Constantino&tinto 2023

Download as pdf or txt
Download as pdf or txt
You are on page 1of 9

RESEARCH LETTER Cook Ice Shelf and Ninnis Glacier Tongue Bathymetry From

10.1029/2023GL103815
Inversion of Operation Ice Bridge Airborne Gravity Data
Key Points:
Renata R. Constantino1 and Kirsty J. Tinto1
• H igh resolution bathymetry model
of Cook Ice Shelf, Ninnis Glacier 1
Lamont-Doherty Earth Observatory, Columbia University, Palisades, NY, USA
Tongue, and surrounding open ocean
from airborne gravity inversion
• New bathymetry model improves
the understanding of water pathways Abstract The seafloor depths under the Cook Ice Shelf and Ninnis Glacier Tongue have not been directly
between the ice shelves and the measured, despite their importance for understanding ocean circulation and ice shelf change. We model the
continental shelf edge bathymetry underneath the floating ice and surrounding ocean using airborne gravity data. Our model is
• Transit flight gravity anomalies
suggest relocation of the edge of the constrained by few ship-based seafloor measurements near the ice front and by ice-base measurements over
continental shelf northwards of the areas of grounded ice from radar data. Localized basins (∼1,400 m deep) are found beneath both ice shelves.
currently attributed position The shallowest modeled bathymetry (∼200 m) represents the offshore extension of Cape Freshfield. Near the
grounding line, seafloor depths are found to be deeper than the observed depth of the modified Circumpolar
Supporting Information: Deep water in the region (<350 m), key factor for basal melt analyses. From transit flight gravity anomalies,
Supporting Information may be found in we suggest the relocation of the mapped edge of the continental shelf and a narrowing of the Cook Shelf
the online version of this article.
Depression.

Correspondence to: Plain Language Summary The knowledge of how deep the ocean floor is under the floating
R. R. Constantino, ice shelves that connect to grounded ice sheets, is crucial for understanding how ocean water circulates and
constantino@ldeo.columbia.edu
interacts with the overlying ice. We present a new bathymetric model of the seafloor beneath two ice shelves
located in East Antarctica: Cook Ice Shelf and Ninnis Glacier Tongue. Both ice shelves are inaccessible to
Citation:
ships due to heavy sea ice conditions, so the data used in our model were collected from airborne surveys. Our
Constantino, R. R., & Tinto, K. J. (2023).
Cook ice shelf and Ninnis Glacier
bathymetry model shows new information on the depth and shape of the seafloor that will help understanding
Tongue bathymetry from inversion of the ocean circulation in the area and how this might impact ice thickness changes.
operation ice bridge airborne gravity
data. Geophysical Research Letters,
50, e2023GL103815. https://doi.
org/10.1029/2023GL103815
1. Introduction
Ice which is grounded well below sea level in the marine basins of Antarctica with an inland-sloping bed is poten-
Received 22 MAR 2023
tially vulnerable to marine ice sheet instability (Miles et al., 2018; Schoof, 2007). Floating ice shelves are crucial
Accepted 19 MAY 2023
to the stability of ice streams and outlet glaciers that drain marine basins because they can exert an important
buttressing effect, slowing the flow of grounded ice toward the coast (Fürst et al., 2016).

The Cook Ice Shelf (CIS) and Ninnis Glacier Tongue (NGT) are located on the George V Coast (Figure 1) and
buttresses ice sheets draining parts of the Wilkes Basin into the Southern Pacific Ocean (Silvano et al., 2016). The
Wilkes Basin is the largest region of bed topography below sea level in EA and contains 3–4 m sea level equiva-
lent of ice grounded below sea level (Mengel & Levermann, 2014). The nomenclature and definition for the CIS,
NGT differ in the literature. Here we use the names derived from the SCAR Composite Gazetteer of Antarctica
(Secretariat SCAR, 1992, updated 2022).

CIS and NGT present a complex, long term view of surface elevation change, as different periods have been
analyzed and different behaviors have been observed. Ice thinning has been reported in studies that use time
series prior to 2007 (e.g., Davis et al., 2005: 1992–2003; Pritchard et al., 2009: 2003–2007; Paolo et al., 2015:
1994–2003). For shorter period analyses (e.g., McMillan et al., 2014: elevation change between 2010 and 2013),
no significant changes were found over CIS, NGT and their outlet glaciers. Both thickening and thinning have
been reported between 2013 and 2019 (Smith et al., 2020). On the most recent timescale a consistent thickening
© 2023. The Authors. signal been observed (Hogg et al., 2020: 2017–2020).
This is an open access article under
the terms of the Creative Commons Similarly, a long term velocity increase was reported in Miles et al., 2018, who used satellite imagery from 1947
Attribution-NonCommercial-NoDerivs
License, which permits use and to 2017 to produce an ice front position and ice velocity from 1973 to 2017, and in Rignot et al., 2019, who
distribution in any medium, provided the suggested mass loss in both CIS and NGT between 1979 and 2017. No observable velocity change was reported
original work is properly cited, the use is over a shorter period (Rignot, 2006, p. 1996 and 2000), suggesting that velocity changes in velocity have been
non-commercial and no modifications or
adaptations are made. intermittent as well as elevation changes.

CONSTANTINO AND TINTO 1 of 9


19448007, 2023, 11, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2023GL103815 by CAPES, Wiley Online Library on [12/06/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Geophysical Research Letters 10.1029/2023GL103815

Figure 1. Free-air gravity anomalies from airborne ice-shelf (A to G) and transit flights (T1 to T6) gravity surveys. Gray
scale shows ice flow velocity (Mouginot et al., 2017). The Cook Ice Shelf and Ninnis Glacier Tongue marked by white
parallel line texture (Fretwell et al., 2013). Main ice streams marked as 1, 2 and 3. The blue dashed line represents the extent
of the final grid shown in the results. Location for bathymetry data outboard from ice shelves used in this work as constraints
are shown in red. Overview map with West Antarctica, East Antarctica with study area in red’.

The bathymetry beneath ice shelves exerts a fundamental control on the ice sheet stability, as the shape of the
water cavity is a first-order control on sub-ice-shelf currents, including the flux of warm, deep-ocean water to
the base of the ice shelves and the grounding line (Jacobs et al., 2011). Cavity shape is also an important factor
controlling the rate of melting close to the grounding line (Milillo et al., 2019; Schoof, 2007). The presence of
warm modified Circumpolar Deep Water (mCDW) beneath CIS and NGT has been suggested as a driver for
increase in ice velocity (Rignot et al., 2019) and ice shelf retreat (Miles et al., 2018). A westward advection of
warm mCDW from the Cook Shelf Depression (CSD) on the continental shelf in front of both ice shelves has
been proposed from numerical modeling (Kusahara et al., 2017), but it is not known if there are any connecting
troughs from this depression that could facilitate the delivery of warm mCDW intrusions toward CIS and NGT
outlet grounding line.

Current maps of regional bathymetry are based on interpolation of very sparse ship constraints (e.g., Dorschel
et al., 2022; Morlighem et al., 2020). The only available data are close to NGT, while none is available at the
CIS (Figure 1). One way to overcome the lack of direct bathymetry measurements is applying the gravity inver-
sion method. The shape of the seafloor influences the earth's gravity field because of the contrast in density
between rock and water. The resulting gravity anomalies can be used to model the bathymetry. These models are
improved by incorporating directly measured soundings or regional depth averages as constraints. The model is
also constrained by ice thickness radar data where ice is grounded. We present a new bathymetric model from a
high-resolution aerogravity inversion constrained by the available seafloor depth measurements. Our goal is to
identify the water pathways estimated from our model to help understand the ocean/ice interaction beneath the
studied ice shelves.

CONSTANTINO AND TINTO 2 of 9


19448007, 2023, 11, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2023GL103815 by CAPES, Wiley Online Library on [12/06/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Geophysical Research Letters 10.1029/2023GL103815

2. Data
2.1. Gravity Anomalies

Between 2009 and 2019, the NASA Operation IceBridge (OIB) Mission collected airborne remote sensing meas-
urements over Earth's polar regions (MacGregor et al., 2021). For this study, we use data from surveys flights of
the 2019 campaign. These survey flights were flown at a height of ∼500 m above the surface at an average speed
of 275 kn (142 m s −1) on NASA's Gulfstream-V aircraft. The gravity measurements were obtained using an iMar
inertial navigation system (Becker, 2016; iMar Navigation, 2012) as a strapdown gravimeter reporting free air
gravity anomalies with a half-wavelength of ∼5 km (Tinto, 2018). We also use gravity measurements obtained
during the transit flights which are flown at higher elevation (up to 12,600 m) and have coarser resolution than
survey data, and have been inverted for bathymetry with wavelengths of 20 km.

2.2. Constraints

The ice surface and ice thickness (Figures S1a and S1b in Supporting Information S1) were measured by the
Multi-channel Coherent Radar Depth Sounder with approximately 10 m accuracy (Leuschen, 2012). For profile
D where radar was partially unavailable over the grounded ice, the ice surface elevation was measured with an
accuracy of ∼0.1 m using the Airborne Topographic Mapper lidar (Krabill et al., 2002), and bed was modeled
from the gravity anomaly.
Ship-based bathymetry depths acquired in 1957 and 1958 by the Vernadsky Institute for Geochemistry and
Analytical Chemistry, Russia (available in Ryan et al. (2009)) are used to constrain the bathymetry model where
the ship track and airborne data are coincident (Figure 1). These data are also included in the final model grid.
Bathymetry depths near the continental shelf break from the International Bathymetric Chart of the Southern
Ocean grid (Dorschel et al., 2022) in areas where data have been directly measured were used to constrain the
transit flight inversion models (Constraints are shown in Figure S2 in Supporting Information S1).

3. Method
3.1. Gravity Inversion

We model the bathymetry beneath the Cook and Ninnis ice shelves and surrounding open ocean by inverting the
gravity anomaly. The inversion was performed using the Geosoft GM-SYS 2D software package. The gravity
response is calculated based on the method of Talwani et al., 1959, while the inversion routine utilizes an imple-
mentation of the Marqardt inversion algorithm (Marqardt, 1963) (Text S1 in Supporting Information S1).
Beneath the ice shelves and surrounding ocean, the inversion was performed along the seven survey profiles
with gravity measurements (Figure 1). For each profile, we first created a forward gravity model by assuming
three main bodies with different densities (ice, water and crust). Radar ice thickness and surface altimetry were
compared to identify the location of the grounding line along track using Depoorter et al. (2013) as guide. Gravity
inversion for bathymetry was performed along. A density of 0.917 g/cm 3 was assigned to the ice and 1.03 g/cm 3
was assigned to water column. Beneath the ice and/or the water column, an initial density value of 2.67 g/cm 3
was used for the crust.
First, forward models were created for profiles A and B, where ice is grounded and bed topography is known from
radar over most of their extent, and used to identify the crustal density. We started with rock as a constant density
block of 2.67 g/cm 3. To minimize the residuals between the observed and calculated gravity anomalies, we then
split the bedrock into more four blocks. With no constraints on the nature of the boundaries between these blocks,
they were defined as simple vertical structures (Figures S2–S5 7in Supporting Information S1) and extended
orthogonally to the coast to the offshore profiles (Figure 2a). The density variation between the blocks is centesi-
mal (2.655 g/cm 3, 2.678 g/cm 3, 2.691 g/cm 3) except for one block of denser material (3.0 g/cm3) inferred at 2 km
depth near NGT (Figure 2a). This denser block was initially identified from the 2D gravity models along profiles
A and B (Figures S2 and S3 in Supporting Information S1) and then extended to profiles C and D (Figures S4 and
S5 in Supporting Information S1). The boundary of this block of denser material is most aligned with the steep
edge of a broad positive magnetic anomaly (amplitudes close to 300 nT, ADMAP2, Golynsky et al., 2018, Figure
S7 in Supporting Information S1), which was interpreted from aeromagnetic maps as marking mafic Jurassic
intrusions (Damaske et al., 2003). From the gravity anomaly interpretation, the boundary of this denser body

CONSTANTINO AND TINTO 3 of 9


19448007, 2023, 11, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2023GL103815 by CAPES, Wiley Online Library on [12/06/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Geophysical Research Letters 10.1029/2023GL103815

Figure 2. (a) Bedmachine (Morlighem et al., 2020). (b) Cook Ice Shelf (CIS) and Ninnis Tongue bathymetry model seaward of grounding line. Model includes
bathymetry obtained from gravity inversion and offshore depth constraints (Figure S2 in Supporting Information S1). Onshore bed topography from Bedmachine
bedrock (Morlighem et al., 2020). The CIS and Ninnis Glacier Tongue marked by white parallel line texture (Fretwell et al., 2013). Main ice streams marked as 1, 2, and
3. Profiles A to G are shown as thin red lines. Red line for available ship bathymetry and red circles for Conductivity, Temperature, and Depth depths.

extends beyond the grounding line, where the magnetic anomaly is of low amplitude (−150 nT) but the presence
of dense rock is inferred from ice-base constraints. After the 2D forward models were built, we performed an
inversion of the gravity residuals to model the bathymetry where the ice is floating.
For the surrounding open ocean, the inversion was performed along six profiles with transit flight gravity meas-
urements (Figure 1). To remove the effect of deep structure, the anomaly was detrended using a first order poly-
nomial fit (Figure S8b in Supporting Information S1). A 20 km wavelength low-pass filter was applied to the
anomaly before inverting for bathymetry (Figure S8c in Supporting Information S1). For each profile we created
a forward gravity model by assuming two main bodies with different densities, water (1.03 g/cm 3) and crust
(2.678 g/cm 3 for T1, 2.691 g/cm 3 for T2 and 2.67 g/cm 3 for the other profiles, consistent with densities estab-
lished for the ice shelf profiles). The inverted bathymetry values from profiles E, F and G were used as constraints
in the inversion where they connect with transit profiles T3, T2, and T1, respectively.

3.2. Model Uncertainty

The uncertainties to be considered due to instrumentation are concerning the radar and the gravimeter. An uncer-
tainty of ±10 m in the radar-derived bed elevation is taken into account (MacGregor et al., 2021). The uncertainty

CONSTANTINO AND TINTO 4 of 9


19448007, 2023, 11, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2023GL103815 by CAPES, Wiley Online Library on [12/06/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Geophysical Research Letters 10.1029/2023GL103815

due to the gravimeter accuracy is estimated by calculating the standard error of the differences between gravity
measurements at the crossover points between lines within this survey, which is 2.7 mGal. To quantify this value
in meters, a simple Bouguer slab correction is applied to the residual considering two scenarios: with the lowest
(1.625 g/cm3) and highest (1.97 g/cm3) density contrast between rock and water, which gives values of ±39 and
±32 m respectively. The uncertainty from variations in geology is estimated by the Root Mean Square Error
(RMSE) of the gravity residuals where topography is known from radar suggesting that these residuals are due to
small-scale topography and density variations.

The RMSE is 5.1 mGal when considering a single density for the crust, and 1.88 mGal when considering variable
densities along profiles A and B. Applying these uncertainties to the bathymetry model adds ±74 m (for single
density) or ±27 m (for variable density) of uncertainty, showing that a 47 m of improvement in accuracy can be
achieved by applying the density model. With the radar data uncertainty of ±10m, this gives a total uncertainty
ranging from ±69 to ±123 m. The uncertainty of the transit flights is in the same range as the surveyed lines.

4. Results
Our new bathymetry beneath the ice shelves is shown in Figure 2b. In order to get a better representation of
the seafloor over the entire area, we have gridded the bathymetry obtained along the profiles (Figure S9a in
Supporting Information S1) with a cell size of 1 km and interpolated between them using minimum curvature.
We have also included the ship and Conductivity, Temperature, and Depth (CTD) depth constraints (Figure S2 in
Supporting Information S1). To avoid transitions at the grounding line, the bathymetry grid was interpolated with
Bedmachine Antarctica (Version 2) (Morlighem et al., 2020) topography where ice is grounded. To highlight the
increased resolution offered by the new data and the contribution of our model not only beneath the ice shelves but
also in the surrounding areas, we show a comparison between Bedmachine and our final bathymetry (Figure S9b
in Supporting Information S1). Our model (Figure 2b) reveals localized basins with water depths of ∼1,400 m
beneath both CIS and NGT, offshore of ice streams 1, 2, and 3. Outlet glacier 2, flows through a 1,300 m deep
trough onshore which the new bathymetry model shows continuing offshore at similar depths. At the grounding
line a distinct ridge with 500 m relief detected by the radar and also shown in the gravity anomaly (Figure S10 in
Supporting Information S1) divides the onshore and offshore parts of the trough, and likely exerts strong control
on the present-day grounding line position. The new model also shows that the deep basin constrained by ship
bathymetry continues farther in both eastward and westward directions than previously known, and could be an
important path for ocean circulation. At the surrounding open ocean (Figure 3), the seafloor is deeper in front of
Cape Freshfield (CF) (∼1,000 m) and shallows eastward (∼400–500 m) forming a ridge in front of the CIS that
could act as a barrier for local circulation. Where bathymetry deepens from ∼700 m to more than 1,000 m over an
horizontal distance of ∼40 km, it likely shows the transition between the continental shelf break (Amblas, 2018)
and the continental slope.

The water column thickness (Figure S11 in Supporting Information S1) estimated from our bathymetry model
and the radar ice base measurements (Figure S1c in Supporting Information S1) shows a water column beneath
most of CIS (∼300 m) and NGT (∼100 m), which results mainly from the thick ice (>500 m) (Figure S1b in
Supporting Information S1). Beneath CIS, a thick water layer under the D and E profile intersection (∼1,300 m)
and along profile F (∼750 m) represent areas of possible deep water access toward the grounding line. The water
column is thinned by a shallow ridge (∼200 m deep) in the offshore area in front of CF.

5. Discussion
5.1. Bathymetry and Water Pathways

A westward advection of warm mCDW from the CSD (Figure 3) on to the continental shelf offshore from CIS
has been proposed as a key component of the local oceanography (Kusahara et al., 2017). The lack of seafloor
depth data between this depression and Cook and Ninnis outlets has limited the understanding of the potential for
transport of the warm mCDW to these ice shelves (Miles et al., 2018).

The shape of the CSD itself is unknown and it is constrained by only two depth values of ∼1,300 m obtained
from isolated soundings (Figure 3) with Bedmachine (Morlighem et al., 2020). Gravity data from OIB transit
flights show no evidence of such a depression beneath the eastern flight lines (T1 and T2), but do not preclude

CONSTANTINO AND TINTO 5 of 9


19448007, 2023, 11, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2023GL103815 by CAPES, Wiley Online Library on [12/06/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Geophysical Research Letters 10.1029/2023GL103815

Figure 3. Bedrock topography derived from gravity inversion and from Bedmachine (Morlighem et al., 2020) everywhere
else with transparency. Free-air gravity anomaly from airborne gravity surveys (see Figure 1) and from transit flights. Circles
represent Conductivity, Temperature, and Depth stations at Ninnis Polynya (NP, pink outline) and Mertz Polynya (MP, white
outline). Red-filled circles show where modified Circumpolar Deep Water was observed (Moreau et al., 2019). Triangle
represents oceanographic observation from Gordon and Tchernia (1972) and black crosses represent isolated sounding data
(Dorschel et al., 2022). Cook Ice Shelf (CIS), Ninnis Glacier Tongue (NGT), Cape Freshfield, Cook Shelf Depression (CSD),
Mertz Glacier. The CIS and NGT marked by diagonal hatching (Fretwell et al., 2013). Light pink arrows represent possible
pathways connecting CIS with CSD, NGT with CSD and CSD with the continental shelf edge. Thin gray lines show the
inferred crustal density boundaries (Figure S7 in Supporting Information S1). The brown line marks the continental shelf
edge (Amblas, 2018) and black line with circles shows the proposed relocation. Possible water density barrier marked by the
dotted purple line. Potential morphological barrier of shallower bathymetry (∼500 m) marked by the yellow arrows.

the existence of a deeper seafloor feature on the western side that could connect NGT and CIS to the continental
shelf edge (Figure 3). From the inverted bathymetry along profiles T1 and T2, we can see that the seafloor is
deeper (∼1,000 m) along a 100 km section of T1, and shallows toward T2 (∼700 m) (Figure 3), constraining the
location of a potential deep water pathway across the continental shelf. There is no gravity data over the CSD,
but based on our results we suggest that there is a deeper pathway between T1 and T2 that does not necessarily
comprise the rounded shape depression as shown in the published maps (e.g., Dorschel et al., 2022), which is
result of interpolation. To better understand this depression, more ship measurements are needed.

The lack of bathymetric constraints in the area of the CSD leads to uncertainty even in the position of the conti-
nental shelf break, which has been inferred from data interpolation (Amblas, 2018). From the transit flights, we
observed a significant change in the gravity anomaly that is repeated on every profile: the anomaly drops ∼4
mGal (from ∼66 to ∼70 mGal), (open black circles, Figure 3), which we interpret as the continental shelf edge,
and suddenly decreases to ∼90 mGal, marking the continental shelf slope at a location 27 km north of the previ-
ously assumed position (Figure 3).

5.2. Oceanographic Implications

Near CIS, the closest available oceanography data including temperature and salinity variations with depth were
collected during the International Geophysical Year (1957–1958) by the Russian R/V Ob (Figure 3, site A),

CONSTANTINO AND TINTO 6 of 9


19448007, 2023, 11, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2023GL103815 by CAPES, Wiley Online Library on [12/06/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Geophysical Research Letters 10.1029/2023GL103815

where no mCDW was observed (Gordon & Tchernia, 1972). Near NGT, Moreau et al. (2019) interpreted the
presence of mCDW between 100 and 350 m deep in CTD samples collected at the Mertz Glacier (North of
67°S and between 149° and 150°E, Figure 3), whereas at CTD stations South of 67°S, no mCDW was observed.
Between the deep basins in front of NGT and the Mertz Polynya (MP), where no ship data are available (Figure
S12 in Supporting Information S1) our new bathymetry model suggests that these deep basins are connected by a
deep trough at 145°E, 67.2°S, with no indication that East-West water flow is being blocked by a morphological
feature on the seafloor. For stations South of 67°S, there is no indication of mCDW at the same depths where
it has been observed to the north (Moreau et al., 2019). In order to meet these observations, a barrier is needed,
which could be either a seafloor feature or a barrier in the water column (e.g., Planchat et al., 2022). Although
there is a shallower bathymetry (∼500 m) at this latitude (yellow arrows, Figure 3), which is covered by ship
data (Figure S12 in Supporting Information S1), it is still deeper than the mCDW depths found for the area. At
the deep basins of NGT and MP, the shallower, warmer and lighter mCDW flow may be being blocked by dense
shelf water (e.g., Moreau et al., 2019; Williams et al., 2010). The presence of dense water can be a barrier to
flow even in the absence of topographic obstacles. Detailed knowledge of such a density water barrier (purple
dotted-line, Figure 3) is critical for ocean models. To better understand the mCDW pathways in the area, more
data is desirable.

The cavity shape near the grounding line and beneath the ice shelf is a key factor when analyzing the basal melt
rates (Rignot & Jacobs, 2002). Our model shows that over most of its extent, except near CF and in front of
Ninnis Polynya, the ice sheet is grounded at ∼400 m below sea level, deeper than mCDW depths (∼100–400 m)
reported nearby over Mertz and NGT (Moreau et al., 2019; Williams et al., 2010). Any mCDW intrusions beneath
CIS and NGT, are likely to influence ocean/ice shelf interactions. From our bathymetry model, the shallowest
seafloor depth observed (∼200 m) is interpreted as the offshore extension of CF. The water column is thinned by
this shallow ridge and may present an obstacle to the ocean circulation between CIS and NGT. In order to fully
understand the processes driving changes at CIS and NGT and the influence of bathymetry on ice sheet history
more extensive studies of ocean properties are required, especially at CIS where no CTD measurements have yet
been conducted.

5.3. Geological Interpretation

With few rock outcrops and in the absence of onshore geological data near the grounding line, the offshore crus-
tal density variation boundaries inferred from this study are difficult to interpret. The models suggest relatively
minor variations in crustal density which could be accounted for by changes in bedrock geology, or denser crustal
intrusion within the crust.

We have modeled a dense body of 3.0 g/cm 3 from the seafloor to 1.5 km depth along profiles C and D (Figures
S5 and S6 in Supporting Information S1). The presence of this dense structure along both profiles is supported by
two main points: the constraint of the measured ice-base overlying it, and the high amplitude magnetic anomaly
observed in the ADMAP2 grid (Figure S7 in Supporting Information S1). A dense body was also identified from
the gravity models beneath NGT along profiles A and B (Figures S3 and S4 in Supporting Information S1), where
the magnetic anomaly is of low amplitude (Figure S7 in Supporting Information S1). Exposures of high-density
Jurassic Ferrar dolorite have been reported in the vicinity of high-amplitude magnetic signals further along the
coast (Damaske et al., 2003) and it is likely that similar lithologies are responsible for the gravity and magnetic
anomalies observed here. The volume of the dense material is consistent between profiles that underlie both
the low (Profile A—Figure S3 in Supporting Information S1 and Profile B—Figure S4 in Supporting Informa-
tion S1) and high magnetic amplitude regions (Profile C—Figure S5 in Supporting Information S1 and Profile
D—Figure S6 in Supporting Information S1).

6. Conclusion
We present a new bathymetry model for the CIS, NGT and the surrounding ocean using airborne gravity data.
Knowledge of the seafloor under these ice shelves is crucial to understand the processes driving ice changes in
the recent past and future projections along EA. The new bathymetry model has a higher resolution than previ-
ously available and can be used for future oceanographic models aiming to investigate the Circumpolar Deep
Water pathways. We use gravity anomalies observed in transit flights across the continental shelf to propose the

CONSTANTINO AND TINTO 7 of 9


19448007, 2023, 11, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2023GL103815 by CAPES, Wiley Online Library on [12/06/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Geophysical Research Letters 10.1029/2023GL103815

relocation of the continental shelf edge ∼27 km north of the currently attributed position. Although coverage by
these lines is sparse, they do not support the existence of a major mid-shelf bathymetric low beneath the transit
lines with the shape that is currently presented on available maps for bathymetry, suggesting that the CSD may
be of different extent than currently drawn, which changes the perspective of pathways between the continental
shelf and the ice shelves.

Data Availability Statement


Data from Operation IceBridge are available from the National Snow and Ice Data Center at http://nsidc.org/
data/icebridge. Bathymetry model data from Operation IceBridge are archived at the National Snow and Ice Data
Center (NSDISC) as data products IGBTH3 (gridded) and IGBTH4 (line data). See https://nsidc.org/data/igbth3
and https://nsidc.org/data/igbth4.

Acknowledgments References
This work was funded by NASA
NNX16AJ65G. We thank GEBCO Amblas, D. (2018). Antarctic continental shelf break (shapefile). Scott Polar Research Institute, University of Cambridge. https://doi.org/10.1594/
Compilation Group (2022) GEBCO_2022 PANGAEA.890863
Grid (https://doi.org/10.5285/e0f0bb80- Becker, D. (2016). Advanced calibration methods for strapdown airborne gravimetry. (PhD thesis), Technische Universität Darmstadt. Retrieved
ab44-2739-e053-6c86abc0289c). We from http://tuprints.ulb.tu-darmstadt.de/5691/
appreciate the helpful conversations with Damaske, D., Ferraccioli, F., & Bozzo, E. (2003). Aeromagnetic anomaly investigations along the Antarctic coast between Yule Bay and Mertz
Dave Porter and Arnold Gordon, and their Glacier. Terra Antartica, 10, 85–96.
discussion contributions. Davis, C. H., Li, Y., McConnell, J. R., Frey, M. M., & Hanna, E. (2005). Snowfall-driven growth in East Antarctic ice sheet mitigates recent
sea-level rise. Science, 308(5730), 1898–1901. https://doi.org/10.1126/science.111066
Depoorter, M. A., Bamber, J. L., Griggs, J., Lenaerts, J. T. M., Ligtenberg, S. R. M., van den Broeke, M. R., & Moholdt, G. (2013). Antarctic
grounding line. PANGAEA. https://doi.org/10.1594/PANGAEA.819150
Dorschel, B., Hehemann, L., Viquerat, S., Warnke, F., Dreutter, S., Tenberge, Y. S., et al. (2022). The international bathymetric chart of the
southern ocean version 2. Scientific Data, 9(1), 1–13. https://doi.org/10.1038/S51597-022-01366-7
Fretwell, P., Pritchard, H. D., Vaughan, D. G., Bamber, J. L., Barrand, N. E., Bell, R., et al. (2013). Bedmap2: Improved ice bed, surface and
thickness data sets for Antarctica. The Cryosphere, 7(1), 375–393. https://doi.org/10.5194/tc-7-375-2013
Fürst, J. J., Durand, G., Gillet-Chaulet, F., Tavard, L., Rankl, M., Braun, M., & Gagliardini, O. (2016). The safety band of Antarctic ice shelves.
Nature Climate Change, 6(5), 479–482. https://doi.org/10.1038/nclimate2912
Golynsky, A. V., Ferraccioli, F., Hong, J. K., Golynsky, D. A., von Frese, R. R. B., Young, D. A., et al. (2018). New magnetic anomaly map of
the Antarctic. Geophysical Research Letters, 45(13), 6437–6449. https://doi.org/10.1029/2018GL078153
Gordon, A. L., & Tchernia, P. L. (1972). Waters of the continental margin off Adélie coast, Antarctica. Antarctica oceanology II: The Australian–
New Zealand sector, 19, 59–69. https://doi.org/10.1029/AR019p0059
Hogg, A. E., Gilbert, L., Shepherd, A., Muir, A. S., & McMillan, M. (2020). Extending the record of Antarctic ice shelf thickness change, from
1992 to 2017. Advances in Space Research, 68(2), 724–731. https://doi.org/10.1016/j.asr.2020.05.030
Jacobs, S. S., Jenkins, A., Giulivi, C. F., & Dutrieux, P. (2011). Stronger ocean circulation and increased melting under Pine Island Glacier ice
shelf. Nature Geoscience, 4(8), 519–523. https://doi.org/10.1038/NGEO1188
Krabill, W. B., Abdalati, W., Frederick, E. B., Manizade, S. S., Martin, C. F., Sonntag, J. G., et al. (2002). Aircraft laser altimetry measurement
of elevation changes of the Greenland ice sheet: Technique and accuracy assessment. Journal of Geodynamics, 34(3–4), 357–376. https://doi.
org/10.1016/S0264-3707(02)00040-6
Kusahara, K., Hasumi, H., Fraser, A. D., Aoki, S., Shimada, K., Williams, G. D., et al. (2017). Modeling ocean–cryosphere interactions off Adélie
and George V Land, east Antarctica. Journal of Climate, 30(1), 163–188. https://doi.org/10.1175/JCLI-D-15-0808.1
Leuschen, C. (2012). IceBridge MCoRDS L2 ice thickness. National Snow and Ice Data Center.
MacGregor, J. A., Boisvert, L. N., Medley, B., Petty, A. A., Harbeck, J. P., Bell, R. E., et al. (2021). The scientific legacy of NASA’s operation
iceBridge. https://doi.org/10.1029/2020RG000712
Marquardt, D. W. (1963). An algorithm for least-squares estimation of nonlinear parameters. Journal of the Society for Industrial and Applied
Mathematics, 11(2), 431–441. https://doi.org/10.1137/0111030
McMillan, M., Shepherd, A., Sundal, A., Briggs, K., Muir, A., Ridout, A., et al. (2014). Increased ice losses from Antarctica detected by
CryoSat-2. Geophysical Research Letters, 41(11), 3899–3905. https://doi.org/10.1002/2014GL060111
Mengel, M., & Levermann, A. (2014). Ice plug prevents irreversible discharge from East Antarctica. Nature Climate Change, 4(6), 451–455.
https://doi.org/10.1038/nclimate2226
Miles, B. W., Stokes, C. R., & Jamieson, S. S. (2018). Velocity increases at cook glacier, east Antarctica, linked to ice shelf loss and a subglacial
flood event. The Cryosphere, 12(10), 3123–3136. https://doi.org/10.5194/tc-12-3123-2018
Milillo, P., Rignot, E., Rizzoli, P., Scheuchl, B., Mouginot, J., Bueso-Bello, J., & Prats-Iraola, P. (2019). Heterogeneous retreat and ice melt of
thwaites glacier, west Antarctica. Science Advances, 5(1), eaau3433. https://doi.org/10.1126/sciadv.aau3433
Moreau, S., Lannuzel, D., Janssens, J., Arroyo, M. C., Corkill, M., Cougnon, E., et al. (2019). Sea ice meltwater and circumpolar deep water
drive contrasting productivity in three Antarctic polynyas. Journal of Geophysical Research: Oceans, 124(5), 2943–2968. https://doi.
org/10.1029/2019JC015071
Morlighem, M., Rignot, E., Binder, T., Blankenship, D., Drews, R., Eagles, G., et al. (2020). Deep glacial troughs and stabilizing ridges unveiled
beneath the margins of the Antarctic ice sheet. Nature Geoscience, 13(2), 132–137. https://doi.org/10.1038/S51561-019-0510-8
Mouginot, J., Scheuchl, B., & Rignot, E. (2017). Updated 2017. MEaSUREs annual Antarctic ice velocity maps 2005–2017, version 1.
[subset 2016–2017]. Boulder, Colorado USA. NASA National Snow and Ice Data Center Distributed Active Archive Center. https://doi.
org/10.5067/9T4EPQXTJYW9. [Date Accessed on February 9th of 2020].
Navigation, IMAR. (2012). iNAV-RQH-1003: inertial measurement system for advanced applications. Tech. rep. Retrieved from http://www.
imarnavigation.de/downloads/NAV_RQH_1003_en.pdf

CONSTANTINO AND TINTO 8 of 9


19448007, 2023, 11, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2023GL103815 by CAPES, Wiley Online Library on [12/06/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Geophysical Research Letters 10.1029/2023GL103815

Paolo, F. S., Fricker, H. A., & Padman, L. (2015). Volume loss from Antarctic ice shelves is accelerating. Science, 348(6232), 327–331. https://
doi.org/10.1126/science.aaa0940
Planchat, A., Jourdain, N. C., & Dutrieux, P. (2022). Spatio-temporal variability of ocean currents at the Amundsen Sea shelf break and their link
to CDW inflow and ice-shelf melt. Authorea Preprints.
Pritchard, H. D., Arthern, R. J., Vaughan, D. G., & Edwards, L. A. (2009). Extensive dynamic thinning on the margins of the Greenland and
Antarctic ice sheets. Nature, 461(7266), 971–975. https://doi.org/10.1038/nature08471
Rignot, E. (2006). Changes in ice dynamics and mass balance of the Antarctic ice sheet. Philosophical Transactions of the Royal Society A:
Mathematical, Physical & Engineering Sciences, 364(1844), 1637–1655. https://doi.org/10.1098/rsta.2006.1793
Rignot, E., & Jacobs, S. S. (2002). Rapid bottom melting widespread near Antarctic ice sheet grounding lines. Science, 296(5575), 2020–2023.
https://doi.org/10.1126/science.1070942
Rignot, E., Mouginot, J., Scheuchl, B., Van Den Broeke, M., Van Wessem, M. J., & Morlighem, M. (2019). Four decades of Antarctic Ice
Sheet mass balance from 1979–2017. Proceedings of the National Academy of Sciences, 116(4), 1095–1103. https://doi.org/10.1073/
pnas.1812883116
Ryan, W. B. F., Carbotte, S. M., Coplan, J., O'Hara, S., Melkonian, A., Arko, R., et al. (2009). Global Multi-Resolution Topography (GMRT)
synthesis data set. Geochemistry, Geophysics, Geosystems, 10(3), Q03014. https://doi.org/10.1029/2008GC002332
Schoof, C. (2007). Ice sheet grounding line dynamics: Steady states, stability, and hysteresis. Journal of Geophysical Research, 112(F3), F03S28.
https://doi.org/10.1029/2006JF000664
Secretariat SCAR. (1992). Updated 2022 composite gazetteer of Antarctica. Scientific Committee on Antarctic Research. GCMD Metadata
Retrieved from http://gcmd.nasa.gov/records/SCAR_Gazetteer.html
Silvano, A., Rintoul, S. R., & Herraiz-Borreguero, L. (2016). Ocean-ice shelf interaction in East Antarctica. Oceanography, 29(4), 130–143.
https://doi.org/10.5670/oceanog.2016.105
Smith, B., Fricker, H. A., Gardner, A. S., Medley, B., Nilsson, J., Paolo, F. S., et al. (2020). Pervasive ice sheet mass loss reflects competing ocean
and atmosphere processes. Science, 368(6496), 1239–1242. https://doi.org/10.1126/science.aaz5845
Talwani, M., Worzel, J. L., & Landisman, M. (1959). Rapid gravity computations for two-dimensional bodies with application to the Mendocino
submarine fracture zone. Journal of Geophysical Research, 64(1), 49–59. https://doi.org/10.1029/JZ064i001p00049
Tinto, K. (2018). Updated 2020. IceBridge LDEO gravimeter suite L1B Geolocated free air gravity anomalies, version 1. Boulder, Colorado
USA. NASA National Snow and Ice Data Center Distributed Active Archive Center. https://doi.org/10.5067/SU3MNDAOQ7LZ. [Accessed
on February, 10th 2022].
Williams, G. D., Aoki, S., Jacobs, S. S., Rintoul, S. R., Tamura, T., & Bindoff, N. L. (2010). Antarctic bottom water from the Adélie and George
V land coast, East Antarctica (140–149 E). Journal of Geophysical Research, 115(C4), C04027. https://doi.org/10.1029/2009jc005812

CONSTANTINO AND TINTO 9 of 9

You might also like