First page Previous page 6 Next page Last page

Early Online Release

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

Oceanography | Early Online Release

to reproduce midwater transect capabilities that had long been

refined using ROVs. The platform is able to travel faster than an

ROV and is quieter (Reisenbichler et al., 2016; Robison et al.,

2017). Other AUVs along these lines are rapidly becoming more

common and trending smaller in size for both water column and

seafloor observations. Just as robots have improved our capacity

for biogeochemical sensing, ROVs and AUVs now offer another

suite of platforms and tools for probing the “large scale geophysi

cal experiment” that Revelle and Suess foretold.

THE DATA DELUGE

Obtaining high-resolution underwater video observations and

conducting in situ experiments have proven to be effective means

for documenting ecosystem changes that are occurring over

time. For example, in Monterey Bay, changes in oxygen in the

water column are linked to observed changes in animal behavior,

which in turn has significant implications for food web dynamics

(e.g., Figure 5; Robison et al., 2017). A key enabling technology

that has made this observation possible is the Video Annotation

and Reference System (VARS; Schlining and Stout, 2006; VARS

Overview). VARS provides the means to expertly identify what is

seen in underwater imagery—a process known as annotation—

and merge it with concurrent measurements of relevant physi

cal and chemical parameters. The result is a searchable database

that contains geolocated quantitative sightings of particular ani

mals cross referenced with the environmental conditions under

which they were observed. VARS is an open-source application

that has been adopted by a number groups, including Australia’s

Commonwealth Science and Industrial Research Organization

(CSIRO), Oregon State University, the University of Hawai‘i School

of Ocean and Earth Science and Technology (SOEST), and the

National Oceanic and Atmospheric Administration (NOAA). At

MBARI, to date, VARS has grown to include nearly 29,000 hours

of underwater imagery from which almost 11 million observations

of over 4,400 unique “concepts” (e.g., animals, debris, geologic for

mations) have been cataloged. Nearly 600 peer-reviewed publica

tions and over 300 new species have been described drawing from

that archive. The Deep-Sea Guide offers a publicly accessible por

tal for accessing a portion of VARS content.

With the ever-growing collection of imagery from a multitude

FIGURE 3. Evolution of platforms used for midwater research and time series studies at MBARI. (a) ROV Ventana’s first launch in 1988 from R/V Point Lobos

concurrent measurements of temperature, oxygen, salinity, and

other ocean variables. The addition of robotic sensors and sam

plers to ROVs also made it possible to collect specimens and con

duct unique in situ experiments. A recent example of the utility of

what ROVs can enable scientifically is particularly well illustrated

in the detailed description of a deep-sea animal new to science that

for years was known only as the “mystery mollusc” (Robison and

Haddock, 2024; Figure 4). Similarly, ROVs have also proven to be

valuable tools for evaluating the impacts of rising levels of CO2 on

ocean biology and chemistry both in the water column and on the

seafloor (e.g., Barry et al., 2017; Brewer et al., 2017; Robison et al.,

2017). In today’s world, ROVs are integral to ocean exploration

and are proliferating. The technology continues to evolve rapidly,

making the platforms more capable, accessible, and affordable.

Operating ROVs is less costly and logistically less complex than

crewed submersibles, but it still requires a surface support ship

and skilled crew. In a step toward reducing the dependency on

crewed ships, AUVs are being modified to conduct similar sur

veys. For example, the i2MAP vehicle built at MBARI (Figure 3c)

carries imaging and acoustic systems along with other sensors

FIGURE 4. This animal was long known as “the mystery mollusc.” Years of

observations, experimentation, and specimen collections using remotely

operated vehicles (ROVs) ultimately led to its formal description as

Bathydevius caudactylus, an entirely new bathypelagic nudibranch genus