June 2025

June 2025 | Oceanography

33

tidal forcing as the initialization states. One generator, GNT→T(·),

translated from the non-tidal to the tidal domain, and the other

generator, GT→NT(·), translated from the tidal to non-tidal

domain. To address the issue of the chaotic, turbulent nature of

the ocean, we considered the simulations to be unpaired (i.e., not

a direct translation between one state and the other). Instead,

the GAN used “cycle-consistency loss,” the mean-squared differ­

ence between the original data sample and the doubly translated

data (Zhu et al., 2017). The cycle-consistency loss was combined

with the traditional GAN losses (i.e., the difference between the

generator and the discriminator output) to train the networks.

The Atlantic Ocean was used as a test-case region; one week of

hourly HYCOM data was split into 90% training data and 10%

validation data.

The GAN results retained the general structure of the tem­

perature and salinity profiles from HYCOM while adding or

removing a semidiurnal tide (Figure 7). The GAN performed

well in the relatively quiescent region of the tropical mid-​

Atlantic (Figure 7b). There, periodic signatures in HYCOM with

tides matched the periodicity of the outputs of GNT→T(·). The

semidiurnal signature was also removed in GT→NT(·) to match

the non-tidally forced HYCOM. It was more difficult to separate

the tidal structure from mesoscale variability in more energetic

regions, such as near the Gulf Stream (Figure 7c,d). For example,

just north of the Gulf Stream (Figure 7c), the GNT→T(·) repro­

duced semidiurnal periodicity of the tidally forced HYCOM,

but there was also periodicity in the nontidal fields. In the Gulf

Stream extension (Figure 7d), the GAN imposed a periodicity to

make the sample like other tidally forced results, but this was a

region dominated by mesoscale variability.

Because the HYCOM output used to train the GAN was sam­

pled from the same region of the globe during the same time of

year, no two samples were completely independent. This intro­

duces the risk of overfitting. Using unpaired data made the

model more robust to overfitting but did not remove the risk

entirely. Additionally, the sound speed structure had a persistent

offset of about 5 m s–1 greater in the GAN-generated results than

the original HYCOM simulations (not shown). Thus, although

this work provides a good starting point, further work will help

revise this approach.

FIGURE 7. Temporal out­

puts of the deep learn­

ing GAN model at the

locations mapped in (a).

For each panel, the first

column shows the non-

tidal (NT) HYCOM results

(Exp 19.2); the second

column shows the NT

results translated into

the tidal domain using

the GAN model; the

third column shows the

tidal (T) HYCOM results

(Exp 19.0); and the fourth

column shows the T

results translated into

the NT domain using a

GAN model. From top

to bottom, rows in (b–d)

show water tempera­

ture, salinity, eastward

velocity, and northward

velocity, respectively.