Early Online Release

Oceanography in the Age of Intelligent Robots and a Changing Climate By Chris Scholin

Oceanography | Early Online Release

calculus. Low-resolution, surface-vessel-based maps can now be

used to guide higher-resolution AUV-based surveys. AUVs can

run in close proximity to the ground compared to a vessel at the

sea surface, thus providing much more detail on what lies below.

The combination of nested surface vessel-AUV-ROV surveys has

greatly aided our understanding of underwater landscapes and

how they evolve, and this approach now informs choices on what

locations to observe more closely and repeatedly to improve the

odds of making new discoveries (e.g., Caress et al., 2008, 2012;

Paull et al., 2010; Paduan et al., 2018, Figure 8).

Even highly detailed bathymetric surveys fail to reveal much

about the animals that inhabit the seafloor. With relatively few

exceptions, most life on the seabed is sub-meter scaled and often

transparent to acoustic energy. By combining high-resolution laser

and optical imagery with acoustic mapping, a truly astounding

view of the seafloor emerges (Figure 9). The systems for acquir­

ing that information can be deployed on ROVs (e.g., Caress et al.,

2025) and are extendable to AUVs, greatly expanding the area that

can be surveyed in detail. Processing the imagery collected using

machine learning techniques also holds promise for significantly

speeding up quantitative assessments of specific animals or other

features of interest even while the vehicle is underway. Further

study of the famed octopus garden provides a stunning example of

what is possible when combining different modes of seafloor visu­

alizations to inform targeted studies that not long ago would have

seemed a pipe dream (Barry et al., 2023; Figure 9). Similar studies

of deep-sea coral and sponge communities found serendipitously

at Sur Ridge and elsewhere paint a similar picture (Girard et al.,

2024; Figure 10). These discoveries highlight what is made pos­

sible by using a combination of hybrid human-machine and fully

autonomous systems for visualizing the seafloor.

Despite that progress, the vast majority of the seabed has never

been mapped at scales needed to reveal underwater landscapes

in detail. Satellite altimetry-derived maps provide ~5 km grid

resolution estimates of seafloor depth for the entire ocean bottom

using gravity anomalies (W.H.F. Smith and Sandwell, 1997), but

those maps provide only a coarse perspective on what lies below,

much like a person viewing a large terrestrial mountain range, deep

valley, or vast plain from a great distance. High-resolution maps

of the seafloor acquired using surface vessel-mounted multibeam

sonar varies linearly with water column depth, typically on the order

of 2 m at 100 m depth to 100 m at 5,000 m depth (Mayer, 2006), but

even those maps currently cover only ~26% of the ocean bottom.

Visualizing deep-sea biological communities requires much higher

resolution, ideally centimeter or even millimeter scale, as shown in

Figure 9. In other words, much of what lies below has never been

seen by human eyes. Although the technology for doing so is avail­

able, actually accomplishing that goal globally is an enormous task

and not likely to come to fruition anytime soon. Once again, robots

offer a path forward for tackling that challenge because they can

work when and where people cannot, dare not, or just prefer to

avoid for many practical and logistical reasons.

A combination of crewed and uncrewed surface and subsurface

vessels are now actively engaged in mapping the entirety of the sea­

floor as a contribution to the Seabed 2030 initiative. Seabed 2030 is

a collaborative project sponsored by the Nippon Foundation and

the General Bathymetric Chart of the Ocean (GEBCO) that aims to

assemble all available bathymetric data into a single, freely accessi­

ble map for the benefit of all. Like the global fleet of profiling floats

returning data on the vital signs of the world’s ocean, Seabed 2030

is a great example of what can be accomplished through public-​

private partnerships, international cooperation, and data sharing to

grow our understanding of seafloor bathymetry. Given the task at

hand and its relevance to society, it speaks to the age-old adage that

“necessity is the mother of invention.” Developing new means for

comprehensively mapping the seafloor is ripe for innovation, fol­

lowing in the footsteps of developing and deploying platforms and

sensors for assessing ocean biogeochemistry on a global scale.

FIGURE 8. Animation demon­

strates the combined use of ship,

AUV, and ROV-based surveys to

obtain

high-resolution

seafloor

bathymetry and imagery at Sur

Ridge within the Monterey Bay

National Marine Sanctuary (Seeing

Sur Ridge). © 2023 MBARI