Marine Geology 463 (2023) 107121

Contents lists available at ScienceDirect

Marine Geology
journal homepage: [Link]/locate/margo

Rising snow line: Ocean acidification and the submergence of seafloor

geomorphic features beneath a rising carbonate compensation depth
Peter T. Harris a, *, Levi Westerveld a, Qianshuo Zhao b, Mark John Costello c
a
GRID-Arendal, P.O. Box 183, N-4802 Arendal, Norway
b
Ocean University of China, Qingdao, People’s Republic of China
c
Nord University, Universitetsalléen 11, 8026 Bodø, Norway

A R T I C L E I N F O A B S T R A C T

Editor: Michele Rebesco Due to burning of fossil fuels, carbon dioxide is being absorbed by the ocean where its chemical conversion to
carbonic acid has already caused the surface ocean to become more acidic than it has been for at least the last 2
Keywords: million years. Global ocean modeling suggests that the carbonate compensation depth (CCD) has already risen by
Carbonate compensation depth nearly 100 m on average since pre-industrial times and will likely rise further by several hundred meters more
Ocean acidification
this century. Potentially millions of square kilometres of ocean floor will undergo a rapid transition in terms of
Seafloor geomorphology
the overlying water chemistry whereby calcareous sediment will become unstable causing the carbonate “snow
Calcareous sediment
Climate change line” to rise.
We carried out a spatial analysis of seafloor geomorphology to assess the area newly submerged below the
rising CCD. We found that shoaling of the CCD since the industrial revolution has submerged 12,432,096 km2 of
ocean floor (3.60% of total ocean area) below the CCD. Further hypothetical shoaling of the CCD by 100 m
increments illustrated that the surface area of seafloor submerged below the CCD has risen by 14% with 300 m of
shoaling, such that 51% of the ocean area will be below the CCD. All categories of geomorphic feature mapped in
one global database intersect the lysocline and will be (or already are) submerged below the CCD with much
regional variation since the rise in CCD depth during the last 150 years varies significantly between different
ocean regions. For seamounts, the highest percentages of increase in area submerged below the CCD occurred in
the Southern Indian Ocean and the South West Atlantic regions (6.3% and 5.9%, respectively). For submarine
canyons we found the South West Atlantic increased from 3.9% in pre-industrial times to 8.0% at the present
time, the highest percentage of canyons found below the CCD in any ocean region.
We also carried out a relative risk assessment for future submergence of ocean floor below the CCD in 17 ocean
regions. In our assessment we assumed that the change in CCD from pre-industrial times to the present is an
indicator of the likelihood and the change in percentage of seafloor submerged below the CCD due to a hypo­
thetical 300 m rise in the CCD is an indicator of the consequences. We found that the western equatorial Atlantic
is at high risk and 9 other Ocean Regions are at moderate risk. Overall, geomorphic features in the Atlantic Ocean
and southern Indian Ocean are at greater risk of impact from a rising CCD than Pacific and other Indian Ocean
regions.
A separate analysis of the Arctic Ocean points to the possible submergence of glacial troughs incised on the
continental shelf within a mid-depth (400–800 m) acidified water mass. We also found that the area of national
Exclusive Economic Zones submerged below the rising CCD exhibits extreme variability; with 300 m of CCD
shoaling we found a > 12% increase in area submerged below the CCD for 23 national EEZs, whereas there was
virtually no change for other countries.

1. Introduction calcium carbonate mineral “calcite” becomes unstable and begins to

dissolve. Its upper limit is the calcite saturation depth (CSD) and its
The “lysocline” is the transition zone in the ocean within which the lower limit is the calcite (carbonate) compensation depth (CCD). Above

* Corresponding author.
E-mail address: [Link]@[Link] (P.T. Harris).

[Link]
Received 9 April 2023; Received in revised form 7 August 2023; Accepted 13 August 2023
Available online 15 August 2023
0025-3227/© 2023 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY license ([Link]
P.T. Harris et al. Marine Geology 463 (2023) 107121

Fig. 1. Map showing the 18 ocean regions included in this study. The 17 regions mapped by Sulpis et al. (2018) were manually edited and clipped in ArcGIS to
stretch all the ocean areas to the respective continents. The Arctic Ocean was included as a separate (18th) region; the Mediterranean Sea is not included.

calcareous minerals with increasing depth (Sulpis et al., 2018).

Western Equatorial Atlantic

Aragonite is another calcium carbonate mineral whose chemistry is

less stable (and hence much more soluble) in seawater than calcite. The
Southern Indian Ocean

East Equatorial Atlantic

Western Indian Ocean

aragonite compensation depth (ACD) is therefore much shallower than

Eastern Indian Ocean
Central Indian Ocean
South West Atlantic

North West Atlantic

South East Atlantic

North East Atlantic

the CCD. Aragonite accounts for about 50% of the calcium carbonate
Philippine Basin

flux in the ocean (mainly the shells of marine snails called pteropods),
essentially equal with the calcite flux (Manno et al., 2007). Sulpis et al.
(2022) suggested that the dissolution of aragonite at the sea floor fa­
vours the preservation of less soluble calcite.
Due mainly to burning of fossil fuels, humans have caused atmo­
spheric carbon dioxide content to rise to levels that have not occurred
naturally over many millions of years. Carbon dioxide is being absorbed
by the ocean where its chemical conversion to carbonic acid has already
caused the surface ocean to become more acidic than it has been for at
least the last 2 million years (Honisch et al., 2009). As a dissolved gas,
CO2 is carried to abyssal depths by the ocean conveyor where acidifi­
Water depth (m)

cation is expected to cause the lysocline to rise (Archer et al., 1998), a

process operating on a multi-century time scale.
Acidification of the oceans is acknowledged as a threat to ocean
health in the United Nations Sustainable Development Goals (SDGs),
Goal 14 to “Conserve and sustainably use the oceans, seas and marine
resources for sustainable development” and SDG indicator 14.3.1:
“Average marine acidity (pH) measured at agreed suite of representative
sampling stations” (United Nations, 2023a). It is also a factor to consider
in the design of any marine protected areas established in the high seas
under the recently adopted UN Ocean Treaty (United Nations, 2023b),
in which protection of deep sea biodiversity is a key goal (eg. Albright
et al., 2023). In the absence of information on deep sea biodiversity,
Fig. 2. Depth of the CCD modelled by Sulpis et al. (2018) showing pre- surrogates (Harris, 2012) such as ocean water properties and seafloor
industrial and present CCD for the 17 ocean regions ranked in order of geomorphic features have been proposed (e.g., Harris and Whiteway,
increasing, pre-industrial CCD. In general, the Pacific is more acidic (shallow 2009; Sayre et al., 2017; Fischer et al., 2019). Here we first review recent
CCD/CSD) than the Atlantic (deep CCD/CSD) but there is variability between
modeling of the CCD and then study how seafloor morphology will affect
ocean regions. Error bars for the present-day CCD (after Sulpis et al., 2018)
the depth of the CCD under future climate change.
are shown.

the CSD ocean water is super-saturated with calcite and seabed sedi­ 1.1. Previous CCD modeling
ments are calcareous rich. Below the CCD seabed sediments contain
little or no carbonate minerals, the carbonate “snow line” (Woosley, In their recent study, Sulpis et al. (2018) used a rate model to derive
2016). Sediments within the lysocline exhibit a decline in content of global benthic calcite dissolution rates for abyssal sediments in 17

2
P.T. Harris et al. Marine Geology 463 (2023) 107121

Table 1
The percentage of the ocean regions’ (Fig. 1) areas exposed below the CCD by hypothetical shoaling of the CCD. The rise in the CCD (m) with estimated error from
Sulpis et al. (2018) was used to estimate the percentage below the pre-industrial CCD and the percentage area presently below the CCD. The percentage areas for +100
m, +200 m and + 300 m were estimated as explained in the methods. Total ocean area follows Sulpis et al. (2018), which excludes the Arctic Ocean and the Med­
iterranean Sea.
Ocean Region Total Area Below CCD Pre- Below CCD Rise of CCD Pre- Below CCD Present Below CCD Present Below CCD Present
km2 industrial times (%) Present (%) industrial to Present (m) +100 m (%) +200 m (%) +300 m (%)

1 Central Equatorial
Pacific 16,580,040 45.70 47.63 20 ± 276 57.14 65.17 71.37
2 Central Indian
Ocean 11,022,170 37.12 38.35 36 ± 261 42.15 45.90 49.08
3 East Equatorial
Atlantic 10,156,820 0.55 0.948 73 ± 161 1.73 3.16 6.05
4 Eastern Equatorial
Pacific 9,440,670 28.06 29.07 19 ± 331 34.52 40.70 47.32
5 Eastern Indian
Ocean 18,149,390 22.23 25.59 80 ± 231 29.93 33.90 37.73
6 North East
Atlantic 13,585,500 6.10 10.65 149 ± 178 13.81 16.91 19.89
7 North Pacific 42,585,460 66.37 67.11 36 ± 293 69.13 71.01 72.76
8 North West
Atlantic 18,237,390 13.20 23.33 316 ± 185 26.39 29.38 32.18
9 Philippine Basin 20,110,530 32.15 32.57 30 ± 316 33.98 35.23 36.39
10 South East
Atlantic 13,512,790 44.15 49.12 133 ± 251 52.61 55.98 59.41
11 South East
Pacific 24,878,730 43.39 47.78 69 ± 308 54.05 59.75 65.06
12 South West
Atlantic 23,764,040 38.25 46.11 227 ± 280 49.59 52.74 55.60
13 South West
Pacific 45,292,370 42.48 45.21 81 ± 287 48.50 51.62 54.50
14 Southern Indian
Ocean 27,405,630 36.27 46.24 216 ± 288 50.05 53.51 56.78
15 Western
Equatorial
Atlantic 9,520,260 22.49 27.11 129 ± 215 30.79 34.75 39.28
16 Western
Equatorial Pacific 23,555,170 49.39 50.42 30 ± 269 53.53 56.66 59.89
17 Western Indian
Ocean 17,104,860 12.31 12.77 18 ± 245 15.92 19.76 23.48
TOTAL (Average) 344,901,850 36.91 40.51 97.8 ± 257 44.13 47.61 50.91

Fig. 3. Conceptual diagram showing two seafloor areas A and B exposed by the rising lysocline. The carbonate content of seafloor sediments decreases within the
lysocline, reaching zero below the CCD (carbonate “snow line”). The present study focuses on areas previously above the CCD but now below it (Exp-A), noting that
areas previously above the CSD but now within the lysocline (Exp-B) will also be effected by acidification of the oceans.

separate ocean regions (Fig. 1). The depth of the CCD in pre-industrial bottom water is formed in locations of downwelling (e.g., adjacent to the
times was estimated to range from as shallow as 3895 ± 377 m in the polar seas) and the CCD is deeper (e.g., Broecker, 1991; Schmitz and
eastern equatorial Pacific to as deep as 5926 ± 155 m in the eastern McCartney, 1993; Johnson, 2008).
equatorial Atlantic (Fig. 2). The depth of the CCD varies with the length Changes in the depth of the CCD from pre-industrial times to its
of time since the bottom water has been exposed to the surface (bottom present position (estimated for 2002) range from a rise (shallowing) of
water “age”). Over time, benthic respiration and oxidation of organic 316 ± 185 m in the North West Atlantic above its pre-industrial depth of
matter add CO2 such that the oldest water masses have the highest ~5400 m, to 18 ± 245 m (essentially no change) in the western Indian
concentrations of CO2 and therefore the shallowest CCD. The youngest Ocean. The average rise in the CCD across all ocean areas was estimated

3
P.T. Harris et al. Marine Geology 463 (2023) 107121

Fig. 4. The changes in percent area of ocean exposed to corrosive bottom waters in 17 different ocean regions in pre-industrial times (− 100 m), at present and
resulting from a hypothetical shoaling of the CCD in three, 100 m increments. The slope of lines gives a relative indication of change in area below the CCD with CCD
shoaling. Estimated error (not shown) for present CCD is as per Table 1.

to be about 98 ± 257 m since pre-industrial times, with considerable sensitivity to these undersaturated conditions, benthic fauna will either
spatial heterogeneity in the modelled shoaling of the CCD (Fig. 2; Sulpis remain and adapt to the changed conditions, or their distribution will
et al., 2018). In general, the greatest change in CCD since post-industrial shallow as they die out in, or otherwise avoid colonizing, areas that they
times occured in areas of downwelling, where the bottom water was cannot tolerate. The literature on the exact response of deep sea fauna to
youngest (Fig. 2) and this was attributed by Sulpis et al. (2018) to the the effects of acidification are mainly speculative due to the lack of data
post-industrial addition of CO2 to the atmosphere. Although the large (Fabry et al., 2008; Widdicombe and Spicer, 2008; Hofmann et al.,
error bars in the estimated CCD depths were acknowledged (Fig. 2 and 2010). The acidification of the abyssal zone is believed to favour
Table 1), it was concluded by Sulpis et al. (2018) that there has been a agglutinated foraminifera over calcifying species, for example. Howev­
general rise in the CCD since pre-industrial times which has been er, studies have focused on carbonate chemistry and its effect of disso­
causing an increased rate of carbonate sediment dissolution (i.e., a rise lution of plankton shells (Milliman et al., 1999). A recent paper by
in the carbonate sediment “snow line”) amounting to 40–100% of the Simon-Lledó et al. (2023) has provided the first evidence that the CCD is
total seafloor dissolution at its most intense locations. a biological boundary located between 4300 and 4800 m depth in the
The thickness of the lysocline (difference in depth between CCD and Clarion-Clipperton zone of the northeast Pacific. The authors found that
CSD) averages around 1081 m over all ocean regions (Fig. 2). However, there was no change in indicators of species diversity across the CCD but
there is considerable spatial variation in estimated lysocline thickness rather that species that build a carbonate shell were absent below the
between ocean regions, ranging from 536 ± 185 m in the Northwest CCD and were replaced by other, non-calcifying species.
Atlantic to 1755 ± 231 m in the Eastern Indian Ocean. Lysocline A key consideration is that the acidification of the deep ocean is
thickness exhibited no significant difference between pre-industrial likely to occur in combination with a decrease in dissolved oxygen,
times and the present day with an average increase of 15.6 m (within decreased POC flux and a weakening of abyssal currents, all attributed to
the error limits for available data; see Sulpis et al., 2018). climate change (Sweetman et al., 2017). To these climate change related
Palaeoceanographers have reported that oceanic acidification events pressures we must add the impacts of pollution (e.g., plastic pollution;
have happened before in the geologic past, the most recent occurring in Galgani et al., 2022) and other human activities such as fishing, cable
association with the Palaeocene-Eocene Thermal Maximum (PETM; laying and seismic surveys among others (Harris, 2020). These cumu­
55.8 million years ago). This global warming event was short-lived lative impacts acting together will have a greater impact on benthic
(~200,000 years) during which temperatures increased by 5–9 ◦ C and biodiversity than any one pressure acting in isolation (Halpern et al.,
the Atlantic CCD shoaled to <2000 m (Zachos et al., 2005). It was also 2008).
marked by the largest deep-sea mass extinction among calcareous There is a popular misconception that carbonate minerals will
benthic foraminifera in the last 93 million years (Thomas, 1998). dissolve and buffer the acidification effect of increased anthropogenic
Although the unprecedented rapid pace of change in the environment CO2. While this is correct over geologic (~1000 year) timescales, there is
during the Anthropocene limits the applicability of the fossil record to a time lag involved that is often overlooked as explained by Archer et al.
inform what we may expect in the coming decades to century (Kump (1998) and Andersson and Mackenzie (2012). Over timescales of human
et al., 2009), the available information indicates that human-induced lifespans, the buffering effect is minimal. Therefore, the recent ~100 m
ocean acidification will impact oceanic ecosystems, although how rise in the CCD since pre-industrial times reported by Sulpis et al. (2018)
significantly remains unknown. is likely to continue this century. Further, the rate of rise of the CCD is
The rising depth of the lysocline will change the chemical environ­ likely to accelerate in the next century as the ocean conveyor delivers
ment of benthic organisms inhabiting depths that were previously above increasingly higher concentrations of dissolved carbon dioxide into the
the lysocline and are now within it as well as organisms that existed deep ocean.
within the lysocline but find themselves below it. Depending on their

4
P.T. Harris et al. Marine Geology 463 (2023) 107121

Fig. 5. Maps showing the changes in area of ocean exposed to

corrosive bottom waters in 17 different ocean regions, showing
area below pre-industrial CCD (dark blue), above lysocline
(light blue) and expanding below CCD (red), where: (A) pre­
sent day; (B) with additional 100 m hypothetical shoaling of
CCD; (C) with 200 m hypothetical shoaling of CCD; (D) with
300 m hypothetical shoaling of the CCD. Since the future CCD
depth will be determined by deep ocean circulation and
changes in ocean chemistry, these maps do not represent a
time series, but rather only show changes in area with hypo­
thetical shoaling of the CCD. (For interpretation of the refer­
ences to colour in this figure legend, the reader is referred to
the web version of this article.)

5
P.T. Harris et al. Marine Geology 463 (2023) 107121

Fig. 6. Percent surface areas of selected geomorphic feature types versus hypothetical future shoaling of the CCD in 100 m increments. Note the Y-axes (percentage
feature area) are plotted on different scales.

6
P.T. Harris et al. Marine Geology 463 (2023) 107121

Table 2
Change in area of geomorphic features located below the CCD in km2 (percentage shown in brackets) at the present time in relation to pre-industrial times listed for 17
ocean regions shown in Fig. 1 and for the total global ocean. Largest percent change indicated by shaded numbers. NP = not present.
Region Abyssal Plain Basin km2 Canyon Fan km2 Plateau Escarp. Seamount Guyot Spreading Rift Valley Trench
km2 km2 km2 km2 km2 km2 Ridge km2 km2 km2

1. C Eq 220,412 29,833 20 1980 2080 60 100 50

Pacific (2.26%) (1.07%) 0 NP (<0.1%) (0.59%) (0.81%) (0.46%) (0.1%) (0.67%) NP
46,614 134,832 40,900 90 2200 780 40 10 10
2. C Indian (1.2%) (2.29%) 0 (1.6%) (0.01%) (0.65%) (0.99%) (3.56%) (<0.1%) (0.11%) NP
3. E Eq 1229 40,433 590 290 180
Atlantic (0.05%) (0.89%) 0 0 0 (0.13%) 0 0 (0.14%) (0.7%) NP
4. E Eq 41,765 95,416 80 290 40 2040 470 6890 260 890
Pacific (1.26%) (1.71%) (0.12%) (1.17%) (<0.1%) (0.47%) (0.45%) NP (1.33%) (0.79%) (0.71%)
205,141 611,304 390 640 17,100 7470 90 2060 370 2740
5. E Indian (3.6%) (5.14%) (0.24%) 0 (0.05%) (2.22%) (2.69%) (1.17%) (2.53%) (3.36%) (3.06%)
6. NE 217,614 618,129 10 15,180 1880 20 20
Atlantic (8.43%) (14.0%) 0 0 (<0.1%) (1.61%) (0.98%) 0 (<0.1%) (0.04%) NP
83,039 18,660 530 510 640 29,680 21,980 3360 170 140 1650
7. N Pacific (0.63%) (0.10%) (0.18%) (0.24%) (0.15%) (1.04%) (1.26%) (0.84%) (0.11%) (0.64%) (0.61%)
8. NW 853,607 1,848,230 1200 1120 2950 74,410 11,210 170 640 260 2080
Atlantic (20.9%) (22.5%) (0.37%) (0.46%) (0.51%) (6.11%) (5.68%) (4.2) (0.47%) (1.19%) (1.79%)
9. Philip. 8296 37,722 1670 610 15,530 4370 10 4060 360 130
Basin (0.75%) (0.47%) (0.32%) 0 (0.1%) (0.62%) (0.92%) (0.06%) (1.12%) (0.76%) (0.03%)
10. SE 135,455 319,316 2490 23,050 170 16,280 10,390 490 9600 1250
Atlantic (4.87%) (4.98%) (1.39%) (8.60%) (0.08%) (2.34%) (2.90%) (1.14%) (2.77%) (3.58%) NP
11. SE 391,892 1,083,490 720 7210 1160 20,070 9700 460 14,280 2870 360
Pacific (5.03%) (6.28%) (0.43%) (3.69%) (0.65%) (1.65%) (1.82%) (1.35%) (1.57%) (4.58%) (0.16%)
12. SW 695,703 241,089 11,760 27,290 4120 56,260 11,190 390 27,690 7580 3520
Atlantic (9.69%) (2.45%) (4.14%) (4.80%) (0.41%) (5.33%) (5.89%) (0.67) (3.99%) (11.17%) (3.48%)
13. SW 365,583 478,147 2340 5120 45,850 28,430 2050 7030 790 720
Pacific (3.53%) (2.75%) (0.63%) NP (0.09%) (1.71%) (2.31%) (1.88%) 0.97%) (1.00%) (0.26%)
1,313,843 1,290,078 5400 15,030 14,790 65,250 16,130 770 25,960 6690
14. S Indian (12.58%) (11.64%) (1.61%) (2.81%) (0.70%) (6.59%) (6.27%) (3.4%) (4.30%) (12.01%) NP
15. W Eq 180,723 427,251 40 55,000 560 10,390 4960 2260 1050
Atlantic (7.79%) (10.46%) (0.03%) (9.09%) (0.16%) (1.87%) (2.89%) 0 (0.82%) (2.81%) NP
16. W Eq 80,709 67,839 390 1420 17,100 13,590 830 890 360 1650
Pacific (1.51%) (0.71%) (0.23%) NP (0.07%) (0.74%) (1.13%) (0.67%) (0.56%) (1.57%) (0.81%)
17. W 26,542 78,603 60 70 2840 530 70 1680 470 200
Indian (0.63%) (1.45%) (0.02%) 0 (<0.1%) (0.26%) (0.18%) (0.19%) (0.23%) (0.54%) (0.63%)
4868,691 7,420,959 27,110 170,430 32,410 392,880 145,200 8800 103,610 22,700 13,860
Total (4.98%) (4.90%) (0.71%) (2.14%) (0.19%) (1.91%) (1.86%) (0.95%) (1.61%) (3.36%) (0.71%)

Fig. 7. Three-dimensional diagram illustrating how a projected 300 m rise in Fig. 8. Three-dimensional diagram illustrating how a projected 300 m rise in
CCD will result in a rise in the “snow line” with more of the Nazca Ridge CCD will result in a rise in the “snow line” with more rift valley area being
(plateau) area being located below the CCD. located below the CCD on the SE Pacific spreading ridge.

1.2. Special situation of the Arctic Ocean other ocean regions, linked to reduction in sea ice cover (Qi et al., 2022).
However, the models also predict an interesting effect of intruding
The analysis of water and seabed sediment samples indicate that the Atlantic Ocean water masses transporting carbonate undersaturated
deep abyssal seafloor of the Arctic Ocean lies wholly above the CCD water into mid-depth regions of the Arctic (Luo et al., 2016; Terhaar
(Jutterström and Anderson, 2005). Modeling suggests that deep water et al., 2020). In addition to the importation of carbonate undersaturated
masses in the Arctic are not expected to become under-saturated with water into the Arctic from the North Atlantic there is the potential added
calcium carbonate within this century. Recent observations indicate that impact of methane seepage from melting of seafloor gas hydrates and
the acidification of the Arctic Ocean is occurring four times faster than permafrost within the Arctic basin (Biastoch et al., 2011; GRID-Arendal,

7
P.T. Harris et al. Marine Geology 463 (2023) 107121

global bathymetry data layer at 15 arc sec (approximately 500 × 500 m

at the equator) resolution. The global bathymetry raster was clipped to
each ocean area to create 17 individual SRTMs for each ocean polygon.
This step was mostly done to increase the analysis speed of the subse­
quent classification of ocean depths. Each of the raster tiles for each
ocean area was reclassified four times to identify: (A) the depth between
pre-industrial and present CCD; (B) present CCD + 100 m; (C) present
CCD +200 m; and (D) present CCD + 300 m. Surface areas were then
calculated using the world Cylindrical Equal Area projection.
For the Arctic there is a special situation whereby acidification is
projected to intersect the depth range 400–800 m, as proposed by Ter­
haar et al. (2020). Thus, in Step 1 we replaced the elevation of the CCD
in depth increments and with this specific depth range in GIS. For Steps 2
and 3, we used the same approach as above within the 400–800 m depth
range.

Fig. 9. Three-dimensional diagram illustrating how a projected 300 m rise in 2.2. Geomorphic features
CCD will result in a rise in the “snow line” with more canyon area being located
below the CCD adjacent to the Argentina continental slope. Using GIS, we overlaid the derived bathymetry polygons (A-D as
defined above) with the global seafloor geomorphic features (GSFM)
2020; El bani Altuna et al., 2021). Hence, although shoaling of the deep database of Harris et al. (2014) to create inventories of features exposed,
water lysocline is not expected to expose any deep seafloor habitats to their numbers and surface areas for each ocean basin. Each of the out­
undersaturated carbonate waters this century, it is rather the mid-depth puts from the geomorphic layers pairwise intersect with each individual
waters in the Arctic, within the 400–800 m depth range (Terhaar et al., CCD layer were merged together to create 4 final polygon layers. Given
2020), that will potentially be impacted. the 30 arc sec, or approximately 1 km, grid resolution of the
SRTM30_PLUS bathymetric model of Becker et al. (2009) used by Harris
et al. (2014) to map geomorphic features, the location of the derived
1.3. Aims and objectives feature boundaries will reflect this resolution. Assuming the precision of
interpretation of the bathymetric model is 3 × 3 grid cells in any
We ask what ocean areas, which geomorphic features and which dimension, Harris et al. (2014) rounded all area estimations to the
nations will be impacted in the next century by the rising lysocline, by nearest 10 km2 which is adopted in this study.
mapping the area of seafloor that will be exposed to more corrosive
ocean water as a result of increased anthropogenic CO2 dissolved in 2.3. National Exclusive Economic Zones and the rising CCD
seawater and attendant rising of the CCD. We will focus on the area of
seabed that was previously located within the lysocline, immediately We overlaid the derived bathymetry polygons (A-D as defined above)
above the CCD, but which is now below it and exposed to corrosive with national Exclusive Economic Zones (EEZs) from VLIZ 2022? to
bottom waters (i.e., area Exp-A in Fig. 3). Our objective is to better create inventories of features exposed, their numbers and surface areas
understand which geographic regions and what categories of geomor­ for each country. Each of the outputs from the geomorphic layers pair­
phic features are most at risk in order to inform better decision-making wise intersect with each individual CCD layer were merged together to
in ocean governance. create four final polygon layers to the nearest 10 km2 as explained
above.
2. Methods
3. Results
2.1. Calculation of ocean regions and their exposure to shoaling CCD
3.1. Ocean areas below the CCD
Global data on the depth of carbonate compensation depth (CCD)
and calcite saturation depth (CSD) were tabulated in Sulpis et al. (2018) For the 17 ocean regions included in the Sulpis et al. (2018) analysis
for 17 different ocean regions (Fig. 1). We mapped the pre-industrial and (excluding the Arctic Ocean) our analysis indicated that around 37%
modern (circa 2002) depths of the CSD and CCD (from Sulpis et al., (127,303,272 km2) of the global ocean floor lies below the pre-industrial
2018) in ArcGIS using the bathymetric depth contours from a recent depth of the CCD. Shoaling of the CCD since the industrial revolution has
global bathymetry map (Tozer et al., 2019). The surface area located submerged an additional 12,432,096 km2 of ocean floor (3.6% of total
between these two isobaths in each of the 17 ocean regions was then ocean area) below the CCD. Conceptually, this is the area “Exp. A" as
measured and mapped in GIS. illustrated in Fig. 3. Thus, at present, 40.5% of the ocean floor
Based on the modern CCD, we assume an increase in elevation of the (139,685,248 km2) is submerged below the CCD.
CCD in three 100 m increments and mapped the affected seafloor area Further hypothetical shoaling (shallowing) of the CCD by an incre­
(for 100, 200 and 300 m rises in CCD). These increments were subjective ment of 100 m illustrates that the surface area of seafloor submerged
and hypothetical but reflect general expectations in the extent that the below the CCD (Exp. A; Fig. 3) rises by 24,947,550 km2 equal to 7.2% of
CCD will shoal within this century (Archer et al., 1998; Sulpis et al., total ocean area. An additional 48,298,879 km2 equal to 14.0% of total
2018), noting that the current depth of the CCD is not known to within ocean area is submerged below the CCD with 300 m of CCD shoaling,
<200 to 300 m depth in most ocean regions (Fig. 2; Table 1). The referenced to the pre-industrial depth of the CCD (Table 1). Thus, if the
amount of shoaling that has occurred since pre-industrial times (albeit CCD shoals by 300 m then 51% of the ocean seafloor area will be deeper
poorly constrained) varies significantly between ocean regions (Fig. 2) than the CCD.
and hence it is expected that the CCD shoaling will occur at different There is much heterogeneity in the areas below the CCD between the
rates in different regions, faster in areas of downwelling and much different ocean regions (Fig. 4). The area below the CCD in pre-industrial
slower (taking probably many centuries) in other regions. times is estimated here to have ranged from 0.55% to 66% of ocean
All the SRTM15 tiles from Tozer et al. (2019) were merged to create a regions, which is a significant difference even allowing for the large

8
P.T. Harris et al. Marine Geology 463 (2023) 107121

Table 3
Feature areas mapped in this study as below the pre-industrial CCD as defined by Sulpis et al. (2017).
Ocean Region Abyssal Hills Abyssal Mountains Bridges Hadal Rise (km2) Slope (km2) Sill Terrace Trough
(km2) (km2) (km2) (km2) (km2) (km2) (km2)

1 Central Equatorial 2,525,430 308,980 5100 100

Pacific
2 Central Indian Ocean 2,077,120 289,760 90 210 503,750 10 37,900
3 East Equatorial 24,960 8090 22,620
Atlantic
4 Eastern Equatorial 1,275,200 209,300 300 10,670 48,880 4450
Pacific
5 Eastern Indian Ocean 2,049,870 503,010 770 214,960 47,940 4780 47,620
6 North East Atlantic 442,590 200,630 38,920 57,380 1060
7 North Pacific 12,190,340 3,731,270 690 147,990 45,750 140,690 6060 97,760
8 North West Atlantic 1,057,500 320,680 30 232,500 38,290 790 4200
9 Philippine Basin 2,907,060 2,116,010 980 578,200 135,400 230 500 147,370
10 South East Atlantic 3,199,360 647,910 720 369,440 860
11 South East Pacific 5,340,510 837,210 410 51,340 866,100 53,160
12 South West Atlantic 3,303,530 959,980 130 202,490 2,658,650 7120 300 62,110
13 South West Pacific 11,347,360 2,175,680 390 223,600 216,590 50,900 980 73,290
14 Southern Indian 3,686,540 805,460 3100 1,319,200 1810 20
Ocean
15 West Equatorial 1,107,790 289,760 3430 359,600 30
Atlantic
16 Western Equatorial 5,918,600 2,275,240 380 352,880 53,340 170 630 193,000
Pacific
17 Western Indian 686,200 279,780 30 940 61,100 970 5280
Ocean
TOTAL Area Below CCD 59,139,970 15,958,740 4210 3,421,620 6,543,780 498,810 1704 7200 674,030
Total Area 146,153,780 54,859,300 7830 3,425,890 28,487,930 17,142,750 37,358 1,992,850 2,566,490

errors in the estimate of CCD depth (Table 1; Sulpis et al., 2018). The industrial and present times and also the areas predicted to be sub­
average pre-industrial percentage area below the CCD of the 17 areas is merged by future CCD shoaling (Table 2; Fig. 6; Supplementary Table 1).
31.7 +/− 17.2 (standard deviation). The North Pacific had the largest Vertically extensive features, like escarpments, guyots, seamounts
area below the CCD before the industrial revolution at 66%, rising to and submarine canyons, partially intersected with the depth range of the
about 73% with 300 m of hypothetical CCD shoaling (Fig. 4). The East rising CCD (e.g., Fig. 3). Only the deepest parts of canyons and fans
Equatorial Atlantic had the smallest percentage area below the CCD extended below the CCD, hence their pre-industrial areas below the CCD
before the industrial revolution at 0.55%. With 300 m of hypothetical were on average <10% (Fig. 6). The effect of shoaling of the CCD could
CCD shoaling the area below the CCD increases to only 6%, the smallest result in some features coming into contact with the CCD for the first
area below the CCD of any ocean region. The North East Atlantic has the time; this can be seen for guyots, spreading ridges, submarine fans and
second smallest area below the CCD at 6.1%, which increases to about canyons (Fig. 6). In such cases, the deepest parts of features occurred
20% with 300 m of hypothetical CCD shoaling (Fig. 4). below the CCD after it has shoaled sufficiently; this is illustrated in Fig. 6
The change in seafloor area below the CCD between pre-industrial as a line emerges mid-way along the X-axis representing the first
times and the present was greatest for the Northwest Atlantic, the occurrence of a feature in that ocean region to extend below the depth of
Southwest Atlantic and the Southern Indian Ocean (Fig. 5A). As noted the CCD.
above, the 200 to 300 m rise in CCD in these areas is attributed by Sulpis In contrast, >90% of the deep-water trenches were already below the
et al. (2018) to strong downwelling of oceanic water masses containing CCD in pre-industrial times (Fig. 6). For abyssal plains and basins the
increased concentrations of dissolved CO2 compared with pre-industrial figures were around 50 to 60% on average (Fig. 6). Seamounts, guyots,
times. The changes in CCD correspond with ocean areas of between escarpments, and rift valleys all had between 20 and 30% of their areas
8.59% and 11.09% (well above the global average of 3.56%) that have submerged below the CCD in pre-industrial times (Fig. 6). The changes
become submerged below the CCD since the industrial revolution. These in CCD affecting each category of geomorphic feature are described in
areas increase to between 19.24% and 20.77% with 300 m of hypo­ the following sections.
thetical shoaling of the CCD (Figs. 4 and 5).
Four ocean regions had <1% increases in area submerged below the 3.2.1. Abyssal plains
CCD between pre-industrial and present: the Eastern Equatorial Atlantic The area of abyssal plains in the 17 ocean regions is 97,843,544 km2.
(0.40%); Philippine Basin (0.42%) Western Indian Ocean (0.46%) and A total of 49% of abyssal plains were submerged below the pre-
the North Pacific (0.74%). These regions are distant from locations of industrial CCD and this has increased to 54.3% at the present time.
downwelling water masses and presumably have relatively “old” bottom The North Pacific and South East Atlantic had the largest areas sub­
water in comparison with other ocean regions. merged below the CCD at 79.3% and 76.3%, respectively in pre-
industrial times and 80.0% and 80.9% at present. In contrast, the
Eastern Equatorial Atlantic had only 0.01% submerged below the pre-
3.2. Geomorphic features submerged below the CCD industrial CCD increasing to 0.05% at present. With 300 m of hypo­
thetical CCD shoaling, the average area of abyssal plains submerged is
All categories of off-shelf seafloor geomorphic features mapped by predicted to increase from 54.3% at present to 67.4% (Fig. 6).
Harris et al. (2014) were found to intersect the depth-range of the The highest percentages of increase in area submerged below the
lysocline in several (if not all) ocean regions in the present study. These CCD since pre-industrial times (steepest slopes in Fig. 6) occurred in the
extended from the deepest hadal features (e.g., ocean trenches) to fea­ Northwest Atlantic and Southern Indian Ocean regions (20.9% and
tures extending onto the continental slope (submarine canyons and fans; 12.6%, respectively). A smaller (<1%) increase in surface area occurred
Fig. 6). In general, there was great variability between ocean regions in in the East Equatorial Atlantic, North Pacific and the Philippine Basin.
terms of the percentages of features submerged below the CCD in pre-

9
P.T. Harris et al. Marine Geology 463 (2023) 107121

Table 4 0.82% submerged below the pre-industrial CCD and increased to 0.96%
List of geomorphic features projected to be submerged within the 400–800 m at present. With 300 m of hypothetical CCD shoaling, the average area of
depth range of a hypothetical intrusion of an acidified mid-depth water mass in escarpment submerged was predicted to increase from 22.7% at present
the Arctic Ocean as proposed by Terhaar et al. (2020). Features listed in to 29.9% (Fig. 6).
decreasing order of percent area submerged. The highest percentages of increase in area of escarpment submerged
Feature Area In 800–400 depth zone below the CCD from pre-industrial times to the present (steepest slopes
(km2) (km2) % in Fig. 6) occurred in the Southern Indian Ocean, North West Atlantic
and South West Atlantic, regions (6.6%, 6.0% and 5.3%, respectively). A
Terrace 146,870 41,590 28.32
Glacial Trough 1,107,880 241,330 21.78 smaller (<0.5%) increase in surface area occurred in the East Equatorial
Slope 642,390 116,140 18.08 Atlantic, Western Indian Ocean and Eastern Equatorial Pacific.
Shelf 3,849,470 368,330 9.57
Plateau 778,390 72,540 9.32 3.2.4. Seamounts
Canyon 328,930 16,920 5.14
Ridge 74,820 2890 3.87
Seamounts cover 7,814,770 km2, and on average 24.8% of seamount
Escarpment 146,430 3950 2.69 area was submerged below the CCD in pre-industrial times, increasing to
26.6% at the present time. For reference, Harris et al. (2014) mapped
9951 seamounts in the world ocean having an average area of 790 km2
(see also Costello et al., 2010; MacMillan-Lawler and Harris, 2016). The
North Pacific and Philippine Basin had the greatest areas below the pre-
industrial CCD at 46.8% and 35.0%, respectively, which has increased to
48.1% and 35.9%, respectively, at the present time. In contrast, the
Eastern Equatorial Atlantic had only 0.08% of seamount area submerged
below the pre-industrial CCD with no change up to the present. With
300 m of hypothetical CCD shoaling, the average area of seamount
submerged was predicted to increase from the present 26.6% to 36.2%
(Fig. 6).
The highest percentages of increase in area of seamount submerged
below the CCD from pre-industrial times to the present (steepest slopes
in Fig. 6) occurred in the Southern Indian Ocean and the South West
Atlantic, regions (6.3% and 5.9%, respectively). A smaller (<0.5%) in­
crease in surface area occurred in the East Equatorial Atlantic, Western
Indian Ocean and Eastern Equatorial Pacific.

3.2.5. Plateaus
Fig. 10. Map showing the area (in red shading) impacted by the hypothetical Plateaus cover 17,229,030 km2 and for the most part they do not
intrusion of an acidified water mass into the central Arctic Ocean and affecting occur below the CCD. On average 0.48% of plateau area was submerged
the 400 to 800 m depth range. Locations of some large areas of glacial troughs, below the CCD in pre-industrial times, increasing to 0.67% at the present
terraces, plateaus and canyons are indicated. (For interpretation of the refer­ time. The North Pacific had the greatest areas below the pre-industrial
ences to colour in this figure legend, the reader is referred to the web version of
CCD at 2.3%, which has increased to 2.4% at the present time. The
this article.)
Eastern Equatorial Atlantic is the only ocean region with no plateau area
submerged below the CCD, either in pre-industrial times or with 300 m
3.2.2. Basins of hypothetical CCD shoaling. With 300 m of hypothetical CCD shoaling,
The area of basins in the 17 ocean regions is 151,588,990 km2. A the average area of plateau submerged below the CCD was predicted to
total of 64.9% of basins were submerged below the pre-industrial CCD increase from 0.67% at present to 1.5% (Fig. 6).
and this has increased to 69.8% at the present time. The North Pacific An interesting phenomenon occurs for the Nazca ridge (plateau) in
and South West Atlantic had the largest areas submerged below the CCD the Southeast Pacific which exhibits a dramatic increase in surface area
at 90.9% and 84.9%, respectively in pre-industrial times and 91.0% and submerged by the shoaling CCD (Fig. 6). The South East Pacific contains
87.3% at present. North Pacific basins were also found at the greatest only one major plateau feature, the Nazca Ridge (Fig. 7), covering
water depths of over 5300 m (Harris and Macmillan-Lawler, 2017). In 179,602 km2 which presently has 1.1% of its area below the CCD but
contrast, the Eastern Equatorial Atlantic had only 1.2% submerged which is predicted to have 10.4% submerged with 300 m of CCD
below the pre-industrial CCD increasing to 2.1% at present. With 300 m shoaling. The large increase in surface area predicted to be submerged
of hypothetical CCD shoaling, the average area of basins submerged was below the CCD is explained by the geomorphology of plateaus; once the
predicted to increase from 69.8% at present to 81.9% (Fig. 6). CCD rises to breach the level of the upper, flat surface of a plateau, the
The highest percentages of increase in area submerged below the surface area below the CCD increases dramatically with only small
CCD from pre-industrial times to the present (steepest slopes in Fig. 6) amounts of CCD shoaling.
occurred in the North West Atlantic, North East Atlantic and Southern
Indian Ocean regions (22.3%, 13.9 and 11.6%, respectively). A smaller 3.2.6. Guyots
(<1%) increase in surface area occurred in the East Equatorial Atlantic, Guyots cover a total area 930,370 km2 in the 17 ocean regions
North Pacific, Western Equatorial Pacific and the Philippine Basin. dominated by the North Pacific Ocean where 43% of guyot area occurs.
Harris et al. (2014) mapped 283 guyots in the world ocean having an
3.2.3. Escarpments average area of 3300 km2; there are no guyots in the Eastern Equatorial
Escarpments cover 20,524,040 km2, of which on average 20.7% Pacific and guyots in three ocean regions (Central Indian Ocean, Eastern
were submerged below the CCD in pre-industrial times increased to Indian Ocean and the North West Atlantic) covered <10,000 km2.
22.7% at the present time. The Philippine Basin and North Pacific had On average 23.5% of guyot area was submerged below the CCD in
the greatest areas below the pre-industrial CCD at 36.6% and 34.7%, pre-industrial times, increasing to 24.4% at the present time. The North
respectively, which has increased to 37.3% and 35.8%, respectively, at Pacific had the greatest areas below the pre-industrial CCD at 38.8%,
the present time. In contrast, the Eastern Equatorial Atlantic had only which has increased to 39.7% at the present time. The Eastern

10
P.T. Harris et al. Marine Geology 463 (2023) 107121

Table 5 Equatorial Atlantic was the only ocean region with no guyot area sub­
List of EEZ’s with >10% increase in area of seafloor submerged below the CCD merged below the CCD in pre-industrial times or at the present; a small
due to hypothetical 300 m shoaling of the CCD since pre-industrial times (listed area of 0.12% was predicted to become submerged with 200 m of CCD
in order of increase in percent area impacted since pre-industrial times). Also shoaling.
listed are the EEZ area, and total percentage area that would be located below With 300 m of hypothetical CCD shoaling, the average area of guyot
the CCD with a 300 m rise of CCD depth. The comments column lists the ocean
submerged below the CCD was predicted to increase from 24.4% at
region together with the percentage area of submarine canyons, seamounts and
present to 30.6% (Fig. 6). A similar situation occurs for guyots as was
other features most impacted by a 300 m rise in CCD depth within each EEZ.
seen for plateaus, both of which exhibit steep sides and a flat top; guyots
EEZ name EEZ area Pre- % EEZ % Comments
in the Central Indian Ocean and in the North West Atlantic exhibited a
(Country) km2 Industrial area Change
% EEZ area below
dramatic increase in surface area submerged below the CCD from 10.5%
below CCD CCD and 4.2% at the present time, respectively, to 37.7% and 20.7%,
with respectively, with 300 m of CCD shoaling (Fig. 6).
300 m
rise in
3.2.7. Trenches
CCD
Deep ocean trenches are a unique category of geomorphic feature in
Bermuda 464,390 26.10 68.15 42.05 Northwest
that, given their extreme water depths, most of their areas are expected
(UK) Atlantic
32.8% of
to be below the present CCD. Only ten ocean regions contain trenches
terraces and they are absent from the Central Equatorial Pacific, Central Indian
25.5% Ocean, Eastern Equatorial Atlantic, North East Atlantic, Western Equa­
Seamounts torial Atlantic, South East Atlantic and the Southern Indian Ocean.
El Salvador 95,100 0.79 37.96 37.17 Eastern
Trenches cover an area of 1,944,060 km2 and on average 91.7% of
Equatorial
Pacific 38.4% trench area was submerged below the CCD in pre-industrial times,
ridge increasing to 92.4% at the present time (Fig. 6).
2.9% canyons Since all the trench area is below the CCD in some ocean regions,
Nauru 309,260 0.03 35.73 35.70 Western
there is no increase in area with shoaling of the CCD possible, and this is
Equatorial
Pacific 4.1%
the pattern seen for the Northwest Atlantic, Philippine Basin, Southeast
seamounts Pacific and the Southwest Pacific (Fig. 6). An interesting case occurs in
Guatemala 110,660 2.22 37.46 35.24 Eastern the Western Indian Ocean in which the Amirante Trench (located to the
Equatorial west of the Seychelles) and another un-named trench located within the
Pacific 74.5%
Carlsberg Ridge intersect the CCD, dividing the trench area of 31,894
ridge
2% canyons km2 into two halves, located above and below the CCD.
Sri Lanka 533,550 0.84 28.95 28.11 Central Indian
Ocean 3.2.8. Spreading ridges
34% fan, Spreading ridges cover an area of 6,426,930 km2, and on average
34.9% rise,
0.4% canyons,
12.3% of spreading ridge area was submerged below the CCD in pre-
19.4% industrial times, increasing to 14.0% at the present time. The Philip­
seamounts pine Basin had the greatest areas below the pre-industrial CCD at 63.1%,
Christmas 327,990 3.97 31.97 28.00 Eastern Indian which has increased to 64.3% at the present time. In contrast, the North
Island Ocean
East Atlantic and Central Indian Ocean both had only 0.02% of
(Aus) 15.8%
seamounts spreading ridge area submerged below the pre-industrial CCD with
Chile - San 450,960 10.37 38.35 27.98 Southeast virtually no change up to the present. With 300 m of hypothetical CCD
Felix and Pacific shoaling, the average area of spreading ridge submerged was predicted
San 30.8% guyot to increase from 14.0% at present to 21.3% (Fig. 6). The Eastern
Ambrosio 16.7%
islands seamounts
Equatorial Pacific exhibited the greatest increase in spreading ridge area
Macquarie 475,720 7.47 34.61 27.14 Southwest predicted to become submerged, increasing from 16.0% at present to
Island Pacific 38.7% (an increase of 22.7%) with 300 m of CCD shoaling.
(Aus) 14% plateau The highest percentages of increase in area of spreading ridges
10.8%
submerged below the CCD from pre-industrial times to the present
canyons
19.6% (steepest slopes in Fig. 6) occurred in the Southern Indian Ocean and the
seamounts South West Atlantic, regions (4.3% and 4.0%, respectively). A smaller
Cocos 467,440 6.80 29.03 22.23 Eastern Indian (<0.5%) increase in surface area occurred in the Central Indian Ocean,
Islands Ocean Central Equatorial Pacific, North East Atlantic, North West Atlantic, East
(Aus) 15.4%
escarpments
Equatorial Atlantic, North Pacific and Western Indian Ocean.
9.2%
seamounts 3.2.9. Rift valleys
French 4,766,850 5.42 27.14 21.72 Central Rift valleys cover an area of 674,750 km2, and 31.8% of rift valley
Polynesia Equatorial
area on average was below the CCD in pre-industrial times, rising to
Pacific
3.6% canyons, 35.1% at present. The Philippine Basin had the greatest area below the
15.6% pre-industrial CCD at 93.6%, which has increased to 94.4% at the pre­
seamounts sent time. In contrast, the North East Atlantic and Central Indian Ocean
Chile 2,488,070 8.52 29.29 20.77 Southeast had only 0.09% and 0.18%, respectively, of rift valley area submerged
Pacific
50.4% rift
below the pre-industrial CCD increasing to 0.13% and 0.30%, respec­
valley, 22.8% tively, at present. With 300 m of hypothetical CCD shoaling, the average
seamounts, area of rift valley submerged is predicted to increase from 35.1% at
19.9% slope, present to 45.2% (Fig. 6). The South East Pacific (Fig. 8) exhibits the
greatest increase in rift valley area predicted to become submerged,

11
P.T. Harris et al. Marine Geology 463 (2023) 107121

Table 5 (continued ) increasing from 41.6% at present to 65.6% (an increase of 28.6%) with
EEZ name EEZ area Pre- % EEZ % Comments 300 m of CCD shoaling.
(Country) km2 Industrial area Change The highest percentages of increase in area of spreading ridges
% EEZ area below submerged below the CCD from pre-industrial times to the present
below CCD CCD (steepest slopes in Fig. 6) occurred in the Southern Indian Ocean and the
with
300 m
South West Atlantic, regions (12.0% and 11.2%, respectively). A smaller
rise in (<0.5%) increase in surface area occurred in the Central Indian Ocean
CCD and in the North East Atlantic.
3.7% canyons,
1.9% trench 3.2.10. Submarine fans
Réunion (Fr) 315,360 6.30 26.40 20.10 Western Submarine fans cover an area of 7,950,110 km2, and occurred only in
Indian Ocean 10 out of the 17 ocean regions. On average, 7.7% of fan area was below
32.2% of one
guyot, 46.2%
the CCD in pre-industrial times, rising to 9.8% at present. The Eastern
spreading Equatorial Pacific had the greatest percent area below the pre-industrial
ridge, 10.5% CCD at 36.4%, which has increased to 37.6% at the present time, but this
seamounts relates only to a small area of fan that occurs in this region (24,760 km2).
Palmyra 353,720 2.04 21.01 18.97 Central
Of greater significance is the area of the Bengal Fan (the largest fan
Atoll (US) Equatorial
Pacific complex on earth) which occurs in the Central Indian Ocean making a
15.9% total fan area of 2,558,480 (32% of the world’s total fan area). We find
seamounts that 20% of fan area in the Central Indian Ocean was below the CCD
Curaçaoa 25,400 7.85 26.12 18.27 Western prior to the industrial revolution, and 21% at present.
(Neth) Equatorial
Atlantic
With 300 m of hypothetical CCD shoaling, the average area of fan
0.2% canyons submerged is predicted to increase from 9.8% at present to 20.3%
Crozet 575,380 16.52 34.71 18.19 Southern (Fig. 6). The South East Atlantic exhibits the greatest increase in fan area
Islands Indian Ocean predicted to become submerged, increasing from 15.9% at present to
(Fr)
37.1% with 300 m of CCD shoaling. The highest percentages of increase
Seychelles 1,341,480 0.67 18.64 17.97 Western
Indian Ocean in area of submarine fan submerged below the CCD from pre-industrial
0.7% canyons, times to the present (steepest slope in Fig. 6) occurred in the Western
4.4% Equatorial Atlantic (increase of 9.1%).
seamounts
Kiribati 1,053,240 1.33 19.19 17.86 Western
3.2.11. Canyons
Equatorial
Pacific Submarine canyons cover an area of 3,794,790 km2 and on average,
2.0% canyons, 2.7% of canyon area was below the CCD in pre-industrial times, rising to
12.8% 3.4% at present. Harris et al. (2014) mapped 9477 canyons in the world
seamounts
ocean having an average area of 460 km2. Although submarine canyons
Western 130,480 5.53 22.92 17.39 Western
Samoa Equatorial are abundant in all ocean basins, they are mainly located on the conti­
Pacific nental slope and hence above the CCD of most ocean regions. Never­
3.9% canyons, theless, we find canyons intersect the lysocline in 12 of the 17 ocean
14.2% regions (Fig. 6).
seamounts
In the Philippine Basin canyons cover an area of 528,640 km2 of
Bahamas 619,790 8.30 22.91 14.61 Northwest
Atlantic which 7.6% was below the pre-industrial CCD, which is the highest
94.7% rise, percentage of canyons found below the pre-industrial CCD in any ocean
1.8% canyons, region. However, canyons in the South West Atlantic (area of 283,870
19.6%
km2) have exhibited an extraordinary increase in area submerged below
seamounts
St. Helena 449,220 6.80 20.95 14.15 Southeast
the CCD from 3.9% in pre-industrial times to 8.0% at the present time.
(UK) Atlantic Thus, at the present time it is the South West Atlantic that has the highest
8.0% percentage of canyons found below the CCD in any ocean region.
seamounts With 300 m of hypothetical CCD shoaling, the average area of
Portugal 315,480 6.30 20.25 13.95 Northeast
canyon submerged was predicted to increase from 3.4% at present to
Atlantic
43.6% rise, 6.5% (Fig. 6). The South West Atlantic (Fig. 9) was predicted to maintain
1.1% canyons, its high rate of increase in area submerged below the CCD, rising to
2.4% 17.3% of canyon area with 300 m of hypothetical CCD shoaling (Fig. 6).
seamounts
Prince 474,900 11.21 24.14 12.93 Southern
Edward Indian Ocean
3.2.12. Other categories of geomorphic features
Islands 18.1% rift In addition to the above, we analysed the following separate
(South valley, 8.7% geomorphic feature categories as mapped by Harris et al. (2014): abyssal
Africa) seamounts hills; abyssal mountains; bridges; hadal area; continental rise; conti­
Aruba 29,970 7.45 19.58 12.13 Western
nental slope; terrace; and trough. We found that some part of every
Equatorial
Atlantic feature category (n = 20) was located below the pre-industrial CCD in at
5.3% canyons least one ocean region (see Table 3).
Guadeloupe 90,850 17.19 29.17 11.98 Northwest
Atlantic
3.3. Ocean acidification impacts in the Arctic
43% fan,
47.4% ridge,
0.5% canyons The intrusion of an acidified mid-depth water mass as proposed by
Terhaar et al. (2020) in the 400–800 m depth range impacts upon 8
categories of seafloor geomorphic features (Table 4). The area of the

12
P.T. Harris et al. Marine Geology 463 (2023) 107121

Fig. 11. Risk matrix for the 17 Ocean Regions

included in this study, assuming that the change in
CCD from pre-industrial times to the present is an
indicator of the likelihood that the CCD will continue
to rise over the next century (X-axis) and the increase
in percentage of seafloor submerged below the CCD
due to a hypothetical 300 m rise in the CCD is an
indicator of the consequences (Y-axis). The dashed
lines separate four quadrants and represent the
average rise (97.8 m) of the CCD from pre-industrial
times to the present and the average change in per­
centage of seafloor submerged below a CCD 300 m
higher than present (11.0%). The western equatorial
Atlantic (Ocean Region 15) is the only Ocean Region
to plot in the high risk zone (error bars for Ocean
Region 15 are shown to illustrate the error in esti­
mated CCD rise for all ocean regions as per Table 1).

Arctic Ocean within the 400–800 m depth range covers 1,296,800 km2 4. Discussion
(Fig. 10), which was equal to about 8.3% of the Arctic basin region. The
intrusion of the hypothetical acidified mid-depth water mass onto the 4.1. Ocean regions most at risk to future rise in CCD
continental shelf occurred within the deeply incised glacial troughs
(Fig. 10). Apart from glacial troughs, our analysis indicates that the To assess the risk of the seafloor area of any region being submerged
geomorphic features potentially most impacted by the hypothetical below the CCD, apart from the rate at which the CCD is rising, the
intrusion of an acidified mid-depth water mass are terraces, the conti­ important factors are the amount of area that remains above the CCD
nental slope, shelf and plateaus (Table 4). Canyons, ridges and escarp­ and the amount of area that will be submerged with a hypothetical 300
ments are also potentially impacted. m rising of the CCD. The pre-industrial east equatorial Atlantic was 94%
above the CCD and obviously has “more to lose” than the North Pacific,
3.4. Shoaling CCD impacts within Exclusive Economic Zones two thirds of which (66%) was already below the CCD in pre-industrial
times (Table 1). The average of all 17 ocean regions is 37% below the
The Exclusive Economic Zone (EEZ’s) of some countries will be CCD in pre-industrial times.
affected more by the shoaling of the CCD than others. Generally, oceanic The rise in the CCD from pre-industrial times to the present has
and island nations lose more, while countries with large continental affected some areas more than others. The Northwest Atlantic and the
shelves lose proportionately less. Bermuda’s EEZ is predicted to be the Southern Indian Ocean both experienced a 10% increase in area sub­
most impacted by a 300 m shoaling of the CCD above the present level, merged below the CCD. The area below the CCD nearly doubled in the
with 68% of that country’s seabed becoming submerged below the CCD Northwest Atlantic from 13% to 23% (Table 1) and this was the region
(Table 5). In contrast, only 6% of the US EEZ and 0.39% of the Russian that had the largest overall rise in CCD of 316 m (Table 1). The overall
EEZ were predicted to be impacted by 300 m shoaling of the CCD above global effect of the rise in the CCD since pre-industrial times is to expose
the present level. Also, around 6.1% of the South China Sea EEZ (subject an additional 12,432,100 km2 of ocean floor (3.6% of total ocean area)
to multiple, overlapping claims) was predicted to be impacted by 300 m below the CCD at the present time. This relatively small (average of 97.8
shoaling of the CCD. m) change in the depth of the CCD has therefore had a large impact on
In terms of geomorphic features, 14.68% of canyons in the small the surface area submerged below it. However, in the North Pacific,
Caribbean island state of Antigua and Barbuda were predicted to Western Indian Ocean, Philippine Basin and the East Equatorial Atlantic,
become submerged below CCD with 300 m of shoaling, as were 33% of there was <1% change in area from pre-industrial times to the present.
Bermuda’s terraces and 25% of one seamount in Bermuda’s EEZ. Also <1% of the East Equatorial Atlantic is presently submerged below the
noteworthy was that 94% of the continental rise in the Bahama’s EEZ CCD.
(an area of 37,632 km2) and 74.5% of ridge features in the Guatemalan The only way to test the relative sensitivity of each ocean region to
EEZ were predicted to become submerged below CCCD with 300 m of future shoaling of the CCD is to assume a uniform rise in the CCD such as
shoaling. In the Chile EEZ, it was predicted that 50.4% of rift valley area, has been carried out in this study. The reason is because the geo­
19.9% of slope area and 1.9% of trench area became submerged below morphology of the ocean floor located at the CCD will vary between
CCCD with 300 m of shoaling (Table 5; Supplementary Table 2). ocean regions. The surface area submerged below the rising CCD will be
Developing countries including El Salvador, Guatemala, Sri Lanka greatest in locations of low-gradient topography and least in areas of
and Kiribati will apparently be more severely impacted by future ocean high-gradient topography.
acidification than most developed countries (US, Russia and China) that For a hypothetical 300 m rising of the CCD the geographic region
are also the main sources of the greenhouse gas emissions (US EPA, that will experience the greatest relative change in percentage of surface
2023) that have caused ocean acidification. For example, the area of area is the Central Equatorial Pacific, where the percent area below the
seabed projected to be submerged below the CCD within the US EEZ CCD increases from 46% in pre-industrial times to 71% (a 26% change in
(lower 48 states) rose from 0.09% to 6.09% with 300 m CCD shoaling, surface area). Heterogeneity of seafloor geomorphology located at the
whilst that of Russia rose from 0.04 to 0.39%. depth of the CCD results in differences in area submerged with a fixed
amount of CCD rise. A 300 m change in CCD in the Central Equatorial
Pacific results in a 26% increase in new area below the CCD, whereas the
same change results in only a 4% increase in area below the CCD in the
Philippine Basin.
Although this analysis does not represent any specific future

13
P.T. Harris et al. Marine Geology 463 (2023) 107121

(caption on next page)

14
P.T. Harris et al. Marine Geology 463 (2023) 107121

Fig. 12. Risk matrices for 11 categories of seafloor geomorphic features included in this study, as per Fig. 12. The dashed lines separate four quadrants and represent
the average rise (97.8 m) of the CCD from pre-industrial times to the present and the average change in percentage of seafloor geomorphic feature submerged below a
CCD 300 m higher than present (which varies between feature types). Note error bars (not shown) for CCD rise extend beyond the limits of the plotted area
(see Table 1).

scenario, it does serve to illustrate how the seafloor in different ocean conservation priorities and targets.
regions will respond to the rising CCD at some time in the future. Based For example, the percent area of submarine canyons submerged
on the rate at which the CCD has risen since pre-industrial times, we can below the CCD will increase by up to 17% in the southwest Atlantic
predict that the NE Atlantic, which has already experienced a > 300 m (more than any other ocean region), whereas virtually no area of sub­
rise in the CCD, might realise an additional >100 m rise in the CCD by marine canyon is predicted to become submerged below a CCD that is
the end of this century. In contrast, the central and eastern equatorial 300 m shallower than at present in the northeast Atlantic or in the
Pacific and western Indian Ocean regions have not experienced any central Indian Ocean (Fig. 6). For seamounts we find that in the North
significant rise in the CCD since pre-industrial times (Table 1) and so it Pacific over 50% of seamount area will be below a CCD that is 300 m
may be some centuries before these regions will realise any significant shallower than at present, whereas virtually no area of seamount is
(~100 m) rise in the CCD. predicted to become submerged below a CCD that is 300 m shallower
A relative risk assessment that considers the likelihood and the than present in the eastern equatorial Atlantic (Fig. 6).
consequences of the CCD rising can be created using the information Using the same approach as for ocean regions, we have estimated the
available. We assume that the change in CCD from pre-industrial times relative risk for 11 categories of geomorphic feature (Fig. 12). The
to the present is an indicator of the likelihood that the CCD will continue likelihood (x-axis) in the risk matrices is based on the change in CCD
to rise at a similar pace over the next century. Next we assume that the from pre-industrial times to the present, which is the same for each
change in percentage of seafloor submerged below the CCD due to a category; only the consequences (y-axis) varies between categories.
hypothetical 300 m rise in the CCD is an indicator of the consequences. Abyssal plains are found to be at high risk of submergence in 5 ocean
Plotting these variables in an X-Y scatter plot provides the basis for a regions and rift valleys are found to be at high risk of submergence in 4
relative risk assessment for the 17 ocean regions included in this study ocean regions. We find that 20 to 30% more of abyssal plains would be
(Fig. 11). submerged by a 300 m rise in the CCD (note that an average of ~49% of
Based on this assessment we find that the area at greatest relative risk abyssal plains are already below the CCD), located in four Atlantic
is the Western Equatorial Atlantic (Ocean Region 15; Fig. 11), which has Ocean regions and in the southern Indian Ocean. Seamounts and sub­
experienced an above average rise in CCD together with an above marine canyons are at high risk in three ocean regions (Fig. 12).
average percentage of seafloor submerged below a CCD that is 300 m Overall we see that geomorphic features in the Atlantic Ocean and
higher than present. We note that 5 of the 6 ocean regions that plot on southern Indian Ocean have a greater risk of being impacted by a rising
the right-hand (high likelihood) side of Fig. 11 are in the Atlantic, along CCD than those located in the Pacific Ocean and in other Indian Ocean
with the southern Indian Ocean (ocean region 14). Apart from ocean regions. However, features in Group B of mainly Pacific Ocean regions
region 3 (Eastern Equatorial Atlantic) all of the Atlantic Ocean regions are still at a moderate risk of impact from a rising CCD (noting the large
plot on the right-hand (high likelihood) side of Fig. 11. estimated error of the CCD rise from pre-industrial times to the present;
Two groups of ocean regions at moderate risk impact from a rising Table 1). The southeast Pacific (ocean region 11), for example, has
CCD are: (A) those that have a high likelihood (i.e., have experienced an witnessed an estimated 69 ± 308 m of CCD rise since pre-industrial
above average rise in CCD since pre-industrial times) but low conse­ times and it’s geomorphic features plot in the moderate risk field for
quences (i.e., are predicted to exhibit a below average increase in per­ 10 out of 11 cases studied here (Fig. 12).
centage of seafloor submerged below a CCD that is 300 m higher than
present); and (B) low likelihood of submergence (i.e., those area that
have experienced a below average rise in CCD since pre-industrial times) 4.3. Impacts of acidification in the Arctic
but high consequences (i.e., are predicted to exhibit an above average
increase in percentage of seafloor submerged below a CCD that is 300 m The introduction of an acidified, mid-depth water mass around the
higher than present). Group A includes the northeast Atlantic, northwest Arctic continental margin could potentially have the greatest conse­
Atlantic, southeast Atlantic, southwest Atlantic and southern Indian quences for marine life of any ocean region. This is because ocean life in
Ocean (Ocean Regions 6, 8, 10, 12 and 14, respectively, in Fig. 11). the Arctic has not previously been exposed to conditions below the CCD;
Group B includes the central equatorial Pacific, eastern equatorial Pa­ and whether the Arctic benthos can adapt to life below the CCD is un­
cific, southeast Pacific and southwest Pacific (Ocean Regions 1, 4, 11 known. We can only speculate on what impact the introduction of an
and 13, respectively, in Fig. 11). All other Ocean Regions are in the low acidified, mid-depth water mass would have.
risk category. In their study of the benthic ecology of a Canadian glacial trough,
Lacharité et al. (2020) noted the significance of a boundary at 460 m
below which iceberg keel marks were absent; where the benthic fauna
4.2. Categories of geomorphic features most at risk was dominated by molluscs and polychaetes. In their review of Arctic
benthic fauna, Vedenin et al. (2021) identified a sublittoral boundary at
To assess the risk of categories of seafloor geomorphic features 450–800 m that the authors suggest indicates significant changes in
impacted by a rising CCD, the important factors are as for the ocean trophic conditions, particularly the flux of organic matter. However, the
regions (i.e., the amount of area that remains above the CCD and amount authors note that the specific environmental factors controlling organic
of area that will be submerged with a hypothetical 300 m rising of the matter flux in this depth zone are unknown and that further research is
CCD). A result of the present study is to clearly illustrate differences required.
between ocean regions in how the rising CCD will impact geomorphic Although some cold-water coral species could possibly occur in the
features (Fig. 6). Each category of geomorphic feature is impacted Arctic Ocean basin, there are no published records to confirm this. One
differently among the different ocean regions. Hence the risk of any (cold water) group of coral species is adapted to a water temperature
category of features becoming submerged below the rising CCD varies range of − 1 to 3 ◦ C but the occurrence of these species is apparently
significantly between ocean regions. This implies that efforts to conserve limited by geographic barriers (deep sea ridges and sills) as well as the
biodiversity associated with different categories of geomorphic feature available food supply (Buhl-Mortensen et al., 2015). It is interesting to
will need to consider each ocean region separately to define note that cold-water coral species are commonly found at depths

15
P.T. Harris et al. Marine Geology 463 (2023) 107121

between 200 and 1000 m (thus overlapping with the depth range of a from 0.09% to 6.09% within the US EEZ (lower 48 states), whilst that of
hypothetical intrusion of an acidified mid-depth water mass). Further Russia rises from 0.04 to 0.39%.
south along the Norwegian continental slope, it has been proposed by
Raddatz et al. (2016) that changes in water chemistry related to glacial
melting may have resulted in a hiatus in cold water coral reef growth Declaration of Competing Interest
between ~6.4 and 4.8 thousand years ago.
The authors declare that they have no known competing financial
5. Conclusions interests or personal relationships that could have appeared to influence
the work reported in this paper.
Based on a spatial analysis of seafloor geomorphology we have
assessed the seafloor area submerged below the rising CCD caused by Data availability
anthropogenic ocean acidification. We find that shoaling of the CCD
since the industrial revolution has submerged 12,432,096 km2 of ocean This study is based on the analysis of previously published spatial
floor (3.6% of total ocean area) below the CCD, making a total of 40.5% data as described in the methods section. Supplementary Tables 1 and 2
of ocean floor below the CCD at present. Uniform 300 m shoaling of the plus the derived GIS polygons for seafloor area located below the CCD
CCD across all ocean regions illustrates that the surface area of seafloor for different ocean regions and for different geomorphic feature types
submerged below the CCD rises by 14% relative to pre-industrial revo­ are stored at [Link].
lution times (total of 51% of ocean floor below the CCD). These results
are significant in the context of species adaptation to climate change, Acknowledgements
since there is now evidence from at least one area of the deep ocean that
the CCD is a biological barrier for some taxa (Simon-Lledó et al., 2023). The work described in this paper was produced with financial sup­
A relative risk assessment for future submergence of ocean floor port from the Norwegian Agency for Development Cooperation (Norad),
below the CCD has been carried assuming that the change in CCD from Ministry of Foreign Affairs.
pre-industrial times to the present is an indicator of likelihood and the
change in percentage of seafloor submerged below the CCD due to a
References
hypothetical 300 m rise in the CCD is an indicator of the consequences.
We find that the western equatorial Atlantic is at high risk and 9 other Albright, R., Hansson, L., Cooley, S.R., Gattuso, J.-P., Marshall, P., Marshall, N.,
Ocean Regions are at moderate risk. While noting the large error of the Fletcher, S., Haraldsson, G., Hoegh-Guldberg, O., 2023. Are we ready for ocean
acidification? A framework for assessing and advancing policy readiness. Environ.
estimated CCD rise from pre-industrial times to the present, overall
Res. Lett. 18 (4), 041001.
geomorphic features in the Atlantic Ocean and southern Indian Ocean Andersson, A.J., Mackenzie, F.T., 2012. Revisiting four scientific debates in ocean
appear to be at greater risk of impact from a rising CCD than Pacific and acidification research. Biogeosciences 9, 893–905.
other Indian Ocean regions. Archer, D., Kheshgi, H., Maier-Reimer, E., 1998. Dynamics of fossil fuel CO2
neutralization by marine CaCO3. Glob. Biogeochem. Cycles 12, 259–276.
Out of 20 categories of geomorphic feature type, we found that some Becker, J.J., Sandwell, D.T., Smith, W.H.F., Braud, J., Binder, B., Depner, J., Fabre, D.,
part of every feature category intersected with the lysocline and will be Factor, J., Ingalls, S., Kim, S.H., Ladner, R., Marks, K., Nelson, S., Pharaoh, A.,
(or is already) submerged below the CCD in at least one ocean region. Trimmer, R., Von Rosenberg, J., Wallace, G., Weatherall, P., 2009. Global
Bathymetry and Elevation Data at 30 Arc. Seconds Resolution: SRTM30_PLUS.
There is much regional variation between ocean regions predicted with Marine Geodesy. 32 (4), 355–371.
a CCD that is 300 m shallower than present. For seamounts we find that Biastoch, A., Treude, T., Rüpke, L.H., Riebesell, U., Roth, C., Burwicz, E.B., Park, W.,
in the North Pacific over 50% of seamount area will be below a CCD that Latif, M., Böning, C.W., Madec, G., Wallmann, K., 2011. Rising Arctic Ocean
temperatures cause gas hydrate destabilization and ocean acidification. Geophys.
is 300 m shallower than at present, whereas virtually no area of Res. Lett. 38 (8).
seamount is predicted to become submerged below in the eastern Broecker, W.S., 1991. The great ocean conveyor. Oceanography 4 (2), 79–89.
equatorial Atlantic (Fig. 6). The percent area of submarine canyons Buhl-Mortensen, L., Olafsdottir, S.H., Buhl-Mortensen, P., Burgos, J.M., Ragnarsson, S.A.,
2015. Distribution of nine cold-water coral species (Scleractinia and Gorgonacea) in
submerged below the CCD will increase by up to 17% in the southwest the cold temperate North Atlantic: effects of bathymetry and hydrography.
Atlantic (more than any other ocean region), whereas virtually no area Hydrobiologia 759 (1), 39–61.
of submarine canyon is predicted to become submerged below a CCD Costello, M.J., Cheung, A., De Hauwere, N., 2010. Surface area and the seabed area,
volume, depth, slope, and topographic variation for the world’s seas, oceans, and
that is 300 m shallower than at present in the northeast Atlantic or in the
countries. Environ. Sci. Technol. 44 (23), 8821–8828.
central Indian Ocean (Fig. 6). Regional variation in the areas of different El Bani Altuna, N., Rasmussen, T.L., Ezat, M.M., Vadakkepuliyambatta, S.,
geomorphic feature types submerged below the CCD implies that sepa­ Groeneveld, J., Greaves, M., 2021. Deglacial bottom water warming intensified
rate conservation priorities and strategies will be needed for each ocean Arctic methane seepage in the NW Barents Sea. Commun. Earth Environ. 2 (1), 188.
Fabry, V.J., Seibel, B.A., Feely, R.A., Orr, J.C., 2008. Impacts of ocean acidification on
region to ensure their adequate representation within future marine marine fauna and ecosystem processes. ICES J. Mar. Sci. 65 (3), 414–432.
protected area networks. Fischer, A., Bhakta, D., Macmillan-Lawler, M., Harris, P.T., 2019. Existing global marine
Nearly 1.3 million km2 of the Arctic continental shelf and slope is protected area network is not representative or comprehensive measured against
seafloor geomorphic features and benthic habitats. Ocean Coast. Manag. 167,
predicted to become submerged within a mid-depth, acidified water 176–187.
mass proposed by Terhaar et al. (2020) by the end of the 21st century. Galgani, F., Michela, A., Gérigny, O., Maes, T., Tambutté, E., Harris, P.T., 2022. Marine
Submarine terraces and glacial troughs incised into the continental shelf Litter, Plastic, and Microplastics on the Seafloor. Plastics and the Ocean. A. L.
Andrady. Wiley, pp. 151–197.
are the most impacted features. Ocean life in the Arctic has evolved in GRID-Arendal, 2020. Coastal and Offshore Permafrost in a Changing Arctic, Rapid
the absence of seafloor area below the CCD, and hence the formation of Response Assessment. [Link]
this acidified water mass may cause significant challenges for Arctic Halpern, B.S., Walbridge, S., Selkoe, K.A., Kappel, C.V., Micheli, F., D’Agrosa, C.,
Bruno, J.F., Casey, K.S., Ebert, C., Fox, H.E., Fujita, R., Heinemann, D., Lenihan, H.S.,
benthos. Madin, E.M.P., Perry, M.T., Selig, E.R., Spalding, M., Steneck, R., Watson, R., 2008.
Finally, we also find that the area of national Exclusive Economic A global map of human impact on marine ecosystems. Science 319, 948–952.
Zones submerged below the rising CCD exhibits extreme variability; Harris, P.T., 2012. Chapter 5, Surrogacy. In: Harris, P.T., Baker, E.K. (Eds.), Seafloor
Geomorphology as Benthic Habitat: GeoHab Atlas of Seafloor Geomorphic Features
with 300 m of CCD shoaling we find >12% increase in area submerged
and Benthic Habitats. Elsevier, Amsterdam, pp. 93–108.
below the CCD for 23 national EEZs, mainly of developing countries. The Harris, P.T., 2020. Chapter 3 - Anthropogenic threats to benthic habitats. In: Harris, P.T.,
five most impacted EEZ’s are Bermuda (UK), El Salvador, Nauru, Baker, E. (Eds.), Seafloor Geomorphology as Benthic Habitat (Second Edition).
Guatemala and Sri Lanka that will see >28% of their EEZ’s submerged Elsevier, pp. 35–61.
Harris, P.T., MacMillan-Lawler, M., 2017. Origin and characteristics of ocean basins. In:
below a rising CCD. In contrast the area of seabed projected to be sub­ Micallef, A., Krastel, S., Savini, A. (Eds.), Submarine Geomorphology. Springer
merged below a CCD that is 300 m more shallow than at present rises Geology. Springer, Cham, pp. 111–134.

16
P.T. Harris et al. Marine Geology 463 (2023) 107121

Harris, P.T., Whiteway, T., 2009. High seas marine protected areas: benthic Schmitz, W.J., McCartney, M.S., 1993. On the North Atlantic circulation. Rev. Geophys.
environmental conservation priorities from a GIS analysis of global ocean 31, 29–49.
biophysical data. Ocean Coast. Manag. 52, 22–38. Simon-Lledó, E., Amon, D.J., Bribiesca-Contreras, G., Cuvelier, D., Durden, J.M.,
Harris, P.T., MacMillan-Lawler, M., Rupp, J., Baker, E.K., 2014. Geomorphology of the Ramalho, S.P., Uhlenkott, K., Arbizu, P.M., Benoist, N., Copley, J., Dahlgren, T.G.,
oceans. Mar. Geol. 352, 4–24. Glover, A.G., Fleming, B., Horton, T., Ju, S.-J., Mejía-Saenz, A., McQuaid, K.,
Hofmann, G.E., Barry, J.P., Edmunds, P.J., Gates, R.D., Hutchins, D.A., Klinger, T., Pape, E., Park, C., Smith, C.R., Jones, D.O.B., 2023. Carbonate compensation depth
Sewell, M.A., 2010. The effect of ocean acidification on calcifying organisms in drives abyssal biogeography in the northeast Pacific. Nat. Ecol. Evol. [Link]
marine ecosystems: an organism-to-ecosystem perspective. Annu. Rev. Ecol. Evol. org/10.1038/s41559-023-02122-9.
Syst. 41 (1), 127–147. Sulpis, O., Boudreau, B.P., Mucci, A., Jenkins, C., Trossman, D.S., Arbic, B.K., Key, R.M.,
Honisch, B., Hemming, N.G., Archer, D., Siddall, M., McManus, J.F., 2009. Atmospheric 2018. Current CaCO3 dissolution at the seafloor caused by anthropogenic CO2. Proc.
carbon dioxide concentration across the mid-Pleistocene transition. Science 324, Natl. Acad. Sci. 115 (46), 11700.
1,551–1,554. Sulpis, O., Agrawal, P., Wolthers, M., Munhoven, G., Walker, M., Middelburg, J.J., 2022.
Johnson, G.C., 2008. Quantifying Antarctic bottom water and North Atlantic deep water Aragonite dissolution protects calcite at the seafloor. Nat. Commun. 13 (1), 1104.
volumes. J. Geophys. Res. Oceans 113 (C5). [Link] Sweetman, A.K., Thurber, A.R., Smith, C.R., Levin, L.A., Mora, C., Wei, C.-L., Gooday, A.
2007JC004477. J., Jones, D.O.B., Rex, M., Yasuhara, M., Ingels, J., Ruhl, H.A., Frieder, C.A.,
Jutterström, S., Anderson, L.G., 2005. The saturation of calcite and aragonite in the Danovaro, R., Würzberg, L., Baco, A., Grupe, B.M., Pasulka, A., Meyer, K.S.,
Arctic Ocean. Mar. Chem. 94 (1), 101–110. Dunlop, K.M., Henry, L.-A., Roberts, J.M., 2017. Major impacts of climate change on
Kump, L., Bralower, T., Ridgwell, A., 2009. Ocean acidification in deep time. deep-sea benthic ecosystems. Elementa Sci. Anthropocene 5.
Oceanography 22 (4), 94–107. Terhaar, J., Kwiatkowski, L., Bopp, L., 2020. Emergent constraint on Arctic Ocean
Lacharité, M., Brown, C.J., Normandeau, A., Todd, B.J., 2020. Chapter 41 - Geomorphic acidification in the twenty-first century. Nature 582 (7812), 379–383.
features and benthos in a deep glacial trough in Atlantic Canada. In: Harris, P.T., Thomas, E., 1998. Biogeography of the later Paleocene benthic foraminiferal extinction.
Baker, E. (Eds.), Seafloor Geomorphology as Benthic Habitat (Second Edition). In: Aubry, M.-P., Lucas, S.G., Berggren, W.A. (Eds.), Late Paleocene-Early Eocene
Elsevier, pp. 691–704. Climatic and Biotic Events in the Marine and Terrestrial Records. Columbia
Luo, Y., Boudreau, B.P., Mucci, A., 2016. Disparate acidification and calcium carbonate University Press, New York, pp. 214–235.
desaturation of deep and shallow waters of the Arctic Ocean. Nat. Commun. 7 (1), Tozer, B., Sandwell, D.T., Smith, W.H.F., Olson, C., Beale, J.R., Wessel, P., 2019. Global
12821. bathymetry and topography at 15 arc Sec: SRTM15+. Earth Space Sci. 6 (10),
MacMillan-Lawler, M., Harris, P.T., 2016. Multivariate Classification of Seamount 1847–1864.
Morphology: Assessing Seamount Morphotypes in Relation to Marine Jurisdictions United Nations, 2023a. Measure and Report Ocean Acidification - Sustainable
and Bioregions. Ocean Solutions, Earth Solutions. D. J. Wright, Los Angeles, Esri. Development Goal 14.3.1 Indicator Methodology. [Link]
[Link] ⋅. ips/measure-and-report-ocean-acidification-sustainable-development-goal-1431-
Manno, C., Sandrini, S., Tositti, L., Accornero, A., 2007. First stages of degradation of indicator.
Limacina helicina shells observed above the aragonite chemical lysocline in Terra United Nations, 2023b. UN delegates reach historic agreement on protecting marine
Nova Bay (Antarctica). J. Mar. Syst. 68 (1–2), 91–102. biodiversity in international waters. UN News, 5 March, 2023. [Link]
Milliman, J.D., Troy, P.J., Balch, W.M., Adams, A.K., Li, Y.H., Mackenzie, F.T., 1999. org/en/story/2023/03/1134157.
Biologically mediated dissolution of calcium carbonate above the chemical US EPA, 2023. Global Greenhouse Gas Emissions Data. accessed June, 2023,. https
lysocline? Deep-Sea Res. I Oceanogr. Res. Pap. 46 (10), 1653–1669. ://[Link]/ghgemissions/global-greenhouse-gas-emissions-data.
Qi, D., Ouyang, Z., Chen, L., Wu, Y., Lei, R., Chen, B., Feely, R.A., Anderson, L.G., Vedenin, A., Galkin, S., Mironov, A.N., Gebruk, A., 2021. Vertical zonation of the
Zhong, W., Lin, H., Polukhin, A., Zhang, Y., Zhang, Y., Bi, H., Lin, X., Luo, Y., Siberian Arctic benthos: bathymetric boundaries from coastal shoals to deep-sea
Zhuang, Y., He, J., Chen, J., Cai, W.-J., 2022. Climate change drives rapid decadal Central Arctic. PeerJ 9, e11640. [Link]
acidification in the Arctic Ocean from 1994 to 2020. Science 377 (6614), Widdicombe, S., Spicer, J.I., 2008. Predicting the impact of ocean acidification on
1544–1550. benthic biodiversity: what can animal physiology tell us? J. Exp. Mar. Biol. Ecol. 366
Raddatz, J., Liebetrau, V., Trotter, J., Rüggeberg, A., Flögel, S., Dullo, W.-C., (1), 187–197.
Eisenhauer, A., Voigt, S., McCulloch, M., 2016. Environmental constraints on Woosley, R.J., 2016. Carbonate compensation depth. In: White, W.M. (Ed.),
Holocene cold-water coral reef growth off Norway: insights from a multiproxy Encyclopedia of Geochemistry: A Comprehensive Reference Source on the Chemistry
approach. Paleoceanography 31 (10), 1350–1367. of the Earth. Springer International Publishing, Cham, pp. 1–2.
Sayre, R.G., Wright, D.J., Breyer, S.P., Butler, K.A., Van Graafeiland, K., Costello, M.J., Zachos, J.C., Röhl, U., Schellenberg, S.A., Sluijs, A., Hodell, D.A., Kelly, D.C., Thomas, E.,
Harris, P.T., Goodin, K.L., Guinotte, J.M., Basher, Z., Kavanaugh, M.T., Halpin, P.N., Nicolo, M., Raffi, M., Lourens, L.J., 2005. Rapid acidification of the ocean during the
Monaco, M.E., Cressie, N., Aniello, P., Frye, C.E., Stephens, D., 2017. A three- Paleocene-Eocene Thermal Maximum. Science 308, 1,611–1,615.
dimensional mapping of the ocean based on environmental data. Oceanography 30
(1), 90–103.

17