As greenhouse gases accumulate in Earth’s atmosphere, the planet holds on to heat that would otherwise dissipate into space. The bulk of that extra warmth is being absorbed by the ocean, and researchers have now turned to an unlikely data source—earthquakes—to study how quickly seawater is warming. Their measurements reveal that the Indian Ocean is heating up by roughly 0.044 K per decade. That’s significantly faster than the rate measured by an array of autonomous floats, the authors reported.
A Climatic Pacemaker
Over 90% of the energy trapped by greenhouse gases ends up warming the ocean. As a result, the global ocean is somewhat of a climatic pacemaker, said Jörn Callies, an oceanographer at the California Institute of Technology in Pasadena and a coauthor of the new study. “The ocean plays an important role in the climate system.”
But getting a handle on how rapidly the ocean is warming has been, until very recently, a bit of a slog. “You’d lower an instrument over the side [of a ship] and take the temperature,” said Callies. That laborious process resulted in spotty and biased coverage because mariners had little interest in sailing the notoriously inclement Southern Ocean, for example.
However, that all changed in 2000 when the first Argo autonomous float was deployed. Each of these roughly meter-long robotic instruments takes repeated conductivity, temperature, and pressure readings of the ocean as it moves up and down the water column. Roughly 4,000 Argo floats now roam the world’s ocean, drifting with the currents.
Going Deeper Than Argo
But a limitation of Argo floats is that they go no deeper than about 2,000 meters. (A program known as Deep Argo, currently being piloted, will send floats to a depth of 6,000 meters.) To get a better sense of how the ocean—including its deeper reaches—is warming, Callies and his colleagues brushed off an idea proposed more than 40 years ago: measure changes in the travel times of sound waves propagating through the water.
Physics dictates that sound waves travel faster through warmer water than colder water. But researchers in the 1990s discovered that generating sound waves using loudspeakers underwater, for instance, comes with its own difficulties, said Bruce Cornuelle, a physical oceanographer at the Scripps Institution of Oceanography in La Jolla, Calif., not involved in the research. “Creating sound sources is expensive, and there were a lot of questions about what it would do to marine mammals.”
Callies and his colleagues opted to turn to a natural source for sound waves. “We’re using earthquakes to generate the sound,” said Callies. When seismic waves shake the seafloor, they generate sound waves in the ocean. Those waves travel at roughly 1.5 kilometers per second—far slower than primary (P) and secondary (S) seismic waves—and they’re converted back into seismic waves when they hit the seafloor again.
Looking for Twins
Callies and his collaborators focused on the region near the Indonesian island of Sumatra, one of the world’s most earthquake prone sites. This area is special, said Wenbo Wu, a seismologist at the California Institute of Technology and lead author of the new study. “The subduction continuously slips there.” As the same section of fault slips repeatedly over time, it generates earthquakes with nearly identical waveforms. These “repeating earthquakes” can be separated in time by hours, days, or years.
Using an earthquake catalog compiled by the International Seismological Centre, the scientists isolated 2,047 pairs of repeating earthquakes that occurred near Sumatra from 2004 to 2016. Finding these repeaters made this research possible, said Callies. The seismic waves of repeating earthquakes originate in the same location, so any changes in travel time can be pegged to changes in ocean temperature, he said. “If we hadn’t used repeaters, the uncertainty in the earthquake location would have swamped any ocean signals.”
For each pair of repeating earthquakes, the researchers calculated the difference in the waves’ travel times between their origin in Sumatra and a seismic receiver on Diego Garcia, an atoll roughly 3,000 kilometers away and part of the British Indian Ocean Territory.
The team found that the travel time differences—typically a few tenths of a second—tended to increase with time. That’s the signature of a warming ocean.
A Pronounced Warming Trend
The researchers calculated that on average, the Indian Ocean had warmed by roughly 0.044 K per decade between 2004 and 2016, they reported last month in Science. That’s about 70% more than the Argo-derived rate of 0.026 K per decade. That discrepancy is somewhat expected, said Callies, because the two data sets are probing different parts of the water column. (Reassuringly, both data sets exhibit fluctuations in water temperature with the same periods—6 and 12 months—that are most likely seasonally driven, the team showed.)
These results are intriguing, but there’s more to investigate, said Scripps’s Cornuelle. For instance, understanding how much of the travel time differences stems from true changes in temperature versus small shifts in the earthquakes’ locations is important, he said.
In the future, it’d be beneficial to use underwater receivers—hydrophones—to detect the traveling sound waves directly, said Callies, rather than relying on them being converted back into seismic waves and picked up by on-land seismic sensors. That’d make it possible to collect measurements from even smaller—and therefore more numerous—earthquakes.
There’s also the option of looking at sound waves of different frequencies, said Callies, to study temperature changes at various water depths. “As you increase the frequency, you’re sensitive to different parts of the water column.” This investigation focused on frequencies between 1.5 and 2.5 hertz, but anything between 1.0 and 10.0 hertz should be possible, the researchers propose. “That’s stuff of the future,” said Callies.
—Katherine Kornei (@KatherineKornei), Science Writer