To investigate ocean currents, scientists usually release GPS-tagged buoys and record their velocities to reconstruct the currents that carry them. This recorded data can also be used to identify areas in which water rises up from below the surface or sinks beneath it (phenomena known as “divergences”).
By accurately predicting currents and identifying divergences, experts can more precisely forecast weather, measure energy transfer rates in the ocean, or estimate how oil will spread after a spill.
Until recently, scientists have largely used a machine-learning technique known as a “Gaussian process” to estimate currents and pinpoint divergences.
Although this method is able to make predictions even in cases when data is sparse, it frequently starts from assumptions that are physically inaccurate, such as that the latitude and longitude components of a current are unrelated. For instance, this model assumes that a current’s divergence and its vorticity (a whirling motion of the water) operate on the same magnitude and length scales –an assumption that ocean scientists know is not true.
To overcome these limitations, a team of computer scientists and oceanographers led by the Massachusetts Institute of Technology (MIT) has recently developed a new model which incorporates knowledge from fluid dynamics to better reflect the physics structuring ocean currents.
Using what is known as a “Helmholtz decomposition,” the new method models ocean currents by breaking them into a vorticity component. This captures the water’s whirling motion, and a divergence component, which captures water sinking or rising.
Although this technique is more computationally intensive, its additional costs will be relatively small. Furthermore, it would help scientists make more accurate estimates from buoy data, enabling them to effectively monitor the transportation of biomass, carbon, plastics, nutrients, and oil in the ocean, and to better understand and track the effects on climate change on currents.
“Our method captures the physical assumptions more appropriately and more accurately. In this case, we know a lot of the physics already. We are giving the model a little bit of that information so it can focus on learning the things that are important to us, like what are the currents away from the buoys, or what is this divergence and where is it happening,” explained senior author Tamara Broderick, an associate professor of Computer Science at MIT.
By evaluating this new model using both synthetic and real ocean buoy data from the Gulf of Mexico, the researchers showed that their method was more accurate in predicting currents and identifying divergences than both the standard Gaussian process and another machine-learning approach which used a neural network.
For instance, in a simulation including a vortex adjacent to an ocean current, the new technique correctly predicted no divergence, while both other methods predicted a divergence with very high confidence. A pre-print of the study can be found here.
In future research, the scientists plan to incorporate a time element into their model (since ocean currents can vary over both space and time), and better capture how noise – such as that produced by wind – can impact the data.
“Our hope is to take this noisily observed field of velocities from the buoys, and then say what is the actual divergence and actual vorticity, and predict away from those buoys, and we think that our new technique will be helpful for this,” Broderick concluded.
Studying ocean currents is crucial for many reasons. They play a significant role in controlling the climate, determining marine life distribution, shaping human activity, and more. Here’s a breakdown of why we study them:
Ocean currents distribute heat around the planet, helping to regulate climate. For example, the Gulf Stream carries warm water from the Gulf of Mexico up to the North Atlantic, moderating the climate in Western Europe. Understanding these currents allows scientists to make accurate climate predictions and models.
Ocean currents impact the distribution of nutrients, which in turn affects where marine organisms can live and thrive. Currents can carry nutrient-rich cold water to the surface in a process called upwelling, supporting biodiversity hotspots. By studying ocean currents, we gain insights into marine ecosystems and the factors influencing their health and biodiversity.
For centuries, mariners have used ocean currents to help in navigation. Understanding ocean currents can lead to more efficient shipping routes, saving time and fuel.
When spills or leaks occur (like oil spills), ocean currents help determine where the pollution will spread. By modeling these currents, we can predict the movements of such pollutants and devise better clean-up strategies.
Many commercial fish species follow particular ocean currents during their life cycles. Knowing these patterns helps in managing and protecting these fisheries.
Ocean currents play a role in weather systems. For example, El Niño, a warm ocean current, affects weather patterns around the globe. Understanding these currents can contribute to more accurate weather forecasting.
Ocean currents are expected to change with climate change, affecting global weather patterns, sea levels, and marine life. Studying them helps us understand the impacts of climate change and guide mitigation strategies.
Currents also impact the dispersion of artifacts from shipwrecks, offering clues into historical events and lost civilizations.
Overall, understanding ocean currents is crucial to a wide array of fields, including climate science, biology, navigation, conservation, and more.