As the world's oceans bear most of the warming burden created by climate change, a study found that half of all added heat from the last 44 years entered through the coldest one-quarter of ocean surfaces, according to a newly debuted framework that sorts oceanic regions by temperature.

The Australian mathematicians and climate scientists behind the research, which was published April 11 in Geophysical Research Letters, said their findings show the large influence of cold ocean waters on the global ocean system and also demonstrate that modern climate models are underestimating how hot some parts of the oceans are.

The Earth's average surface temperature has risen substantially since preindustrial times — by about 1 degree Celsius — because of immense greenhouse gas emissions, but it would be spiking even more quickly if not for the enormous buffer that the ocean provides by absorbing a large fraction of the heat.

Warmer waters, however, are not without consequences, and they can contribute to other forms of climate change, such as ocean acidification, disrupted ecosystems and sea-level rise.

"As the climate warms, we estimate that around 90% of the excess heat in the climate system ends up in the oceans," said Taimoor Sohail, a postdoctoral researcher of mathematics and statistics at the University of New South Wales and the paper's lead author. "This means tracking the heat content of the ocean is critical to making future projections of climate change."

There are a few ways to track heat through the oceans by dividing it into regions. One approach is the fixed-depth framework, which some previous research has used to determine how much heat flows in and out of a region of water down to a certain depth. The framework, however, cannot distinguish between heat added from outside the ocean and heat redistributed from below the chosen depth.

To focus on only added heat, Sohail and his co-authors developed an alternative approach they call the "tracer-percentile framework." It organizes the ocean by a selected property, or "tracer," and divides it into percentiles that can be independently tracked. They chose temperature as the tracer in this study.

"This is a departure from traditional ways of tracking heat content in the ocean," Sohail explained, "for example, tracking heat content in the top 2,000 meters or tracking heat content in the volume of water warmer than 14 degrees Celsius."

The framework was then applied to historical ocean observations made between 1970 and 2014, as well as historical projections for the same period from CMIP6 models, a collection of the latest climate models. The researchers followed how oceanic regions of different temperatures gained and lost heat over time.

The authors determined that of all the heat that entered the oceans in the 44-year span, about half was taken up by the coldest 23% of surface area — primarily the North Atlantic and Pacific waters near the Arctic and the Southern Ocean that surrounds Antarctica. These polar regions also experience the most rapid global warming compared with the rest of the world.

This portion of surface waters is linked to heat-related changes in the coldest 90% of ocean volume, according to the study. These interactions amplify the influence of cold ocean surfaces, and minor changes in sea surface temperature or the rate of heat absorption through the surface "may have profound impacts on the ocean circulation," the authors wrote.

The CMIP6 models matched fairly closely with the historical observations, according to the study, but they also sported a "cooling bias" and underestimated the heat in the 5% to 20% warmest volume of ocean. Inaccurate models can lead to incorrect predictions about the ocean's behavior, Sohail said. 

"This will have knock-on effects on how the models behave in modeling future climate projections," the mathematician said. "Given that the ocean absorbs about 90% of the excess heat in the climate system, getting the distribution of the change in heat content of the ocean is very important."

Sohail plans to apply the new framework in several different ways and is currently using it to conduct a similar analysis on changes in the water cycle. He also plans to substitute temperature for other tracers, such as salinity, oxygen and density, and figure out how to use it in regional applications rather than in only global analyses.

The study, "Fifty year trends in global ocean heat content traced to surface heat fluxes in the sub‐polar ocean," published April 11 in Geophysical Research Letters, was authored by Taimoor Sohail, Damien B. Irving, Jan D. Zika, Ryan M. Holmes and John A. Church, University of New South Wales.