GPUs Help Monitor Rising Sea Levels with Pinpoint Accuracy

Editor’s note: This is one in a series of profiles of five finalists for NVIDIA’s 2016 Global Impact Award, which provides $150,000 to researchers using NVIDIA technology for groundbreaking work that addresses social, humanitarian and environmental problems.

While rising sea levels due to climate change have the potential to directly impact a quarter of the world’s population, accurately measuring this metric can be a vexing challenge for climate scientists.

Sea levels are traditionally measured by determining water levels with respect to marks on land. The problem: Parts of the earth’s crust move, too — a result of tectonics. So, while a gauge may report that sea level is constant, it could be because movements of the continent are botching the data.

Thomas Hobiger and fellow researchers at Sweden’s Chalmers University of Technology are tackling this problem with GPS receivers and parallel computing. Instead of measuring water levels against the relative reference frame of land, they’re using reflected signals from a static source: global navigation satellite system reflectometry, known as GNSS-R.

Schematics of the data flow for a software-defined radio GNSS-R solution. Direct (RHCP) and reflected (LHCP) signals are received, A/D converted and sent over a 1 Gbit Ethernet connection to a host PC, where a Tesla K40 GPU handles signal processing.
Schematics of the data flow for a software-defined radio GNSS-R solution. Direct (RHCP) and reflected (LHCP) signals are received, A/D converted and sent over a 1 Gbit Ethernet connection to a host PC, where a Tesla K40 GPU handles signal processing.

Their work could hardly be more relevant. Two separate research teams revealed last week their findings that sea levels are rising at their fastest pace in nearly 3,000 years, the New York Times reported.

The research of Hobiger and team have placed them among five finalists for NVIDIA’s 2016 Global Impact Award. Each year, we award a $150,000 grant to researchers using NVIDIA technology for groundbreaking work that addresses social, humanitarian and environmental problems.

Hobiger got his inspiration for this work back in 2009, at NVIDIA’s first annual GPU Technology Conference, where developers from around the world gather to share insights on GPU computing. Since then, he’s been using GPUs to process data on snow depth, water levels and atomic clocks.

For the past year and a half, Hobiger’s team has been focusing on the sea level project using GNSS-R. This method improves measurement precision by using GPS receivers along the coastline in combination with reflections of GPS signals that bounce off the water’s surface. NVIDIA GPUs then crunch those data signals to compute the water level in real time.

Hobiger chose the reflectometry technique for its centimeter-level accuracy but needed greater back-end processing power to make the solution practical. With NVIDIA GPUs and the CUDA programming model, the GNSS-R solution has become both inexpensive and efficient.

Results from a 36-hour test run last March. Sea-level heights from the software receiver (red) are plotted with measurement in a close-by tide gauge station (blue). Comparing the two results yields an RMS of less than 9 mm and a correlation higher than .995.
Results from a 60-hour test run last March. Sea-level heights from the software receiver (red) are plotted with measurement in a close-by tide gauge station (blue). Comparing the two results yields an RMS of less than 9 mm and a correlation higher than .995.

“With hardware solutions, it’s difficult to implement, takes time and is also very inflexible. We saw that if we go to signal processing in software, our opportunities grow infinitely,” said Hobiger. “And that’s where the GPU and, in particular, CUDA came in.”

The team uses CUDA’s cuFFT libraries to process their data. As a result, Hobiger notes that their solution is much cheaper than similar hardware methods used by other research institutes.

Data from the reflectometry systems comes in a constant stream up to 800 megabits per second. Using the parallel processing power of NVIDIA Tesla and GeForce GPUs, Hobiger’s team can now keep pace with the flood of information.

“Without the use of GPUs, we would not have been able to process all our signals in real time,” said Hobiger.

The software used by Hobiger’s team is open source, and can be used on a desktop in tandem with sea-level instruments in other regions.

“Understanding our planet and how it might evolve is important,” said Hobiger. “Raising awareness of earth science should be in everyone’s interest, because it’s our planet.”

The winner of the 2016 Global Impact Award will be announced at the GPU Technology Conference, April 4-7, in Silicon Valley.

More Global Impact Award 2016 nominees

How GPUs Help Eye Surgeons See 20/20 in the Operating Room

How Haiti’s Earthquake Inspired New Ways to Map Structural Safety Using GPUs

How Imperial College Uses GPUs to Spot Brain Damage

Check out the work of last year’s NVIDIA Global Impact Award winner.

Similar Stories