conditions; it does not establish the accuracy of every later ocean-wind retrieval. Comparing returned products with suitable independent observations is part of demonstrating the scientific measurement, not an optional consequence of having a successful launch.

Two ways to reuse navigation signals

Imagine three fictional matched wind estimates from a satellite: 8, 10 and 12 meters per second. Nearby reference measurements for the same times and places are 7, 9 and 11. Each satellite estimate is one meter per second higher, giving a mean difference — open definition of +1 meter per second. This is an illustrative validation exercise, not a reported TRITON result.

A consistent difference suggests investigating a bias, but three comparisons do not prove the same error everywhere. Check whether the instruments sampled comparable locations and times, whether the reference is reliable and whether performance changes across conditions. The goal is a defensible wind product with known limits. TRITON broadens the lesson of the FORMOSAT missions: useful space science comes from matching a measurement method to a question and testing the chain from signal to interpretation.