No audio available for this content.
In recent years, scientists have shown that detecting changes in navigation signals from GPS and Galileo after they bounce off Earth’s surface (GNSS reflectometry, or GNSS-R) can deliver valuable information on sea ice. Now research drawing on data from Spire Global has enabled the generation of Arctic-wide sea ice maps, marking a major step forward for the emerging technique.
Spire Global‘s sea ice freeboard maps use data captured by Spire’s GNSS-reflectometry multipurpose listening constellation.
The research — enabled by the Third Party Missions (TPM) programme of the European Space Agency (ESA) — suggests that harnessing reflected navigation signals could become an important complement to established ice-monitoring altimetry missions.
The study leveraged Spire’s GNSS-R data to retrieve sea ice freeboard measurements across an entire winter season. The results show strong alignment with established altimetry datasets, including the ESA’s CryoSat mission, validating the complementary role of commercial satellite data alongside government missions.
The study was led by Felix Müller at the Technical University of Munich (DGFI-TUM) and Robert Ricker at the Norwegian Research Centre, experts in GNSS-R.
“The primary purpose of signals emitted from GNSS is to fix the location of a device at any point on Earth,” Müller explained. “However, when these signals bounce off Earth’s surface, their properties change. By analyzing these changes, we can infer information about the characteristics of Earth’s surface.”
“Previous research has shown that this technique works well experimentally,” Ricker added. “Using the Spire constellation, we aimed to demonstrate whether it would hold up on a larger scale by generating an Arctic-wide map of sea ice freeboard, which is a measure of how far ice protrudes above the waterline.”
Spire’s GNSS-R constellation
Spire’s constellation was first used to sample the atmosphere for weather forecasting. Then scientists began exploring other applications. Spire started collecting reflected signals arriving at shallow angles using a technique called grazing-angle GNSS-R. This method is particularly well suited for ice monitoring.
The research team analyzed data detected over the Arctic Ocean and surrounding seas between October 2023 and July 2024. The data was obtained via the TPM program, through which ESA disseminates data from a range of commercial and institutional partners on a free basis for research and development purposes.
The team focused on one of the most critical challenges in sea ice altimetry: reliably identifying narrow openings in the ice pack, known as leads. These openings are reference points for determining sea surface height and, ultimately, sea ice freeboard.
In turn, sea ice freeboard can be used to infer sea ice thickness — an essential parameter for tracking climate change, estimating sea level, and modeling ocean and weather patterns.
Classifying surface properties
“In the initial phase of the project, we used two complementary methods to identify surface properties based on GNSS-R data, with the aim of identifying leads,” Müller said.
The first — known as the adaptive threshold technique — involved measuring the power of the reflected navigation signal to classify surface type as either water or ice. This method allows rapid processing of the entire GNSS-R dataset, while remaining robust to changes in signal conditions.
The second method — known as unsupervised clustering — offers a more complex approach to classifying surface conditions. In addition to signal power, it considers multiple other signal features that tease out more nuanced information on surface type, including identifying thin or refrozen ice.
Both methods were compared with co-located CryoSat surface-type classifications and Sentinel-1 imagery, confirming that the GNSS-R classifications were largely comparable against conventional satellite products.
Mapping sea ice freeboard
“Building on this classification work, we then took the research to the next step by producing Arctic-wide sea ice freeboard maps from GNSS-R data,” Ricker said.
The team corrected ice surface height measurements generated from GNSS-R data for tidal variations, sea surface height, and atmospheric delays, which is standard practice in altimetry. A refined algorithm then identified where leads in the ice were likely to occur, with the lowest points in these areas revealing estimated sea surface height. Sea surface height estimates were then subtracted from ice surface heights to retrieve freeboard. Using this approach, monthly gridded freeboard products were generated for the full winter season.
The team reported that the GNSS-R datasets showed strong agreement with CryoSat freeboard datasets across much of the Arctic, confirming that GNSS-R can reproduce large-scale patterns previously observed by dedicated altimetry missions. Independent validation against upward-looking sonar measurements in the Beaufort Sea further supported the accuracy of the retrieved freeboard values.
However, as expected, the GNSS-R estimates became less reliable during spring, when surface melt alters reflection characteristics. This limitation is consistent with earlier GNSS-R and radar altimetry studies and remains an active area of research.
The contribution of commercial data
While GNSS signals have long been used for positioning, this research highlights how reflected signal analysis can extend their value into large-scale Earth observation applications, delivering persistent coverage independent of sunlight or weather conditions, said Theresa Condor, Spire Global CEO.
“Advances in miniaturization, digital signal processing, and machine learning have fundamentally changed what’s possible in RF sensing,” Condor said. “Commercial constellations can now deliver persistent, high-quality RF data that complements traditional government systems with greater flexibility and cost efficiency.
“As environmental monitoring requirements intensify, we’re seeing agencies increasingly integrate commercially sourced RF datasets into operational architectures, reflecting the continued maturation of this market and the growing role of commercial infrastructure in government missions.”
“By producing analysis-ready gridded datasets, this work marks an important milestone in the progress of grazing angle GNSS-R from an experimental method to a reliable technique for mapping Arctic sea ice freeboard at scale,” said Matthieu Talpe, Remote Sensing Product Engineer, Spire Global. “In doing so, it strengthens the case for the grazing angle GNSS-R technique employed by the Spire constellation as a valuable complement to existing ESA and partner missions, helping to close observational gaps in one of Earth’s most rapidly changing regions.”
