Long‐Term Observations of Subseafloor Temperatures and Pressures in a Low‐Temperature, Off‐Axis Hydrothermal System in North Pond on the Western Flank of the Mid‐Atlantic Ridge

Abstract

Basement formation pressures and temperatures were recorded from 1997 to 2017 in four sealed‐hole observatories in North Pond, an isolated ∼8 × 15 km sediment pond surrounded by thinly sedimented basement highs in 7–8 Ma crust west of the Mid‐Atlantic Ridge at ∼23°N. Two observatories are located ∼1 km from the southeastern edge of North Pond where sediment thickness is ∼90 m; the other two are ∼1 km from the northeastern edge where sediment thickness is 40–50 m. Sediments are up to 200 m thicker in the more central part of the pond. The borehole observations, along with upper basement temperatures estimated from seafloor heat flux measurements, provide constraints on the nature of low‐temperature ridge‐flank hydrothermal circulation in a setting that may be typical of sparsely sedimented crust formed at slow spreading ridges. Relative to seafloor pressures, basement formation pressures are modestly positive and increase with depth, except for a slight negative differential pressure in the shallowest 30–40 m in one northeastern hole. Although the observatory pairs are ∼6 km apart, the lateral pressure gradient in basement between them is very small. Formation pressure responses to seafloor tidal loading are consistent with high basement permeability that allows for vigorous low‐temperature circulation with low lateral pressure gradients. In contrast, there is significant lateral variability in upper basement temperatures, with highest values of ∼12.5°C beneath the thickly sedimented southwest section, lower values near the edges, and lowest values near the southeast edge. The results are key to assessing past and recent models for the circulation system.

Full text

1. Introduction
Even before the late 1970's discovery of low- and high-temperature hydrothermal vents at mid-ocean ridge axes
(Corliss etal., 1979; Spiess etal., 1980), the existence of low temperature off-axis hydrothermal systems was
suggested by geothermal surveys on well sedimented ridge flanks in the nor theastern and eastern equatorial Pacific
(e.g., Anderson & Hobart,1976; Davis & Lister,1977; Lister,1972; Sclater etal.,1974; Williams etal.,1974).
Those surveys showed significantly lower average seafloor heat flux values than predicted by conductive plate
cooling models, an observation that was quickly verified by global ridge flank data and data compilations (e.g.,
Abstract Basement formation pressures and temperatures were recorded from 1997 to 2017 in four
sealed-hole observatories in North Pond, an isolated ∼8×15km sediment pond surrounded by thinly
sedimented basement highs in 7–8Ma crust west of the Mid-Atlantic Ridge at ∼23°N. Two observatories are
located ∼1km from the southeastern edge of North Pond where sediment thickness is ∼90m; the other two are
∼1km from the northeastern edge where sediment thickness is 40–50m. Sediments are up to 200m thicker
in the more central part of the pond. The borehole observations, along with upper basement temperatures
estimated from seafloor heat flux measurements, provide constraints on the nature of low-temperature
ridge-flank hydrothermal circulation in a setting that may be typical of sparsely sedimented crust formed at
slow spreading ridges. Relative to seafloor pressures, basement formation pressures are modestly positive
and increase with depth, except for a slight negative differential pressure in the shallowest 30–40m in one
northeastern hole. Although the observatory pairs are ∼6km apart, the lateral pressure gradient in basement
between them is very small. Formation pressure responses to seafloor tidal loading are consistent with high
basement permeability that allows for vigorous low-temperature circulation with low lateral pressure gradients.
In contrast, there is significant lateral variability in upper basement temperatures, with highest values of
∼12.5°C beneath the thickly sedimented southwest section, lower values near the edges, and lowest values near
the southeast edge. The results are key to assessing past and recent models for the circulation system.
Plain Language Summary Low temperature hydrothermal systems in young oceanic seafloor
have important integrated thermal and chemical effects, but flow patterns are weakly constrained by direct
observations and are poorly understood, especially in sparsely sedimented settings that are characteristic of
much of the world's seafloor. We present 20years of observations of seafloor and formation pressures and
temperatures recorded by sealed borehole observatories beneath a North Atlantic ridge-flank sedimented
basin (“North Pond”) surrounded by sparsely sedimented basement highs—the first such long-term data from
this common type of setting. The observations indicate that there is a vigorous low-temperature hydrothermal
flow system beneath the sediments of North Pond as previously concluded, but lateral pressure gradients in
uppermost basement are very low; these observations require the permeability of uppermost basement to be
very high. Our determinations of uppermost basement temperatures show greater variations than previously
inferred, with the highest temperatures beneath the most thickly sedimented section. This is inconsistent with
prior conceptual models of unidirectional hydrothermal flow in uppermost basement beneath the sediment
pond, but more supportive of recent models with upflow beneath the most thickly sedimented section of
the pond, accompanied by lateral flow toward the pond perimeter where mixing with seawater occurs.
BECKER ETAL.
© 2022. The Authors.
This is an open access article under
the terms of the Creative Commons
Attribution-NonCommercial-NoDerivs
License, which permits use and
distribution in any medium, provided the
original work is properly cited, the use is
non-commercial and no modifications or
adaptations are made.
Long-Term Observations of Subseafloor Temperatures and
Pressures in a Low-Temperature, Off-Axis Hydrothermal
System in North Pond on the Western Flank of the
Mid-Atlantic Ridge
K. Becker1 , E. E. Davis2 , and H. Villinger3
1Department of Marine Geosciences, Rosenstiel School of Marine, Atmospheric, and Earth Science, University of Miami,
Miami, FL, USA, 2Geological Survey of Canada, Natural Resources Canada, Sydney, BC, Canada, 3Department of
Geosciences, University of Bremen, Bremen, Germany
Key Points:
• Formation pressures in igneous
basement beneath North Pond
sediments are positive except in the
upper few tens of meters near the
pond edge
• Lateral pressure gradients in upper
igneous basement are very low,
implying very high permeabilities
(∼10
−10m
2) and vigorous flow
• Basement temperatures are highest
beneath the thickest sediments,
unsupportive of prior models for
unidirectional flow in igneous
basement
Correspondence to:
K. Becker,
kbecker@rsmas.miami.edu
Citation:
Becker, K., Davis, E. E., & Villinger,
H. (2022). Long-term observations of
subseafloor temperatures and pressures in
a low-temperature, off-axis hydrothermal
system in North Pond on the western
flank of the Mid-Atlantic Ridge.
Geochemistry, Geophysics, Geosystems,
23, e2022GC010496. https://doi.
org/10.1029/2022GC010496
Received 26 APR 2022
Accepted 20 JUL 2022
Author Contributions:
Conceptualization: K. Becker
Data curation: K. Becker
Funding acquisition: K. Becker
Investigation: K. Becker
Methodology: K. Becker, E. E. Davis,
H. Villinger
Project Administration: K. Becker
Resources: H. Villinger
Writing – original draft: K. Becker, E.
E. Davis, H. Villinger
Writing – review & editing: K. Becker,
E. E. Davis, H. Villinger
10.1029/2022GC010496
RESEARCH ARTICLE
1 of 24

Geochemistry, Geophysics, Geosystems
BECKER ETAL.
10.1029/2022GC010496
2 of 24
Anderson etal.,1979). It soon became apparent from the ridge flank heat flux deficits and sediment geochem-
ical profiles that low-temperature circulation extends out to crustal ages of many tens of millions of years, with
net global thermal and geochemical effects as important as those of axial hydrothermal systems (e.g., Mottl &
Wheat,1994; C. A. Stein & Stein,1994; Wheat & Mottl,2004).
Much of our understanding of ridge-flank hydrothermal systems has been drawn from the eastern Pacific ridge
flanks, where high sediment accumulation rates on the Juan de Fuca plate and in the eastern equatorial Pacific
have created relatively continuous burial of underlying basement topography. This sediment cover acts as a less
permeable and therefore confining cap on the circulation system in underlying basaltic basement (e.g., Karato
& Becker,1983; Snelgrove & Forster,1996; Spinelli etal.,2004). On the other hand, ridge flanks that formed
at slow spreading rates, representing about half the ocean floor, are generally marked by rougher basement
topography and lie mostly in more discontinuously sedimented regions. As a result, we understand much less
about off-axis hydrothermal systems in such settings. Here, we report long-term observations of subseafloor
temperatures and pressures from sealed borehole observatories in an unusually well studied location in the “North
Pond” sediment pond on the western flank of the mid-Atlantic Ridge (Figure1). The observations span 20years,
from 1997 to 2017, and provide intriguing clues regarding the nature of low temperature (<25°C) circulation in
oceanic crust created at slow spreading rates.
2. North Pond Geological and Hydrogeological Setting
First mapped and named by Hussong etal. (1978), North Pond is a roughly 8×15 km basement depression
filled with sediments up to 200–300 thick on crust about 7–8M.y. old on the western flank of the Mid-Atlantic
Ridge at ∼22°45′N (Figure1). The basin lies about 110km south of the Kane Fracture Zone, is elongated in a
Figure 1. Location of North Pond and its borehole observatories in the North Atlantic Ocean. (a) Location of North Pond in the region that encompasses the Kane
Fracture Zone (dotted line) and the segment of the Mid-Atlantic Ridge to the south (dashed line). The base map was made using GeoMapApp (www.geomapapp.org)/
CC by (Ryan etal.,2009). (b) Detailed multi-beam bathymetry around North Pond (Villinger etal.,2018). Locations of the 4 borehole observatories within North Pond
are marked with two large stars, and the location of Hole 1074A is marked with a smaller star. Dashed lines show locations of seismic profiles 4–6 and 12–14 used in
Section6 and Figure9.

Geochemistry, Geophysics, Geosystems
BECKER ETAL.
10.1029/2022GC010496
3 of 24
south-southwest to north-northeast direction, subparallel to the mid-Atlantic Ridge, and is completely surrounded
by much more thinly sedimented or exposed basement topographic highs. The deepest sill depth is on the south-
west side, elevated nearly 200m relative to the 4,470–4,500m average seafloor depth within North Pond. The
detailed tectonic origin of the North Pond basin has not been resolved, but it appears to be related to spatial and
temporal variations in extensional tectonic processes and lower order ridge discontinuities along the strike of the
slow-spreading Mid-Atlantic Ridge south of the Kane Fracture Zone (Figure1a). Features like North Pond are
common on ridge flanks created at slow spreading rates.
Near the southeast edge of North Pond, the Deep Sea Drilling Project cored 576m of basement beneath 92 m
of sediment in Hole 395A (Shipboard Scientific Party,1978). The hole was subsequently visited multiple times
for logging (summarized by Becker etal.[2001]), and North Pond was revisited in 1989 and 2009 for detailed
geophysical surveys that collected heat flux, seismic reflection, and multibeam bathymetry data (Langseth
etal.,1992; Schmidt-Schierhorn et al.,2012). During the logging visits, borehole temperatures were consist-
ently observed to be nearly isothermal at about ocean bottom-water temperature, both in the casing that extends
through the sediment section and into the uppermost ∼300m of open hole in basement, before transitioning to a
higher conductive gradient deeper in the hole (Becker etal.,1984; Gable etal.,1992; Kopietz etal.,1990). This
indicated that cold bottom seawater flowed down the hole into permeable upper basement for over two decades at
rates on the order of 20m/hr (2,000L/hr) (Becker etal.,2001; Morin etal.,1992), but that the deeper basement
section was much less permeable.
These observations also suggested an active lateral circulation system in the permeable section of upper basement
beneath the North Pond sediment cap (Langseth etal.,1984). The 1989 survey showed generally low heat flux
values throughout North Pond except for higher values near a basement ridge that extends into North Pond near
the center of the northwestern border of the basin (Figure1b). This pattern led Langseth etal. (1992) to infer
a direction of flow in permeable basement from the southeast to northwest. Further support for that inference
was provided in 1997, when the Ocean Drilling Program (ODP) cored Hole 1074A near the basement ridge
(Figure1b), with downhole temperature probe data indicating a temperature of ∼6.5–7°C in uppermost basement
beneath 63m of sediments (Becker etal., 2001; Shipboard Scientific Party,1998). The 2009 survey mapped
additional higher heat flux values near the northern edge of the pond (Schmidt-Schierhorn etal.,2012), suggest-
ing a component of basement flow in a direction closer to south-north, that is, more parallel to the ridge axis and
to the orientation of the major basement structural trends that can be recognized in Figure1b. Recent numerical
simulations of the circulation in basement beneath North Pond (Price etal.,2022) suggest that the hydrothermal
system could be (a) spatially more complex than previous single-pass conceptual descriptions of flow and (b)
temporally variable at timescales from tens of thousands of years to as short as a few decades.
3. North Pond Borehole Observatory Configurations and Methods
North Pond was revisited twice by scientific ocean drilling for installation of subseafloor sealed-hole observa-
tories originally termed “CORKs” (Circulation Obviation Retrofit Kits; Davis etal.,1992). Figure1b shows the
locations of the North Pond observatory holes and Figure2 illustrates their downhole configurations. In 1997,
the Ocean Drilling Program (ODP) installed an original single-seal CORK in Hole 395A (Becker etal.,2001). It
was visited for data recovery during single dives of the manned submersibles Nautile in 1998 and Alvin in 2001,
and it continued recording until its batteries ran low in 2007. Its 2001–2007 data were accessed in 2011 when the
Integrated Ocean Drilling Program (IODP) returned to North Pond for recovery of the original CORK from Hole
395A. This was followed by its attempted replacement plus establishment of two new reentry holes with newer
generation “CORK-II” observatories that added microbiological, geochemical, and hydrological objectives in
multiple zones (Edwards etal.,2012).
Unfortunately, during the IODP operations, the wellhead of the replacement CORK-II in Hole 395A broke off
at the last stage of the installation process. Therefore, a new, shallower Hole U1382A was sited about 50m to
the west and instrumented with a single-zone CORK-II into the upper ∼100m of basement. Then, a second new
site (U1383) was established near the northern edge of North Pond about 6km away. The location of Site U1383
was selected partly to provide a test of the hypothesis of south to north flow in permeable upper basement, using
the original CORK observations in Hole 395A and new CORK-II data from Hole U1382A as a reference. An
initial attempt to establish a reentry hole in Hole U1383B was aborted after drill bit failure 45m into uppermost

Geochemistry, Geophysics, Geosystems
BECKER ETAL.
10.1029/2022GC010496
4 of 24
Figure 2. Configurations of the four borehole observatories in North Pond.

Geochemistry, Geophysics, Geosystems
BECKER ETAL.
10.1029/2022GC010496
5 of 24
basement. A new reentry hole was established about 25m to the southwest at Hole U1383C, where nearly 300m
of basement were successfully drilled followed by installation of a three-zone CORK-II. The new borehole obser-
vatories were revisited three times for data, sample, and instrument recoveries using the remotely operated vehi-
cle (ROV) Jason from R/V Maria S. Merian in 2012 and 2014 and from R/V Atlantis in 2017. During the 2012
ROV operations, a simpler “CORK-Lite” with samplers and pressure and temperature loggers was installed in
Hole U1383B (Wheat etal.,2012). The installation of the CORK-Lite sealed that hole, eliminating the possibility
of hydrologic perturbations at the adjacent CORK-II in Hole U1383C.
For pressure monitoring, all of the installations utilized Paroscientific Digiquartz absolute gauges rated to
70MPa, models 8B7000-110 in Hole 395A and 8B7000-2 in the 2011 holes (see www.paroscientific.com). For
both models, factory calibrations provide absolute accuracies of ∼0.01% of full-scale range, or ∼7kPa. Resolu-
tion of temporal pressure variations with a given sensor is much better, particularly in the case of those deployed
in 2011, where a more precise digital converter provided resolution of a few Pa (Davis etal.,2018). Hydrostatic
calibrations during submersible visits, as described in Section4, provided improvements over the absolute cali-
brations for determining differences in pressure between the formation and seafloor and assessing possible drift
of gauge readings. Large (∼5kPa) tidal signals resulting from seafloor loading are present in all of the forma-
tion pressure records. These are of intrinsic interest but must also be removed to resolve average pressure levels
accurately. This was done using harmonic analyses and filtering of the tidal signals in the seafloor and formation
gauge readings with the open-source code TAPPY (Tidal Analysis and Prediction in Python, Cera,2011).
Beyond improvements to the electronics, the original CORK and CORK-II designs are quite different in terms of
their physical configurations, as summarized in the remainder of this section. For detailed design descriptions,
see Davis etal.(1992) for the original CORK design, and Edwards etal.(2012) for the new CORK-II installations
in Holes U1382A and U1383C.
The original CORK in Hole 395A incorporated four redundant passive rubber “cup packer” seals over a 1m long
section within uppermost casing for monitoring in a single sealed zone below the deepest extent of cemented-in
casing. It included two pressure gauges, one for seafloor reference and a second just beneath the seals for forma-
tion pressures averaged through the open-hole section of basement deeper than the casing, or ∼112–606m below
seafloor (mbsf). The two pressure gauges were mounted at vertical positions that are about 3m different, such
that the borehole pressure gauge should register ∼30kPa greater pressure under hydrostatic conditions. For
temperature monitoring, the installation included a 600-m-long 10-thermistor cable suspended in the sealed
section. Signals from each of the thermistors were carried by independent pairs of conductors connected to the
same data logger that recorded pressures. The logger recorded a digital count for each thermistor corresponding
to a voltage drop applied across the sensor, as well as the voltage drop across two reference resistors for calibra-
tion. The thermistor digital counts were converted to temperatures using the count-resistance calibration function
for the logger and a resistance-temperature calibration for each thermistor, yielding a resolution of ∼0.02K at the
low values (<25°C) through the basement section in Hole 395A. The pressure and temperature sampling interval
was set at 1hr throughout the 1997–2007 recording period. Upon its recovery in 2011, it was evident that the
original CORK wellhead had landed about a half meter higher within the reentry cone than planned owing to a
discrepancy in diameters at the throat of the reentry cone, but the rubber cup seals were still deep enough to seal
the installation. Unfortunately, there was a similar diameter discrepancy in the case of the replacement CORK-II,
and this was likely the cause of its failure when it was put under compression during installation.
The CORK-II design in Holes U1382A and U1383C includes downhole packer assemblies that were used to
isolate uppermost basement in Hole U1382A and three different zones in basement in Hole U1383C. The upper-
most packer assemblies near the bottom of casing each included an inflatable rubber packer to seal quickly on
deployment, plus a “swellable” packer made with a rubber compound that expands on exposure to seawater for
a long-term seal. In Hole U1383C, single inflatable packers were used to seal deeper zones within the open-hole
section. Geophysical logs were used to identify competent and smooth-caliper sections of this borehole to select
the deeper packer seats, although the logs provided no indications for hydrologically distinct units isolated
between the downhole seals. Pressures from screens within the isolated zones are monitored via hydraulic tubing
that runs up from those zones to pressure gauges and a data logger mounted on the wellhead; the wellhead assem-
bly also includes an additional pressure gauge for a seafloor reference pressure signal at each location. Valves
on the wellhead allow for switching the input into the formation gauges to the open ocean reference signal for
determining calibration offsets among gauge readings as well as individual gauge drift; this was done during each

Geochemistry, Geophysics, Geosystems
BECKER ETAL.
10.1029/2022GC010496
6 of 24
of the three ROV visits. For the 2011–2017 pressure data reported in this paper, the sampling interval was set at
2min.
Bottom-water temperatures are sensed by platinum Resistance-Temperature Devices (RTDs) coupled to an endcap
of each pressure data logger and recorded at the same 2-min interval as the pressure data. (see Becker etal.[2021]
for a detailed analysis of variations in the bottom water temperature records from 2011 to 2017 and implications
for renewal of North Pond bottom waters by incursions of cold, dense bottom water over the sill on the southwest
side of North Pond). Downhole temperatures were logged by autonomous-memory temperature loggers attached
to downhole geochemical and microbiological samplers. Because of their much more limited battery capacity,
these were set to longer (1.5 or 2-hr) sampling intervals than that used for the pressure logging systems. Note that
the pressure and bottom-water temperature data from the wellhead data logger could be accessed on each ROV
visit, whereas the downhole temperature data required recovery of the downhole sensor strings (unlike the case
of the early data recorded in Hole 395A). That was attempted during the most recent ROV revisit in late 2017,
but both sensor strings broke off at relatively shallow depths (Figure2), with the result that only one tempera-
ture logger each was recovered from Holes U1382A and U1383C. At the end of the 2017 ROV operations, the
pressure sampling rate was reset to 20min to conserve battery life for the possibility of a future revisit if that
opportunity ever arises. The CORK loggers are configured to allow application of power from a submersible for
data downloading after batteries are depleted.
The CORK-II that broke off in Hole 395A included three downhole packer assemblies separating three monitor-
ing zones as in Hole U1383C, but it extended even deeper into basement. The break occurred after the downhole
components were fully installed and the packers had been inflated. The wellhead broke off at the throat of the
reentry cone, and during the 2014 and 2017 ROV dives the upper end of the remaining assembly was visually
observed to remain in position in the throat of the reentry cone. This indicates that the packers have remained
inflated and are supporting the downhole assembly in place, such that the hole is mostly sealed and probably not a
source of major hydraulic interference with the observations in the adjacent Hole U1382A installation. However,
also observed were several broken small-diameter (1/8” to 1/2“ O.D.) umbilicals that were originally connected
to ports in the formation. It is not clear if these umbilicals still provide intact connections to the formation; if so,
they might allow for minor perturbations to observations in the adjacent Hole U1382A installation.
The CORK-Lite in Hole U1383B incorporated a single seal at the seafloor and a short sensor string suspended
in the hole. A pressure logging system of the same design as that in Holes U1382A and U1383C was deployed
on the landing platform, with one gauge attached to a valve on the wellhead to monitor formation pressure and
another gauge open to the seafloor reference signal. This system was also set to a 2-min sampling interval. The
downhole sensor string suspended one instrument cluster in the top of open basement, including two autonomous
temperature loggers for redundancy at the same depth, set at 2-hr sampling intervals. This string was recovered
in 2017, after which the CORK-Lite seal and pressure logger were reinstalled, the latter reset to 20-min sampling
interval.
4. Results
4.1. Formation Pressure and Temperature Records From the Original
CORK in Hole 395A
Figures3 and4 show the 1997–2007 records of pressures and temperatures,
respectively, obtained with the original CORK in Hole 395A. It is immedi-
ately apparent from Figure3 that there was some sort of early instrumental
issue with the formation pressure gauge reading. This began about 37days
after deployment and apparently self-resolved just 3days before the 1998 visit
by Nautile. We suspect this arose from an issue with the pressure integration
module associated with the gauge, but the readings were reliable after that
episode. Throughout the 10-year records, the difference between the forma-
tion and seafloor gauge readings remained close to the 30 kPa difference
expected for the ∼3m difference in absolute depths of the gauge sampling
ports, although there might have been some temporal drift of the seafloor
gauge reading that introduced uncertainties of a few kPa. The change in trend
Figure 3. Ten-year record of pressures recorded from 1997 to 2007 with the
original CORK in Hole 395A.

Geochemistry, Geophysics, Geosystems
BECKER ETAL.
10.1029/2022GC010496
7 of 24
for the seafloor gauge readings around 2001 is puzzling, but if it involves drift, the apparent drift rates before and
after that change are within typical limits for Paroscientic Digiquartz gauges reported by Polster etal. (2009).
Both pressure signals show tidal loading effects, similar to those seen in the initial 37days and analyzed by
Davis etal.(2000). The attenuation of the formation gauge tidal amplitudes relative to seafloor tidal amplitudes
is significant, suggesting a robust seal at the original CORK, and appears to increase slightly over the 10-year
period. The tidal signals are discussed further in Section5.3.
The shift in the borehole pressure reading apparent at the time of the 2001 Alvin visit (Figure3) resulted from an
unsuccessful attempted calibration of the offsets between seafloor and formation gauges. This involved connect-
ing an independent pressure gauge to a formation sampling valve and opening the valve for ∼8min while also
recording readings from both seafloor and formation gauges at 1-min intervals. A negative shift in the borehole
pressure reading was observed immediately after the sampling valve was reclosed, suggesting that the valve open-
ing during the calibration attempt may have vented ∼3–4kPa of positive differential formation pressure, and the
ensuing record suggests it required at least 1year for the borehole pressures to recover to previous levels. Based
on the long history of downhole flow in Hole 395A, initial interpretations were that the formation was slightly
negatively pressured (Becker etal.,2001). However, a subsequent analysis by Davis etal.(2010) indicates that
drilling with cold seawater can induce a “runaway” downhole flow that can persist in an open or incompletely
sealed hole for years even when the natural formation state is positively pressured. During the 2001 dive, the
sampling valve was initially observed to be slightly rotated (a few degrees) from a fully closed position. The
valve was turned fully closed after the calibration attempt. The long duration of the subsequent pressure recovery
probably resulted from prior leakage from 1998 to 2001.
In the case of the long-term temperature records (Figure4), no data are shown from thermistors T2 at 545 mbsf or
T8 at 245 mbsf. Their readings were erratic from the start, especially T2, suggesting penetration of seawater into
either their molded housings or their independent conductors. The early parts of the records for each sensor show
borehole thermal recovery periods of well over a year, such that temperatures recorded at the time of the 1998
Nautile visit and shown by Becker etal.(2001) were not yet close to equilibrium with formation conditions. The
longer record shows a puzzling year-long increase in the temperature of the deepest sensor after ∼3years, plus
some small events at about 1 and 5years that are discussed in the next paragraph. Otherwise the records show
greater thermal stability by 10 years after deployment. Figure5 shows the temperatures at 10years compared
Figure 4. Ten-year record of temperatures recorded from 1997 to 2007 with the original CORK in Hole 395A.

Geochemistry, Geophysics, Geosystems
BECKER ETAL.
10.1029/2022GC010496
8 of 24
to a profile collected with the ODP DVTP (Davis-Villinger Temperature Probe; Davis etal.,1997) just before
installation of the CORK that indicated isothermal conditions in the uppermost ∼300m of basement but a steeper
gradient below. This is associated with a transition from high permeabilities in the uppermost 300m of basement
to much lower permeabilities below ∼500m into basement (Becker,1990; Becker etal.,2001). It is clear that
temperatures in that uppermost section of basement remained quite low after 10 years and were not likely to ever
re-equilibrate to a purely conductive profile. Instead they are consistent with an inferred natural state (modified
slightly from Davis and Becker[2004]) with (a) relatively low temperatures in the upper section of basement
associated with persistent lateral flow of cold seawater and (b) a deeper section dominated by vertical conduc-
tive heat flux of ∼145mW/m
2, about 80% of the value predicted for conductive cooling of crust ∼7–8M.y. old.
Thus, these observations seem to confirm the continued vigor of the lateral fluid flow system of cold seawater
in permeable uppermost basement beneath the North Pond sediment cap, even if they offer no resolution for the
direction, rate, or other details of flow.
Figure 5. Comparison of temperatures recorded in Hole 395A with the DVTP probe in 1997 and CORK in 2007, 10 years
after its deployment. Also shown are the Davis and Becker(2004) estimate of the natural pre-drilling temperature profile,
based on 1997–2001 CORK temperature data, and an interpretation from Becker etal.(2001) of zones that accepted inflow
from the 1976–1997 downhole flow based on anomalies in a 1997 downhole log of spontaneous potential. Those zones
correlate well to brecciated and probably highly transmissive boundaries between major eruptive units delineated by other
logs such as resistivity (Bartetzko etal.,2001; Hyndman & Salisbury,1984).

Geochemistry, Geophysics, Geosystems
BECKER ETAL.
10.1029/2022GC010496
9 of 24
It is also notable from Figure4 that sensors from the middle part of the cable show small apparent drops in
temperature at about 1year into the record and again 5years into the record. No concurrent pressure changes
are seen (Figure3). The temperature changes occur over timescales of 1–2days and there are additional similar
smaller amplitude events apparent at other times. Not showing such effects are the deepest sensor (T1) in the rela-
tively impermeable section or the shallowest sensor (T10) within the cased section. The sensors that show these
effects are all in either the permeable section of the hole that probably accepted downhole flow of bottom seawa-
ter before the hole was sealed or the underlying transitional 100m to much lower permeability very deep in the
hole. Thus, it is possible that the apparent events recorded by thermistors in this section reflect some combination
of (a) thermally or turbidity induced convective overturn in that section of the borehole, (b) flow into or out of one
or more of the inferred inflow zones, or (c) temporal variability in the overall lateral fluid flow system in upper-
most basement beneath the North Pond sediment cover. The last would be consistent with the indications from
recent numerical simulations by Price etal.(2022) for temporal variability, although natural variations would be
expected to take place over longer timescales (e.g., decades or longer as estimated by Price etal.[2022]).
4.2. Formation Pressure Records From the Observatories in Holes U1382A, U1383C, and 1383B
Unprocessed pressure records from the three new North Pond observatories are presented in Figures6a–6c,
subsampled at hourly intervals, along with superimposed versions from which the tidal components have been
filtered out. Pressures generally stabilized soon after installation, with the possible exceptions of small events in
the record from Hole U1383C (see next paragraph), a slow decrease in the formation reading from Hole U1383B
that is probably a gauge drift effect, and slow parallel increases in the readings from the two gauges at Hole
U1382A that could be similar gauge drift effects. The last might be interpreted to suggest slow sinking of the
U1382A installation at a rate of ∼1.5kPa/a or 0.15m/a, but there were no visual indications of this or corrobo-
rating changes in the depths recorded by the ROV pressure gauges during the 2012, 2014, and 2017 ROV visits.
Like the earlier record from Hole 395A, all the pressure readings show effects of seafloor tidal loading, but the
attenuation of the formation tidal variation relative to the seafloor reference is smaller. Figures6d and6e show
the differences between detided formation and seafloor pressures, uncorrected and corrected, respectively, for the
offsets revealed by hydrostatic checks in 2012, 2014, and 2017.
Unlike the data from Hole 395A, the records from the new CORKs unequivocally resolved non-hydrostatic
formation pressures in Holes U1382A, U1383B, and U1383C. This is evident in Figure6e and during the hydro-
static checks of gauge offsets during ROV operations in 2012, 2014, and 2017 (Figures7a–7c). The hydrostatic
checks in Holes U1382A and U1383C indicated positive differential pressures between formation and seafloor
hydrostatic reference of up to 13kPa, with the differential increasing with depth among the three isolated zones
in Hole U1383C. The hydrostatic checks were associated with small negative offsets in the formation pressure
readings (Figure6e), likely because the three-way valves allow brief venting of positive formation pressures
when turned to sense seafloor pressure. The corrected pressure differences (Figure6e) show a few additional
small pressure drops unassociated with any ROV operations. The inset Figure6f shows an expanded view of the
most prominent example, that in the U1383C middle formation gauge record in late October 2015, showing that
the offset occurred over a time period of about one day. Another example in the U1382A formation record in
late summer of 2012 occurred over a similar time interval, as did earlier variations in Hole 395A temperatures
described in Section4.1. As the instruments remain in place, we cannot ascertain whether the short events in the
pressure records might reflect real changes in the formation pressures or might have been caused by instrumental
issues.
Attempted hydrostatic checks at the CORK-Lite in Hole U1383B in 2014 and 2017 showed no changes in pres-
sure when the formation gauge valve was switched to seafloor pressure, possibly indicating an issue with the
valve mechanism. Nevertheless, an effective hydrostatic check there was achieved in 2017 when the data logger
was disconnected from the installation prior to conducting a tracer injection experiment (Wheat etal., 2020).
This check showed a clear increase in the formation gauge reading upon disconnection, indicating a negative
differential pressure prior to unsealing of ∼2.5kPa in the shallow section of uppermost basement penetrated
there (Figure7c). A downward shift of the formation gauge reading on the order of ∼3–4kPa had been registered
at the time of original deployment of the CORK-Lite in Hole U1383B, also suggesting a negative differential
pressure (Wheat etal.,2012). The larger magnitude of that negative differential might have been a residual effect

Geochemistry, Geophysics, Geosystems
BECKER ETAL.
10.1029/2022GC010496
10 of 24
of the hole having been left open to possible downhole flow of cold seawater for ∼6months before installation
of the CORK-Lite.
A negative differential pressure in Hole U1383B might seem surprising in light of the positive differential pres-
sure registered in the upper monitoring zone in Hole U1383C 25m away, but the open basement section in Hole
U1383B is shallower than the upper screen and interval in Hole U1383C (Figure2). Furthermore, local flow
patterns and the local pressure regime may be complex near the thinly sedimented edges of the sediment pond
(Villinger etal.,2019). As shown in Figure7d, the differential pressures registered in Holes U1383B and U1383C
are nearly colinear with depth in the upper 150m of basement, suggesting a crossover from negative to positive at
∼30m into basement. The small underpressure in uppermost basement at Site U1383 is consistent with results of
two sediment pore pressure deployments in 1989 near the thinly sedimented northwestern edge of the sediment
pond (Langseth etal.,1992). Also as shown in Figure7b, the positive differential pressures in Holes U1382A
and U1383C seem to be slowly decreasing over time. Like the temperature records from the permeable section
of Hole 395A, this may be an indication of temporal variability in the lateral flow system in permeable basement
beneath the North Pond sediment cap, consistent with the recent numerical simulations by Price etal.(2022).
Figure 6. Pressure records from the borehole observatories at Sites U1382 and U1383. (a–c) Six-year records of pressures recorded through 2017 with the CORK-II
installations in Holes U1382A and U1383C and CORK-Lite in Hole U1383B. Each trace shows original readings in color, and detided versions with central white lines.
The same vertical axis scale is used for each plot, and “H” denotes the offsets recorded during hydrostatic checks. Note that the data shown here are original readings
not corrected for the hydrostatic checks shown in Figure7, because the corrections bring readings into partial overlap, obscuring the nature of the tidal response in some
of the zones. (d) Differences between detided formation and seafloor pressure readings, uncorrected for the offsets revealed by the hydrostatic checks. (e) Corrected
differences between detided formation and seafloor, calculated as follows: For data from Holes U1382A and U1383C by applying linear (in time) corrections (a) to
the 2011–2014 data based on the 2012 and 2014 hydrostatic checks and (b) to the 2014–2017 data based on the 2014 and 2017 hydrostatic checks. For data from Hole
U1383B, a constant offset was applied based on the 2017 hydrostatic check shown in Figure7a. (f) Inset showing expanded view of the enigmatic offset registered in
pressures from the middle interval of Hole U1383C in late October 2015.

Geochemistry, Geophysics, Geosystems
BECKER ETAL.
10.1029/2022GC010496
11 of 24
5. Discussion
5.1. Lateral Variations in Upper Basement Temperature and Implications
During the 2017 attempts to recover the sensor strings from Holes U1382A and U1383C, the sensor strings parted
at shallow depths, which left most of the temperature sensors in the holes. Immediately apparent in the temporal
records provided by the single recovered sensors from each of the holes (Figure8a) is the contrast between the
smooth temperature recoveries in Holes U1382A and U1383B, and the abrupt change registered in Hole U1383C
about 1year after deployment. That sharp change occurred in only 6hr, and there is no resolvable accompanying
signal in the pressure records from that zone or the zone below. Given its short duration, it is difficult to imagine
an isolated convective overturn of the fluids within the casing or a sudden natural formation temperature change
of that magnitude. Since most of the sensor string was left in the hole during the 2017 recovery attempt, the cause
of this anomaly may never be resolved.
Figure 8b shows the resolution of the in situ temperature-depth profiles at the CORK sites and Hole 1074A
provided by the downhole data. Notable is the consistency of the single temperature reading in Hole U1382A
with the nearly isothermal profile recorded earlier in the adjacent Hole 395A. This leads to the conclusion that
in situ temperatures in uppermost basement are significantly lower (by ∼3–4K) at Sites 395 and U1382 than
at Site U1383 or Hole 1074A despite the greater sediment thickness at the former two sites. This observation
Figure 7. Indications for basement differential pressures recorded at Sites U1382 and U1383. (a–c) Records of the valid
hydrostatic checks performed at the CORK installations in Holes U1382A, U1383B, and U1383C during the ROV visits
in 2012, 2014, and 2017. Note that the data shown were corrected by small amounts to bring the readings to a common
reference at each installation, assumed here to be provided by the values recorded by the seafloor gauge. Note also that the
timescales for panels (a and b) are much shorter than for panel (c), partly because the 2017 hydrostatic check at Hole U1383C
extended over two ROV dives when the first was cut short by a sudden ROV issue. (d) Temporal trend in the indications for
positive formation pressure differentials in Holes U1382A and U1383C provided by the hydrostatic checks in 2012, 2014,
and 2017.
U1382A 2012
U1382A 2014
U1382A 2017
U1383C 2012
U1383C 2014
U1383C 2017
U1383B 2017
d
Pressure Differential (kPa)
0
50
100
150
200
0510 15 20
-5
44980
44970
44960
a
45710
45700
44690
44680
Time (UTC)5April2014
b
U1382A
5April
Seafloor
U1383C
13 March
Seafloor
Time (UTC)31March 2014
c
45010
45000
44990
44980
44970
44960
45710
45700
44690
44680
44670
44660

Geochemistry, Geophysics, Geosystems
BECKER ETAL.
10.1029/2022GC010496
12 of 24
would be consistent with a single-pass flow system in uppermost basement as previously modeled by Langseth
etal.(1984, 1992), with flow directions from southeast to northwest or south to north. It would also be consistent
with the numerical results of Price etal.(2022) that alternatively suggest a primary upward plume beneath the
thickly sedimented central basin and lateral flow in uppermost basement toward all edges of the sediment pond.
Additional constraints on the variation of uppermost basement temperatures beneath North Pond sediments can
be gleaned by extrapolation of seafloor heat flux measurements to the sediment-basement interface using seismic
profiles of Schmidt-Schierhorn etal.(2012). Some of the seismic profiles display side echoes from the surround-
ing rough topography that obscure true basement reflections. Therefore, we selected the six profiles least affected
by side echoes, three along the strike of North Pond and three across strike (Figure1b) and used heat flux data
collected along those profiles to estimate upper basement temperatures (Figure9). DSDP, ODP, and IODP coring
data do not constrain seismic velocities or thermal conductivities deeper than 100m, so we used constant values
of 1.0W/m-K for sediment thermal conductivity and 1.55km/s for sediment seismic velocity in our extrapola-
tion. In addition to uncertainties in estimating uppermost basement temperatures arising from these assumptions,
uncertainties also arise from the poor seismic definition of the sediment/basement interface. The largest errors in
estimating temperatures at the sediment-basement interface probably arise from uncertainties in basement depth
where basement reflections were difficult to pick or absent. Quantifying potential errors in estimated tempera-
tures at the top of the basement is not possible due to uncertainties in resolving the basement reflector as well as
the lack of observational constraints on the depth variations and possible lateral variations of thermal conductiv-
ity and seismic velocity in the sediments. Furthermore, there are also some local anomalies associated with both
the basement ridge that plunges into the northwest side of North Pond and basement topography in the northeast
section of North Pond, especially along SCS line 4 (Figure9a). Despite the uncertainties, a general pattern is
evident (Figures9 and10), with: (a) higher temperatures, on the order of 10K above bottom-water temperature,
beneath the most thickly sedimented southwestern areas of North Pond (Figure9a), (b) upper basement tempera-
tures of 3–4K above bottom water values near the northeast, northwest, and southwest edges of the pond, and (c)
lowest upper basement temperatures near the southeast edge in the area where Sites 395 and U1382 are located.
These results show significantly different lateral variations in uppermost basement temperatures beneath North
Pond than previously inferred for a single-pass lateral-flow heat exchange model by Langseth etal.(1984, 1992),
or those indicated by the borehole data alone at Sites 395, 1074, U1382, and U1383, all of which are within
roughly 1km from basement outcrops at the pond perimeter. Taken together, the estimates from extrapolated heat
flux observations and the borehole observations are more consistent with the general flow patterns simulated
by Price etal. (2022) with a central upward plume beneath the thickly sedimented section of North Pond and
lateral flow in uppermost basement toward all edges of the pond (although it should be noted that those models
Figure 8. Summary of subseafloor temperature data recorded in North Pond. (a) Temperature-time records provided by the
three memory temperature loggers recovered in 2017 from Holes U1382A, U1383B, and U1383C (only one from each hole
for reasons described in the text). (b) Apparent temperature-depth profiles from Holes 395A, 1074A, U1382A, U1383B, and
U1383C.

Geochemistry, Geophysics, Geosystems
BECKER ETAL.
10.1029/2022GC010496
13 of 24
do not incorporate local basement topography like the basement ridge under the northwest section of North
Pond). The borehole and heat flux observations are also consistent with those made on the sedimented flank
of the Juan de Fuca Ridge near locations of extensive basement outcrops, specifically as illustrated in Figures
8.3a and 8.3b of Davis and Becker(2004) where uppermost basement temperatures extrapolated from seafloor
heat flux increase above bottom-water temperature systematically by ≳10K over distances of ∼5km away from
locations of basement outcrops. Given the limits of the North Pond seismic data in determining the depth of the
Figure 9. Temperatures in sediments and uppermost basement estimated by extrapolating surface heat flux measurements collected along single-channel seismic (SCS)
profiles across North Pond (see Figure1 for profile locations). (a–c) Along-strike profiles 4–6. (d–f) Cross-strike profiles 12–14. Circles denote sediment-basement
contacts interpreted at each heat flux station from the SCS profiles. Estimated temperatures are represented with colored bars at 2.5K increments above bottom-water
temperature (Tbw) of 2.5°C, color-coded as follows: white=Tbw, blue=+2.5K, green=+5K, yellow=+7.5K, and red=+10K. Each heat flux measurement
location is annotated with station number from Schmidt-Schierhorn etal.(2012) or Langseth etal.(2018), heat flux value in mW/m
2 and estimated temperatures at
the sediment-basement interface (in °C). (For a few stations with very low heat flux values, the blue bars appear deep in basement or may be off-scale, but those cases
should not be taken as accurate representations of temperatures that deep in basement because they are conductive extrapolations using sediment properties.)

Geochemistry, Geophysics, Geosystems
BECKER ETAL.
10.1029/2022GC010496
14 of 24
sediment/basement interface, and the unquestionably large and complex variations of permeability in the igneous
crust, details of the circulation cannot be resolved (e.g., influences on the thermal state associated with cellular
convection, and from local buried basement relief as has been identified on the Juan de Fuca Ridge and Costa
Rica Rift flanks; see Figures 8.3b–8.3e in Davis and Becker[2004]). Nevertheless, it is clear that temperatures
beneath the parts of North Pond farthest from the surrounding basement outcrops are also warmest, ≳10K above
bottom water temperature. Nowhere, however, do uppermost basement temperatures rise to a level that would be
commensurate with the deep-seated lithospheric flux and local sediment thickness. This is to be expected, given
the small size of the pond relative to the >20km scale over which heat is mined by lateral hydrothermal circula-
tion as observed at other locations (e.g., Davis & Becker,2004; Davis etal.,1999; Hutnak etal.,2008).
Figure 9. (Continued)

Geochemistry, Geophysics, Geosystems
BECKER ETAL.
10.1029/2022GC010496
15 of 24
5.2. Basement Formation Pressures and Lateral Pressure Gradient
Beneath North Pond, and Possible Implications
The clear indications for positive vertical differential pressures in Holes
U1382A and U1383C across the sediment layer between igneous basement
and the seafloor were initially unexpected, given the long history of down-
hole flow in Hole 395A and prior CORK results from holes on well sedi-
mented ridge flanks in the northeast Pacific and eastern equatorial Pacific. In
most of those Pacific locations, positive differential formation pressures (up
to 20kPa in magnitude) were registered in CORKs in sediment-covered base-
ment highs, while negative differential formation pressures (up to 25–30kPa)
were registered in CORKs situated in nearby sediment-covered basement
lows (i.e., in cases most analogous to North Pond). In contrast, temperatures
at the sediment-basement interface at these locations were nearly identical for
the closely spaced basement high/basement low paired CORKs despite the
difference in sediment thicknesses, implying vigorous lateral hydrothermal
circulation in permeable uppermost basement (Becker etal.,2004; Davis &
Becker,1994, 2002, 2004). In those cases, converting equilibrated differen-
tial pressures measured vertically across the sediment cover at the CORKs to
the formation over- or under-pressures that operate to drive lateral hydrologic
flow within basement required using the temperature data to calculate the
head of a hydrostatic reference column at densities appropriate for the in situ
temperature profile along the lateral flow path (e.g., Davis & Becker,1994,
2002; J. S. Stein & Fisher,2001). For the Davis and Becker(2002) example
in crust ∼3.5M.y. old, the resulting lateral pressure differentials were found
to be small, on the order of 2kPa, in contrast with the vertical pressure differ-
entials of +19kPa at the basement high CORK and −26kPa at the basement
low CORK. In this hydrologically isolated case, the temperature difference
between the two locations was unresolvably small, despite their separation
of over 2km and the contrast of sediment thickness of more than a factor of two. High permeability and high
Rayleigh number convection were inferred.
The North Pond hydrological structure is fundamentally different from the eastern Pacific examples, in that the
sediment cap over North Pond is restricted laterally. The surrounding basement highs are very thinly sedimented
and thus open to free exchange with cold ocean bottom water. In this type of setting, the North Pond sediments
serve as a vertical permeability barrier and geothermal insulator, the lateral dimensions of which are hydrologi-
cally small. This should cause upper basement formation temperatures beneath the pond at locations well away
from its perimeter to be higher—and formation fluid densities lower—than in the surrounding regions that are
kept very close to bottom water temperature by free circulation of cold bottom water. Considering only the effects
of temperature structure on formation fluid densities, this would produce positive lateral gradients of static pres-
sure from the surrounding basement toward basement beneath North Pond—and positive differential pressures
across the sediment layer of the pond, including at the CORK locations (Figure11a).
If the approach described above to calculate lateral pressure gradients (Davis & Becker,1994, 2002) is applied
using the apparent thermal gradients in Holes U1382A and U1383C (Figure8b) compared to cold isothermal
gradients inferred to apply in the surrounding thinly sedimented hydrothermally cooled basement, it yields much
smaller predicted positive differential pressures with depth than actually registered during the hydrostatic checks
(Figure11b). These were calculated assuming temperature profiles in the two holes as follows: (a) from 2.5 to
3°C through 90m of sediments, then from 3 to 5°C through the upper 300m of basement in Hole U1382A and
(b) from 2.5 to 5°C through 50m of sediments, then from 5 to 12.5°C through the upper 300m of basement in
Hole U1383C. We used these assumed profiles to calculate the variation with depth of formation fluid densities,
using the equation of state of Millero etal.(1980) with a salinity of 34.8‰ appropriate for bottom water at the
location, then integrated the density profiles to calculate pressure-depth profiles using a value of 9.788 m/s
2
for the gravitational acceleration at 23°N latitude. For the deep intervals in both holes, the differences between
the measured pressures and the hydrostatic profiles defined for the respective geothermal profiles (“U1382A T
model” and “U1383C T model” on Figure11b) are nearly identical at ∼11kPa in 2017 (dashed horizontal lines
Figure 10. Compilation of estimated temperatures at the sediment basement
interface (open circles) versus estimated sediment thickness for all heat flux
measurements shown on the seismic profiles in Figure9. Also shown are
contours of temperature above bottom-water value of 2.5°C, in increments of
2.5K color-coded as: blue=+2.5K, green=+5K, orange=+7.5K, and
red=+10K.

Geochemistry, Geophysics, Geosystems
BECKER ETAL.
10.1029/2022GC010496
16 of 24
Figure 11.

Geochemistry, Geophysics, Geosystems
BECKER ETAL.
10.1029/2022GC010496
17 of 24
on Figure11b). This suggests that the lateral pressure gradient between the two sites available to drive lateral flow
in basement is very small, unresolvable in light of the uncertainties arising from the ill-constrained temperature
variation along the hydrologic path connecting the sites.
Also shown in Figure11b are similar calculations for two models for warmer conditions in the thickly sedimented
southwestern section of North Pond. These are based on the indications described above for temperatures on the
order of 12.5°C at the sediment-basement contact beneath ∼300m of sediment of constant thermal conductivity.
We then modeled two example linear temperature gradients within basement. These were based on reaching 22°
(the highest temperature recorded ∼500m into basement in Hole 395A) at 300m (SW Model #1), and 100m
(SW Model #2) into basement. There are no data constraints on temperatures within basement in this section of
North Pond, but the second case approximates reaching the predicted lithospheric conductive flux beneath the
thick sediments. The results (Figure11b) are much closer to the magnitudes of positive differential pressures
measured in the deeper basement penetrated in Hole U1383C, although they do not match the depth variations of
differential pressures at that hole or U1382A.
There are two likely contributions to the higher positive differential pressures that were measured in Holes
U1382A and U1383B. First, a key result of the numerical simulations by Price etal.(2022) is the indication for
a central upward plume in basement beneath more thickly sedimented regions of North Pond, characteristically
producing positive vertical pressure gradients within permeable basement of several Pa/100m (Figure 8.1 of
Davis and Becker[2004]). Second, if recharge is occurring through exposed, elevated basement surrounding
North Pond, it would augment any differential pressures as a consequence of the loading by cool pond-perimeter
crustal fluids. The two effects are probably linked; for example, the Price etal. results show that the recharge
through the surrounding basement highs directly links to the central upward plume that produces characteristic
positive differential pressures across the sediment section and vertical pressure gradients within basement.
5.3. Response of Subseafloor Pressures to Seafloor Tidal Loading
As noted in Section4, the formation pressure readings from all of the North Pond CORKs displayed attenuated
responses to the seafloor tidal loading signal. The relative amplitudes can be discerned visually on Figures4
and6, and the records also show formation phase lags that cannot be discerned on those figures. After apply-
ing harmonic analysis to the data from each pressure gauge, Figure12 shows representative amplitude ratios
and phase lags of the North Pond formation responses to seafloor tidal loading for the dominant diurnal and
semi-diurnal constituents. In general, the amplitude ratios and phase lags are quite consistent among all the
formation zones in Holes U1382A, U1383B, and U1383C, and remained quite constant over the 2011–2017
data period. As noted in Section4, for Hole 395A the attenuation of the formation signal seemed to increase
with time, perhaps a consequence of the formation sampling valve possibly not having been completely closed
until 2001, so formation response data from before then may be less reliable. The subsequent Hole 395A data
showed significantly lower amplitude ratios and greater phase lags compared to the results from the newer CORK
installations at Sites U1382 and U1383. This may be partly because the response in Hole 395A is averaged over
a thicker basement interval that includes relatively low permeability basement in intervals deeper than basement
depths penetrated at Sites U1382 and U1383.
The theoretical basis for this response is well documented by van der Camp and Gale (1983) and Wang and Davis
(1996). The formation response to the seafloor tidal loading has two components: (a) an in-phase elastic response
that depends on the formation loading efficiency, a function of porosity and the compressibilities of the formation
matrix and fluids and (b) an out-of-phase diffusive response that depends on the path over which diffusion occurs
and the hydraulic diffusivity of the formation. If there is no phase difference, then the amplitude ratio provides a
good measure of the loading efficiency, a purely poroelastic reflection of the formation properties. Where there
is a laterally continuous sediment cover, the vertical diffusive response quickly dies out with depth, but in a
Figure 11. Hydrologic/geothermal model for an isolated sediment pond like North Pond. (a) Southwest to northeast cartoon of the sediment pond based schematically
on SCS profile 5, along which Site U1383 was positioned in the northeast. Note that it passes ∼2km northwest of Holes 395A and U1382A; hence the projected
positions of the latter two holes are not shown because the profile does not reflect the true sediment thickness of ∼90m there. Their projected position would be at
∼9km on the distance scale. (b) Calculated positive differential pressures at the two sites consistent with buoyancy effects relative to assumed isothermal formation
fluids (at bottom-water temperature) in basement surrounding North Pond. For comparison, also shown are hydrostatic pressure profiles in basement in the thickly
sedimented southwest section of North Pond, based on a temperature of ∼12.5°C in uppermost basement and two models (SW Models #1 and #2) for the temperature
profile with basement (described in Section5.2), all calculated relative to a cold (bottom-water temperature) hydrostat.

Geochemistry, Geophysics, Geosystems
BECKER ETAL.
10.1029/2022GC010496
18 of 24
Figure 12. Representative results of harmonic analyses for the primary diurnal and semi-diurnal tidal constituents for all of the formation intervals isolated by the
North Pond CORKs. (a) Formation to seafloor amplitude ratios and (b) phase lags of the formation signal relative to seafloor tidal loading. The key shown in panel (b)
applies to both panels. For Hole 395A, results are based on the last 3 years of data recorded at 1-hr intervals; for the other CORKs, results are based on the year of 2-min
data from May 2014 through April 2015 inclusive.
0.5
0.6
0.7
0.8
0.9
1.0
0
5
10
15
20
Q1 O1 K1 N2 M2 S2
395A '04-'07
U1382A
U1383C Deep
U1383C Middle
U1383C Shallow
U1383B
Constituent
(periodinhrs)
(26.867) (25.820) (23.934) (12.658) (12.421) (12.000)
a
b

Geochemistry, Geophysics, Geosystems
BECKER ETAL.
10.1029/2022GC010496
19 of 24
setting like North Pond, the nearby basement exposures offer a pathway for transmission of a significant diffusive
response laterally in permeable basement beneath the sediment cap, resulting in a significant phase difference
between the ocean loading and formation response signals.
Davis etal.(2000) developed a conceptual model for lateral transmission in permeable basement that depends
on distance from the nearest outcrop, and showed that the amplitude ratio and phase of the formation responses
can constrain average formation permeability—as long as the data span an appropriate range of distances from
the nearest basement outcrops. They incorporated amplitude ratios and phases for the brief period of good 1997
data available from the CORK in Hole 395A at the time. The amplitude ratios and phases they obtained with
this short data segment are reasonably consistent with the values we obtained for Holes U1382A, U1383B, and
U1383C, but significantly different from the values we obtained for the 2004–2007 data from Hole 395A. The
Davis etal. (2000) model and conclusions are best constrained by tidal response at scaled distances from the
nearest outcrop, x′= distance/(π к T)
1/2, over a range of 0.2–1, where к is the basement hydraulic diffusivity
and T is period of the seafloor loading constituent. Given the limited range of loading signal periods provided
by the semi-diurnal and diurnal tidal constituents that dominate the North Pond response, as well as the limited
and weakly constrained range of distances to outcrop, the North Pond sites by themselves do not provide good
constraints on the nature of any diffusive signal. When combined with observations from other sites, however, a
general consistency can be seen among the results, suggesting that the Davis etal.(2000) estimate for an effective
upper basement permeability on the order of 10
−10m
2 remains appropriate. In a separate detailed study in prepa-
ration, we will explore whether using amplitude ratios and phases for the more extensive range of tidal constitu-
ents resolved by our very long pressure records can better constrain specific values of hydraulic diffusivity and
effective permeability at individual sites.
5.4. Bottom Water Temperature Variations and Implications for Heat Flux Measurement Accuracy
Becker etal.(2021) described a characteristic pattern of variations of bottom water temperatures recorded with
platinum RTDs in the CORK data loggers on the wellheads of Holes U1382A and U1383C, and interpreted
these patterns in terms of geothermal heating with interspersed episodes of cold, dense bottom water incursions
over the deepest sill that bounds North Pond on its southwest side. Nearly identical variations from 2012 to 2017
were also shown by the RTD in the data logger at Hole U1383B, and the 10-year record from Hole 395A shows
similar variations when smoothed suitably to overcome the coarse digital resolution of individual hourly read-
ings (Figure13). These patterns indicate slow heating at a rate consistent with the average seafloor heat flux in
North Pond over time periods of months to as much as 1.5years, interrupted by shorter periods of relatively rapid
drops in temperature as the bottom water is renewed from the southwest. The largest amplitudes of the temporal
variations are on the order of 0.02–0.03°C, which could affect the accuracy of any heat flux measurements made
with very short probes. Davis etal.(2003) presented an analysis of the propagation of bottom-water temperature
variations into the subseafloor and resultant perturbations in temperature gradients below seafloor. That analysis
indicates that variations like those recorded at all four of our sites are unlikely to perturb gradients below 1m
subseafloor by an amount that is significant relative to the sediment geothermal gradients in North Pond. As the
1989 and 2009 detailed heat flux surveys were made with multi-thermistor probes 5 or 6m long, the bottom water
variations recorded in North Pond should have had very little effect on the accuracy of the reported heat flux
values (Langseth etal.,1992; Schmidt-Schierhorn etal.,2012), and those authors reported no indications for any
anomalous temperature readings in the upper m.
6. Summary and Conclusions
Basement formation pressures and temperatures recorded over two decades (1997–2017) in four sealed borehole
observatories in North Pond provide new constraints on the nature of low-temperature hydrothermal circulation
beneath isolated sediment ponds that may be broadly characteristic of circulation in ocean crust formed at slow
spreading rates in thinly and discontinuously sedimented regions. North Pond is located in crust ∼7–8M.y. old
west of the mid-Atlantic Ridge and is ∼8×15km in extent, elongated in a south-southwest to north-northeast
direction subparallel to the strike of the mid-Atlantic Ridge. The borehole observatories were positioned as two
closely spaced pairs, one ∼1km from the southeast edge of North Pond where sediments are ∼90m thick (orig-
inal location of DSDP Site 395) and the other ∼6km away ∼1km from the northeast edge where sediments are

Geochemistry, Geophysics, Geosystems
BECKER ETAL.
10.1029/2022GC010496
20 of 24
40–50m thick. Closer to the center of North Pond, sediments are up to 300m thick, particularly in its southwest
section, so the borehole observatory results are likely to be primarily diagnostic of hydrologic conditions and
processes near the edges of the sediment pond. However, additional constraints on upper basement temperatures,
including in the more thickly sedimented sections of North Pond, are provided by extrapolation of 74 heat flow
measurements collected in 1989 and 2009 that lie along six seismic profiles (three along North Pond strike, three
across strike) collected in 2009 (Langseth etal.,2018; Schmidt-Schierhorn etal.,2012).
Inferences drawn from these results include the following:
• Before it was sealed in 1997, Hole 395A exhibited a 21-year history of downhole flow of ocean bottom
water into uppermost permeable basement, leading to initial interpretations that basement formation fluids
were likely underpressured relative to hydrostatic (e.g., Becker etal.,2001). However, results from the three
borehole observatories that penetrate >50m into basement show super-hydrostatic basement fluid pressures,
on the order 10 kPa at 100m subbasement. A simple model shows that positive differential pressures in
permeable basement are probably a consequence of the hydrologic structure beneath the thermally insulat-
ing, relatively impermeable but laterally restricted sediment cover in the sediment pond that is surrounded
by unsedimented or thinly sedimented, cool, and elevated igneous crust. In contrast, the fourth observatory,
Figure 13. Temporal variability of bottom-water temperatures recorded at Holes U1383B and 395A. (a) Variability recorded
from 2012 to 2017 at Hole U1383B. See Becker etal.(2021) for nearly identical records from Holes U1382A and U1383C
and interpretation in terms of slow geothermal heating and episodic renewal of bottom water by incursions of cold, dense
bottom water over deepest sill on southwest side of North Pond. (b) Similar variability in the bottom-water temperature record
displayed from 1997 to 2007 at Hole 395A, revealed when the original data (gray dots) were smoothed with a 48-hr boxcar
filter.

Geochemistry, Geophysics, Geosystems
BECKER ETAL.
10.1029/2022GC010496
21 of 24
which penetrates only the uppermost 45m of the igneous crust, displayed a slight negative differential pres-
sure in the uppermost few tens of meters of basement, which we interpret as being due to the local hydrologic
regime in the immediate vicinity of the sediment pond perimeter. These observations are consistent with
recent numerical simulations of the circulation system (Price etal., 2022) that indicate upflow beneath the
thickly sedimented central region of North Pond, with rapid lateral fluid flow or convective heat exchange in
permeable upper basement toward the thinly sedimented edges of the basin.
• When adjusted for the thermal states at the two pairs of observatories, the formation pressure observations
indicate only a very small lateral pressure gradient in the uppermost igneous crust beneath the sediment fill
of North Pond. This is consistent with observations in more continuously sedimented examples in the eastern
and northeastern Pacific (e.g., Becker etal.,2004; Davis & Becker,1998, 2002, 2004), and is suggestive of
high permeability enabling significant lateral flow driven by small differences in pressure.
• Lateral temperature gradients in the uppermost igneous crust beneath North Pond are larger than recorded
in the eastern Pacific examples under regionally continuous sediment cover, but similar to those seen where
sediment covered igneous basement lies adjacent to areas of extensive basement outcrop (e.g., Davis &
Becker,2004). Highest upper basement temperatures occur beneath the most thickly sedimented sections of
North Pond where they are ∼10K above the bottom water value of 2.5°C. Upper basement temperatures are
lower near the southwest, northwest, and northeast edges at ∼2.5–5K above bottom water temperature, and
lowest near the southeast edge where Holes 395A and U1382A are located. The general pattern is consistent
with rapid lateral fluid flow or convective heat exchange in permeable basement between the central region
of the basin and the edges (Price etal.,2022), although there are also some local effects associated with both
the basement ridge underneath the northwest section of North Pond and shallow basement topography in the
northeast section. The presence of any regional directional trend of circulation in uppermost basement as
suggested by earlier models (Langseth etal.,1984, 1992) cannot be resolved.
• The borehole observations show temporal variability in formation temperatures and pressures that may be
consistent with indications for nonsteady state circulation modeled by Price etal.(2022).
• The four observatories show basement formation responses to seafloor tidal loading that remain rela-
tively constant in amplitude and phase over time and are consistent with high upper basement permeability
(∼10
−10m
2) as previously concluded by Davis etal.(2000) using the earliest pressure data from Hole 395A.
• All four observatories recorded quasi-regular variations in bottom water temperatures marked by long periods
(several months to >1year) of slow temperature rises punctuated by sharper temperature drops over times-
cales of days to weeks. These are interpreted by Becker etal.(2021) to be caused by slow geothermal heating
interrupted by episodic intrusions of modified Antarctic Bottom Water (AABW) from the west over the deep-
est sill that bounds North Pond on its southwest side. Using the analysis of Davis etal.(2003), we show in this
paper that these variations are too small to be responsible for gradient perturbations that would have affected
the accuracy of seafloor heat flux measurements in the area.
When IODP installed the new observatories in 2011 (Edwards etal.,2012), part of the rationale was to use the
original Hole 395A location as a reference for testing the hypothesis of south-to-north flow in basement from this
location to the monitoring holes at Site U1383 ∼6km to the north. With hindsight based on both our results and
those of Price etal.(2022), a site in the thickly sedimented southwest section of the pond might have been highly
illuminating in terms of hydrogeological and microbiological processes. If another borehole monitoring site were
ever to be established in North Pond, we would suggest this as the best location, and similar reasoning would
apply if another isolated sediment pond is ever selected for future drilling and borehole observatory installation.
For North Pond, such a new location would also shed further light on the bottom-water renewal processes inferred
by Becker etal.(2021) to be caused by episodic incursion of cold, dense bottom water from the west over the
deepest sill that bounds North Pond on its southwest edge.
Data Availability Statement
The CORK pressure and temperature data presented in this manuscript are all freely available from the Marine
Geoscience Data System (MGDS) at www.marine-geo.org (Becker & Davis,2022; Becker & Wheat, 2022a,
2022b, 2022c; Becker etal.,2015a, 2015b, 2015c, 2018a, 2018b, 2018c). Geothermal data from the 1989 survey
are available at www.pangaea.org (Langseth etal.,2018). Geothermal and single-channel seismic data from the
2009 survey are available as appendices in Schmidt-Schierhorn etal. (2012). The base map for Figure1a was

Geochemistry, Geophysics, Geosystems
BECKER ETAL.
10.1029/2022GC010496
22 of 24
produced using GeoMapApp (www.geomapapp.org)/CC by (Ryan etal.,2009). Multibeam bathymetry data used
to produce Figure1b are available at Pangaea.de (Villinger etal.,2018).
References
Anderson, R. N., & Hobart, M. A. (1976). The relation between heat flow, sediment thickness, and age in the eastern Pacific. Journal of Geophys-
ical Research, 81(17), 2968–2989. https://doi.org/10.1029/JB081i017p02968
Anderson, R. N., Hobart, M. A., & Langseth, M. G. (1979). Geothermal convection through oceanic crust and sediments in the Indian Ocean.
Science, 204(4395), 828–832. https://doi.org/10.1126/science.204.4395.828
Bartetzko, A., Pezard, P., Goldberg, D., Sun, Y.-F., & Becker, K. (2001). Volcanic stratigraphy of DSDP/ODP Hole 395A: An interpretation using
well-logging data. Marine Geophysical Researches, 22(2), 111–127. https://doi.org/10.1023/A:1010359128574
Becker, K. (1990). Measurements of the permeability of the upper oceanic crust at Hole 395A, ODP Leg 109. In Proceedings ODP, Scientific
Results (Vol. 106/109,pp.213–222). Ocean Drilling Program. https://doi.org/10.2973/odp.proc.sr.106/109.146.1990
Becker, K., Bartetzko, A., & Davis, E. E. (2001). Leg 174B synopsis: Revisiting Hole 395A for logging and long-term monitoring of off-axis
hydrothermal processes in young oceanic crust. Proceedings ODP, Scientific Results, 174B, 1–12. https://doi.org/10.2973/odp.proc.
sr.174B.130.2001
Becker, K., & Davis, E. E. (2022). Pressure and temperature time-series data from the CORK in Hole U395A in North Pond acquired from
1997–2007. IEDA. https://doi.org/10.26022/IEDA/330948
Becker, K., Davis, E. E., Spiess, F. N., & de Moustier, C. P. (2004). Temperature and video logs from the upper oceanic crust, Holes 504B
and 896A, Costa Rica Rift flank: Implications for the permeability of upper oceanic crust. Earth and Planetary Science Letters, 222(3–4),
881–896. https://doi.org/10.1016/j.epsl.2004.03.033
Becker, K., Langseth, M. G., & Hyndman, R. D. (1984). Temperature measurements in Hole 395A, Leg 78B. Initial Reports DSDP (Vol.
78B,pp.689–698). https://doi.org/10.2973/dsdp.proc.78b.105.1984
Becker, K., Thomson, R. E., Davis, E. E., Villinger, H., & Wheat, C. G. (2021). Geothermal heating and episodic cold-seawater intrusions
into an isolated ridge-flank basin near the Mid-Atlantic Ridge. Communications Earth & Environment, 2(1), 226. https://doi.org/10.1038/
s43427-021-00297-2
Becker, K., Villinger, H., & Davis, E. (2015a). CORK-II Pressure Data from Hole U1382A along the western flank of the Mid-Atlantic Ridge
(2011–2014). IEDA. https://doi.org/10.1594/IEDA/322273
Becker, K., Villinger, H., & Davis, E. (2015b). CORK-II Pressure Data from Hole U1383C along the western flank of the Mid-Atlantic Ridge
(2011–2014). IEDA. https://doi.org/10.1594/IEDA/322274
Becker, K., Villinger, H., Davis, E., & Wheat, C. G. (2015c). CORK-Lite Pressure Data from Hole U1383B along the western flank of the
Mid-Atlantic Ridge (2012–2014). IEDA. https://doi.org/10.1594/IEDA/322275
Becker, K., & Wheat, C. G. (2022a). Temperature time-series data from the CORK-II in Hole U1382A in North Pond acquired from 2011–2017.
IEDA. https://doi.org/10.26022/IEDA/330949
Becker, K., & Wheat, C. G. (2022b). Temperature time-series data from the CORK-Lite in Hole U1383B in North Pond acquired from 2012–2017.
IEDA. https://doi.org/10.26022/IEDA/330950
Becker, K., & Wheat, C. G. (2022c). Temperature time-series data from the CORK-II in Hole U1383C in North Pond acquired from 2011–2013.
IEDA. https://doi.org/10.26022/IEDA/330951
Becker, K., Wheat, C. G., & Davis, E. (2018a). CORK-II Pressure Data from Hole U1382A along the western flank of the Mid-Atlantic Ridge
(2014–2017). IEDA. https://doi.org/10.1594/IEDA/324535
Becker, K., Wheat, C. G., & Davis, E. (2018b). CORK-II Pressure Data from Hole U1383C along the western flank of the Mid-Atlantic Ridge
(2014–2017). IEDA. https://doi.org/10.1594/IEDA/324536
Becker, K., Wheat, C. G., & Davis, E. (2018c). CORK-Lite Pressure Data from Hole U1383B along the western flank of the Mid-Atlantic Ridge
(2014–2017). IEDA. https://doi.org/10.1594/IEDA/324537
Cera, T. (2011). Tidal analysis and prediction in Python (TAPPY). Retrieved from https://github.com/timcera/tappy
Corliss, J. B., Dymond, J., Gordon, L. I., Edmond, J. M., Von Herzen, R. P., Ballard, R. D., et al. (1979). Submarine thermal springs on the
Galápagos Rift. Science, 203(4385), 1073–1083. https://doi.org/10.1126/science.203.4385.1073
Davis, E. E., & Becker, K. (1994). Formation temperatures and pressures in ODP Holes 857D and 858G: One year of CORK observations in the
Middle Valley sedimented rift. Proceedings ODP, Scientific Results, 139, 649–666. https://doi.org/10.2973/odp.proc.sr.139.255.1994
Davis, E. E., & Becker, K. (1998). Borehole observatories record driving forces for hydrothermal circulation in young oceanic crust. EOS, Trans-
actions American Geophysical Union, 79(31), 369–370. https://doi.org/10.1029/98EO00275
Davis, E. E., & Becker, K. (2002). Observations of natural-state fluid pressures and temperatures in young oceanic crust and inferences regarding
hydrothermal circulation. Earth and Planetary Science Letters, 204(1–2), 231–248. https://doi.org/10.1016/S0012-821X(02)00982-2
Davis, E. E., & Becker, K. (2004). Observations of temperature and pressure: Constraints on ocean crustal hydrologic state, properties, and flow.
In E. E. Davis & H. Elderfield (Eds.), Hydrogeology of the Oceanic Lithosphere (pp.225–271). Cambridge University Press.
Davis, E. E., Becker, K., Kyo, M., & Kimura, T. (2018). Foundational experiences and recent advances in long-term deep-ocean borehole observa-
tories for hydrologic, geodetic, and seismic monitoring. Marine Technology Society Journal, 52(5), 74–86. https://doi.org/10.4031/mtsj.52.5.4
Davis, E. E., Becker, K., Pettigrew, T., Carson, B., & MacDonald, R. (1992). CORK: A hydrologic seal and downhole observatory for deep ocean
boreholes. Proceedings of the Ocean Drilling Program, Initial Reports, 139, 43–53. https://doi.org/10.2973/odp.proc.ir.139.103.1992
Davis, E. E., Chapman, D. S., Wang, K., Villinger, H., Fisher, A. T., Robinson, S. W., etal. (1999). Regional heat flow variations across the sedi-
mented Juan de Fuca Ridge eastern flank: Constraints on lithospheric cooling and lateral hydrothermal heat transport. Journal of Geophysical
Research, 104(B8), 17675–17688. https://doi.org/10.1029/1999JB900124
Davis, E. E., LaBonte, A., He, J., Becker, K., & Fisher, A. T. (2010). Thermally stimulated “runaway” downhole flow in a superhydrostatic
ocean crustal borehole: Observations, simulations, and inferences regarding crustal permeability. Journal of Geophysical Research, 115(B7),
B07102. https://doi.org/10.1029/2009JB006986
Davis, E. E., & Lister, C. R. B. (1977). Heat flow measurements over the Juan de Fuca Ridge: Evidence for widespread hydrothermal circulation
in a highly heat transportive crust. Journal of Geophysical Research, 82(30), 4845–4860. https://doi.org/10.1029/JB082i030p04845
Davis, E. E., Villinger, H., MacDonald, R. D., Meldrum, R., & Grigel, J. (1997). A robust rapid-response probe for measuring bottom-hole
temperatures in deep-ocean boreholes. Marine Geophysical Researches, 19(3), 267–281. https://doi.org/10.1023/A:1004292930361
Acknowledgments
This research was generously supported
by the Geological Survey of Canada, the
DFG (German Science Foundation, for
R/V Maria S. Merian shiptime in 2012
and 2014), and the following grants
from the National Science Founda-
tion: OCE-9530426, OCE-0946795,
OCE-1060855, and OCE-1536601 to
KB, OCE-1060634 to K. J. Edwards
for ROV support in 2012 and 2014, and
OCE-1536623 to C. G. Wheat for ship-
time and ROV support in 2017. We thank
the officers and crews of D/V JOIDES
Resolution, R/V Maria S. Merian, R/V
Atlantis, and the ROV Jason team for
outstanding support of at-sea operations.
Early versions of this manuscript bene-
fited from discussions with C. G. Wheat,
and the final version was improved
further after helpful reviews by A. N.
Price and an anonymous reviewer.

Geochemistry, Geophysics, Geosystems
BECKER ETAL.
10.1029/2022GC010496
23 of 24
Davis, E. E., Wang, K., Becker, K., & Thomson, R. E. (2000). Formation-scale hydraulic and mechanical properties of oceanic crust
inferred from pore pressure response to periodic seafloor loading. Journal of Geophysical Research, 105(B6), 13423–13435. https://doi.
org/10.1029/2000JB900084
Davis, E. E., Wang, K., Becker, K., Thomson, R. E., & Yashayaev, I. (2003). Deep-ocean temperature variations and implications for errors in
seafloor heat flow determinations. Journal of Geophysical Research, 108(B1), 2034. https://doi.org/10.1029/2001JB001695
Edwards, K. J., Wheat, C. G., Orcutt, B. N., Hulme, S., Becker, K., Jannasch, H., etal. (2012). Design and deployment of borehole observatories
and experiments during IODP Expedition 336, Mid-Atlantic Ridge flank at North Pond. In Proceedings IODP (Vol. 336,p.2). https://doi.
org/10.2204/iodp.proc.336.109.2012
Gable, R., Morin, R. H., & Becker, K. (1992). Geothermal state of DSDP Holes 333A, 395A, and 534A: Results from the Dianaut program.
Geophysical Research Letters, 19(5), 505–508. https://doi.org/10.1029/92GL00333
Hussong, D. M., Fryer, P.B., Tuthill, J. D., & Wipperman, L. K. (1978). The geological and geophysical setting near DSDP Site 395, North
Atlantic Ocean. Initial Reports DSDP, 45, 23–27. https://doi.org/10.2973/dsdp.proc.45.102.1979
Hutnak, M., Fisher, A. T., Harris, R., Stein, C., Wang, K., Spinelli, G., etal. (2008). Large heat and fluid fluxes driven through mid-plate outcrops
on ocean crust. Nature Geoscience, 1(9), 611–614. https://doi.org/10.1038/ngeo264
Hyndman, R. D., & Salisbury, M. H. (1984). The physical nature of young upper oceanic crust on the Mid-Atlantic Ridge. In Deep Sea Drilling
Project Hole 395A. Initial Reports DSDP (Vol. 78B,pp.839–848). https://doi.org/10.2973/dsdp.proc.78b.118.1984
Karato, S.-I., & Becker, K. (1983). Porosity and hydraulic properties of sediments from the Galapagos Spreading Center and their relationship to
hydrothermal circulation in the oceanic crust. Journal of Geophysical Research, 88(B2), 1009–1017. https://doi.org/10.1029/JB088iB02p01009
Kopietz, J., Becker, K., & Hamano, Y. (1990). Temperature measurements in Hole 395A, Leg 109. Proceedings of the Ocean Drilling Program,
Scientific Results, 106/109, 197–203. https://doi.org/10.2973/odp.proc.sr.106109.144.1990
Langseth, M. G., Becker, K., Von Herzen, R. P., & Schultheiss, P. (1992). Heat and fluid flux through sediment on the western flank of the
Mid-Atlantic Ridge: A hydrogeological study of North Pond. Geophysical Research Letters, 19(5), 517–520. https://doi.org/10.1029/92GL00079
Langseth, M. G., Hyndman, R. D., Becker, K., Hickman, S. H., & Salisbury, M. H. (1984). The hydrogeological regime of isolated sediment
ponds in mid-oceanic ridges. Initial Reports DSDP, 78B, 825–837. https://doi.org/10.2973/dsdp.proc.78b.117.1984
Langseth, M. G., Von Herzen, R. P., Becker, K., Müller, P., & Villinger, H. (2018). Heat flux in sedimented ponds, MAR-flanks 22–24°N, RV
Atlantis II 123-2, 1989 [Dataset]. PANGAEA. https://doi.org/10.1594/PANGAEA.896168
Lister, C. R. B. (1972). On the thermal balance of a mid-ocean ridge. Geophysical Journal International, 26(5), 515–535. https://doi.org/10.1111/
j.1365-246X.1972.tb05766.x
Millero, F. J., Chen, C., Bradshaw, A., & Schleicher, K. (1980). A new high pressure equation of state for seawater. Deep Sea Research Part A,
27(3–4), 255–264. https://doi.org/10.1016/0198-0149(80)90016-3
Morin, R. H., Hess, A. E., & Becker, K. (1992). In situ measurements of fluid flow in DSDP Holes 395A and 534A: Results from the DIANAUT
program. Geophysical Research Letters, 19(5), 509–512. https://doi.org/10.1029/91GL02947
Mottl, M. J., & Wheat, C. G. (1994). Hydrothermal circulation through mid-ocean ridge flanks: Fluxes of heat and magnesium. Geochimica et
Cosmochimica Acta, 58(10), 2225–2237. https://doi.org/10.1016/0016-7037(94)90007-8
Polster, A., Fabian, M., & Villinger, H. (2009). Effective resolution and drift of Paroscientific pressure sensors derived from long-term seafloor
measurement. Geochemistry, Geophysics, Geosystems, 10(8), Q08008. https://doi.org/10.1029/2009GC002532
Price, A. N., Fisher, A. T., Stauffer, P.H., & Gable, C. W. (2022). Numerical simulation of cool hydrothermal processes in the upper volcanic crust
beneath a marine sediment pond: North Pond, North Atlantic Ocean. Journal of Geophysical Research: Solid Earth, 127(1), e2021JB023158.
https://doi.org/10.1029/2021JB023158
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
Schmidt-Schierhorn, F., Kaul, N., Stephan, S., & Villinger, H. (2012). Geophysical site survey results from North Pond (Mid-Atlantic Ridge). In
Proceedings IODP (Vol. 336,p.2). https://doi.org/10.2204/iodp.proc.336.107.2012
Sclater, J. G., Von Herzen, R. P., Williams, D. L., Anderson, R. N., & Klitgord, K. (1974). The Galapagos Spreading Centre: Heat-flow low on
the North Flank. Geophysical Journal International, 38(3), 609–626. https://doi.org/10.1111/j.1365-246X.1974.tb05432.x
Shipboard Scientific Party. (1978). Site 395: 23°N, Mid-Atlantic Ridge. Initial Reports DSDP, 45, 131–264. https://doi.org/10.2973/dsdp.
proc.45.107.1979
Shipboard Scientific Party. (1998). Site 1074. In Proceedings ODP. Initial Reports (Vol. 174B,pp.25–46). https://doi.org/10.2973/odp.proc.
ir.174B.103.1998
Snelgrove, S. H., & Forster, C. B. (1996). Impact of seafloor sediment permeability and thickness on off-axis hydrothermal circulation: Juan de
Fuca Ridge eastern flank. Journal of Geophysical Research, 101(B2), 2915–2925. https://doi.org/10.1029/95JB03115
Spiess, F. N., Macdonald, K. C., Atwater, T., Ballard, R., Carranza, A., Cordoba, D., etal. (1980). East Pacific Rise: Hot springs and geophysical
experiments. Science, 207(4438), 1421–1433. https://doi.org/10.1126/science.207.4438.1421
Spinelli, G. A., Giambalvo, E. R., & Fisher, A. T. (2004). Sediment permeability, distribution, and influence on fluxes in oceanic basement. In E.
E. Davis & H. Elderfield (Eds.), Hydrogeology of the Oceanic Lithosphere (pp.151–188). Cambridge University Press.
Stein, C. A., & Stein, S. (1994). Constraints on hydrothermal heat flux through the oceanic lithosphere from global heat flow. Journal of Geophys-
ical Research, 99(B2), 3081–3095. https://doi.org/10.1029/93JB02222
Stein, J. S., & Fisher, A. T. (2001). Multiple scales of hydrothermal circulation in Middle Valley, northern Juan de Fuca Ridge: Physical constraints
and geological models. Journal of Geophysical Research, 106(B5), 8563–8580. https://doi.org/10.1029/2000JB900395
van der Kamp, G., & Gale, J. E. (1983). Theory of Earth tide and barometric effects in porous formations with compressible grains. Water
Resources Research, 19, 538–544. https://doi.org/10.1029/WR019i002p00538
Villinger, H., Strack, A., Gaide, S., & Thal, J. (2018). Gridded bathymetry of North Pond (MAR) from multibeam echosounder EM120 and
EM122 data of cruises MSM20/5 (2012) and MSM37 (2014) [Dataset]. Department of Geosciences, Bremen University, PANGAEA. https://
doi.org/10.1594/PANGAEA.889439
Villinger, H. W., Müller, P., Bach, W., Becker, K., Orcutt, B. N., Kaul, N., & Wheat, C. G. (2019). Evidence for low-temperature diffuse
venting at North Pond, western flank of the Mid-Atlantic Ridge. Geochemistry, Geophysics, Geosystems, 20(6), 2572–2584. https://doi.
org/10.1029/2018GC008113
Wang, K., & Davis, E. E. (1996). Theory for the propagation of tidally induced pore pressure variations in layered subseafloor formations. Journal
of Geophysical Research, 101, 11483–11495. https://doi.org/10.1029/96JB00641
Wheat, C. G., Becker, K., Villinger, H., Orcutt, B. N., Fournier, T., Hartwell, A., & Paul, C. (2020). Subseafloor cross-hole tracer exper-
iment reveals hydrologic properties, heterogeneities, and reactions in slow-spread crust. Geochemistry, Geophysics, Geosystems, 21(1),
e2019GC008804. https://doi.org/10.1029/2019GC008804

Geochemistry, Geophysics, Geosystems
BECKER ETAL.
10.1029/2022GC010496
24 of 24
Wheat, C. G., Edwards, K. J., Pettigrew, T., Jannasch, H. W., Becker, K., Davis, E. E., etal. (2012). CORK-Lite: Bringing legacy boreholes back
to life. Scientific Drilling, 14, 39–43. https://doi.org/10.2204/iodp.sd.14.05.2012
Wheat, C. G., & Mottl, M. J. (2004). Geochemical fluxes through mid-ocean ridge flanks. In E. E. Davis & H. Elderfield (Eds.), Hydrogeology
of the Oceanic Lithosphere (pp.627–658). Cambridge University Press.
Williams, D. L., Von Herzen, R. P., Sclater, J. G., & Anderson, R. N. (1974). The Galapagos Spreading Centre: Lithospheric cooling and hydro-
thermal circulation. Geophysical Journal International, 38(3), 587–608. https://doi.org/10.1111/j.1365-246X.1974.tb05431.x