Potential on-shore and off-shore reservoirs

for CO2 sequestration in Central Atlantic

magmatic province basalts
David S. Goldberga, Dennis V. Kenta,b,1, and Paul E. Olsena
a
Lamont-Doherty Earth Observatory, 61 Route 9W, Palisades, NY 10964; and bEarth and Planetary Sciences, Rutgers University, Piscataway, NJ 08854.

Contributed by Dennis V. Kent, November 30, 2009 (sent for review October 16, 2009)

Identifying locations for secure sequestration of CO2 in geological seafloor (16) may offer potential solutions to these additional
formations is one of our most pressing global scientific problems. issues that are more problematic on land. Deep-sea aquifers
Injection into basalt formations provides unique and significant are fully saturated with seawater and typically capped by imper-
advantages over other potential geological storage options, includ- meable sediments. The likelihood of postinjection leakage of
ing large potential storage volumes and permanent fixation of car- CO2 to the seafloor is therefore low, reducing the potential
bon by mineralization. The Central Atlantic Magmatic Prov- impact on natural and human ecosystems (8). Long after CO2
ince basalt flows along the eastern seaboard of the United States injection, the consequences of laterally displaced formation water
may provide large and secure storage reservoirs both onshore and to distant locations and ultimately into the ocean, whether by
offshore. Sites in the South Georgia basin, the New York Bight engineered or natural outflow systems, are benign. For more than
basin, and the Sandy Hook basin offer promising basalt-hosted a decade, subseabed CO2 sequestration has been successfully
conducted at >600 m depth in the Utsira Formation as part of

SUSTAINABILITY
reservoirs with considerable potential for CO2 sequestration due
the Norweigan Sleipner project (17). The Central Atlantic Mag-

SCIENCE
to their proximity to major metropolitan centers, and thus to large
industrial sources for CO2. Onshore sites are suggested for cost- matic Province (CAMP) basalt formations formed during rifting
effective characterization studies of these reservoirs, although of the Pangea supercontinent (18, 19) and are present near the
offshore sites may offer larger potential capacity and additional highly industrial US and western European coastlines. In this
long-term advantages for safe and secure CO2 sequestration. study, we further consider the viability of continental flood basalts
as potential targets for geological CO2 sequestration (9)—speci-
Eastern United States ∣ greenhouse gas ∣ Jurassic ∣ lavas ∣ rift fically identifying CAMP basalts in both onshore and near-shore
locations along the eastern North American seaboard.

I
Downloaded from https://www.pnas.org by 173.219.167.170 on June 3, 2023 from IP address 173.219.167.170.

n recent years, scientific interest in targeting geological reser-

voirs for CO2 sequestration in order to stabilize our increasing In Situ Mineral Carbonation
greenhouse gas concentrations in the atmosphere has been grow- Mineral carbonation involves the chemical combination of CO2 ,
ing. Among these targets, the injection of industrial CO2 into water, and metal cations to form carbonates. Oelkers et al. (6)
deep saline aquifers or depleted oil and gas reservoirs with large recently summarized the chemistry of mineral fixation of CO2 ,
potential storage capacity and geographic ubiquity are common and Matter and Kelemen (13) highlight the geochemical poten-
choices (1–5). Several basalt and ultramafic rock reservoirs also tial of various basalt and ultramafic formations for CO2 se-
have been proposed in a variety of tectonic settings, including, for questration. Takahashi et al. (20) present a general geochemical
example, midoceanic islands (6, 7), deep-sea basalt crust (8), and model for mineral carbonation in basalt and suggest that miner-
continental flood basalts (9). The effectiveness of any of these alization rates may be rapid. Shaef et al. (21) show that dissolu-
reservoirs depends strongly on the storage capacity, retention tion and precipitation reactions in basalt samples tested under
time, reservoir stability, and the risk for leakage (10, 11). Basalt laboratory conditions can vary considerably, depending on the
and ultramafic sequestration targets provide unique advantages glass content and composition of the host minerals. Matter et al.
over other geological storage options due to their potential for (22) suggest that fracture porosity and connectivity in the
chemical reaction of injected CO2 with the host formation to basalt aquifer influence the in situ rates, which will differ signifi-
produce stable, nontoxic ðCaþþ ; Mgþþ ; Feþþ ÞCO3 void-filling cantly from those measured in the laboratory. Carbonate precip-
minerals. In nature, the same processes occur in association with itation may also reduce in situ porosity and permeability within
serpentinization and surface weathering in ultramafic and mafic basalt aquifers over time (6).
Field injection tests at basalt sites are essential to evaluate
rocks exposed to water, the breakdown of silicates into clays, and
these processes in each potential environment, however, only
the precipitation of carbonates (12, 13). Such geochemical con-
one in situ experiment has been conducted in a basalt aquifer
version decreases the risk of leakage over the long term, so long
to date. Matter et al. (23) conducted a small-scale injection
as sufficiently large volumes of CO2 can be stored in these for-
experiment in the Palisades sill (CAMP diabase in Newark Basin)
mations and retained long enough to allow for the chemical
to investigate the in situ rates of reaction using dissolved CO2 at
reactions and mineral precipitation to occur (14, 15). Several
low partial pressures. They demonstrated large decreases in
pilot injection projects in basalt rock provinces are currently
Mgþþ and Caþþ concentration in recovered water samples within
underway—such as the CarbFix project in Iceland (6) and the
200 h of injection and suggested that two processes, mixing be-
Columbia River plateau in the United States (9)—that will pro-
tween the injected solution and aquifer water and the release of
vide critical new information about the injectability and rates of
cations from water–rock dissolution, neutralized the introduced
CO2 –water–rock reactions in basalt-hosted reservoirs.
carbonic acid. In situ precipitation rates were not measured in
Several additional challenges remain for geological sequestra-
tion in basalt aquifers, including (i) adequate availability of H2 O
in the subsurface to allow for mineral carbonation, (ii) sufficient Author contributions: D.S.G. and D.V.K. designed research; D.S.G., D.V.K., and P.E.O.
flow and outlet for water displaced by injected CO2 , and (iii) as- performed research; D.S.G., D.V.K., and P.E.O. analyzed data; and D.S.G. wrote the paper.
sured integrity of cap rocks for CO2 to remain buried for long The authors declare no conflict of interest.
periods of time. CO2 sequestration in sediments below the 1
To whom correspondence should be addressed. E-mail: [email protected].

www.pnas.org/cgi/doi/10.1073/pnas.0913721107 PNAS ∣ January 26, 2010 ∣ vol. 107 ∣ no. 4 ∣ 1327–1332

 this experiment, however, and laboratory results from Shaef et al. parated by sedimentary interbeds. The boundaries between suc-
(21) indicated that the most rapid precipitation rates among sev- cessive flows are usually highly vesicular and represent intervals
eral flood basalt samples were for those from the Newark that have undergone rapid cooling as each flow comes in contact
Basin. If large and secure geological reservoirs are identified with an underlying one and, importantly, the atmosphere or water
and tested in flood basalts in close proximity to major industrial above. This often creates highly fractured and brecciated inter-
centers, such as the CAMP basalt formations along the US east vals at the bottoms and especially the tops of flow unit boundaries
coast, enormous volumes of CO2 could be permanently stored as compared with their interiors. Fig. 2 illustrates a schematic of a
and sequestered by mineral fixation. flow-top and interior sequence and actual core images recovered
across a flow boundary from the Orange Mountain basalt in the
Central Atlantic Magmatic Province Newark Basin. The interior of some flows are densely fractured as
Along the eastern North American seaboard, numerous onshore well, notably the thick lower Preakness Basalt in the Newark Ba-
and offshore basins hold thick sequences of continental sedi- sin (Fig. 3). Not only is this fracture pattern spread throughout
ments that were deposited during the Triassic and Jurassic. These the known geographic extent of this flow, it is present in other
basins parallel the margin of the US east coast and, along with basins, specifically in correlative flows within the Deerfield,
basins on the African and European conjugate margins, were Hartford, and Culpeper basins (27).
transected by a huge magmatic event in the Early Jurassic In the continuously cored Orange Mountain basalt, high vesic-
(≈201 Ma) that included radiating dike swarms, associated sills ular porosity is observed at the flow tops; dense, low-porosity, and
and plutons, and massive volumes of basalt flows presently dis- low-permeability basalt is evident in the flow interior. The flow-
persed on four continents (18, 19, 24, 25). Fig. 1 illustrates the top boundaries of flood basalt sequences are potential injection
current-day extent of CAMP basalt formations. Marzoli et al. reservoirs for CO2 sequestration, sealed above and below by thick
(18) suggest that the CAMP was the largest Large Igneous impermeable flow-interior basalt and often by interbedded sedi-
Province in the world, and although mostly eroded now, the mentary units. Within the CAMP, the best-studied basalt for-
original CAMP lava flows may have been equal in volume to mation is the Orange Mountain basalt due to the availability
the Siberian or Deccan traps but distributed more widely of complete drill cores and logs in the Newark Basin, extensive
(≈11 M km2 ). CAMP flows individually tend to have similar outcrops, and other geophysical data. Flow-top boundaries simi-
basalt geochemistry and to be thick (as much as 200-m thick lar to the Orange Mountain basalt formation are proposed here
per cooling unit) (26, 27). as potential targets for CO2 sequestration. Highly fractured flow
Within the major rift basins along the US eastern coast, CAMP interiors such as in the Preakness basalt are another potential
basalt flow formations have been identified as being synchronous target, but quantitative data are thus far lacking.
(19, 28). Several studies (18, 28, 29) provide evidence that the
emplacement for CAMP basalts began ≈202 Ma, recently up- Newark Rift Basin
dated to 201.5 Ma (30), and show similar transected stratigraphy Cores and in situ geophysical measurements in the Newark Basin
and igneous mineralogy across the basins. Many flood basalts illustrate the fractured and porous nature of basalt in CAMP
Downloaded from https://www.pnas.org by 173.219.167.170 on June 3, 2023 from IP address 173.219.167.170.

such as the CAMP and the Columbia River plateau basalts are flows, which are the thickest among known flood basalt provinces.
emplaced during phased eruptive events. This results in a stacked Three flow units have been identified in this area—upper, mid-
sequence of basalt flows with different thicknesses sometimes se- dle, and lower units of the Orange Mountain basalt (31). In Fig. 4,

Fig. 1. Distribution of rift basins in eastern North America and Morocco and the distribution of the CAMP flood basalt, modified from ref. 19. Dotted lines
represent major strike slip fault zones. Detail areas (Inset A and B) are discussed in the text.

1328 ∣ www.pnas.org/cgi/doi/10.1073/pnas.0913721107 Goldberg et al.

 SUSTAINABILITY
Fig. 2. Schematic profile of multiple flow units and core photographs from the Orange Mountain basalt, modified from ref. 28. Flow-top boundary zones

SCIENCE
show considerable vesicular and rubbly pore space as compared to the dense, low-porosity flow interior. Scales are presented in original measurement units
(i.e., core depth in feet).

the middle/lower basalt units are shown as measured by in situ flow interiors—and reaches a porosity value of 20% in the log and
geophysical well logs near Martinsville, NJ (32). Porosity profiles 10% on core samples. Note, however, that core measurements do
are computed from density and neutron measurements as well as not sample the fracture porosity or large void space (secondary
from core samples. Cores were subsampled over the flow bound- porosity) in most formations and often underestimate the mea-
ary and measured using He-pyncnometry to obtain bulk and grain sured in situ porosity. The best estimate of porosity in the inter-
Downloaded from https://www.pnas.org by 173.219.167.170 on June 3, 2023 from IP address 173.219.167.170.

density estimates for comparison with the logs. These data pro- flow zone is greater than the core estimates and likely averages
vide the minimum porosity of each sample, assuming that He 15%. The interpretation of these data suggests massive basalt
invades all of the connected pore space. Very little chemical alter- flow interiors bounded by fractured flow tops and porous inter-
ation is observed by above-baseline natural gamma ray measure- flow contact zones typically occur between the CAMP flows.
ments, and the borehole conditions were consistent through most
of this interval (32). Neutron porosity data are erroneously high Offshore Long Island Rift Basins
over the interval due to the presence of clay minerals that in- Along the eastern North American seaboard, the early Atlantic
crease the neutron count. Importantly, the interflow interval from rifting also generated a series of offshore basins filled with
830–880 ft indicates elevated porosity measurements— Mesozoic sediments, previously identified by seismic, gravity,
a factor of two or more greater than the overlying and underlying and magnetic surveying. Hutchinson et al. (33) mapped four
buried rift basins on the Long Island platform, extending from
the New York Bight to Nantucket, and each extending more than
1; 000 km2 in area and filled with >800 m of sediments (Fig. 5A).
Hutchinson and Grow (34) identified a postrift fault northwest of
the buried New York Bight basin, with no specific evidence of
recent activity, but suggest that this relic fault may be associated
with the origins of the Mesozoic rift basins along the coastline.
Maguire et al. (35) investigated a similar, smaller off-shore basin
near Sandy Hook, NJ. The Sandy Hook and New York Bight
basins have been proposed to be offshore extensions of the
basalt-hosting Hartford rift system and are thus potential loca-
tions of buried CAMP flow basalts. Maguire et al. (35) conducted
a modeling study of the Sandy Hook basin and predicted that
basalt exists under several hundred meters of flat-lying sediments.
These deposits have never been drilled or dated, however their
proposed existence explains seismic reflections observed across
the basin as well as gravity and magnetic anomalies in the region.
Maguire et al. (35) hypothesized that these buried basalt flows are
associated with the eastern extent of the CAMP intrusion and
synchronous with the Orange Mountain basalt flows. Similar
gravity and magnetic anomalies in the New York Bight basin
could be explained by buried basalt flows that exist under a thick
Fig. 3. Photograph of the highly fractured Preakness basalt in the Newark Triassic–Jurassic sediment pile, and associated faults, although
Basin. Photograph courtesy of author (PEO). the early seismic surveys did not adequately resolve these deep

Goldberg et al. PNAS ∣ January 26, 2010 ∣ vol. 107 ∣ no. 4 ∣ 1329
Downloaded from https://www.pnas.org by 173.219.167.170 on June 3, 2023 from IP address 173.219.167.170.

Fig. 4. Geophysical log profiles through lower, middle, and upper flows of the Orange Mountain basalt. Density and porosity profiles indicate an increase
from ≈10% to 20% porosity over the 15-m (50-ft) thick flow-top boundary zone, which amounts to ≈2.25 × 106 m3 open pore volume per km2. Scales are
presented in original measurements units (i.e., log depth in feet and hole diameter in inches).

features (34, 35). Drilling in the Nantucket basin also sampled clay-rich sediments, both advantages for secure storage and rapid
basalt flows of the CAMP (36). carbonation of injected CO2 . For example, considering the small
CAMP extrusives both onshore and offshore were contem- Sandy Hook basin off of New Jersey and assuming that basalt
poraneous, formed episodically, and contain interflow zones like flow-top boundary zones have average bulk porosity of 15%
those observed in the Orange Mountain basalt, potentially pro- and extend over one-third of its area, we estimate that 7 km3
viding a buried storage reservoir with considerable flow-top of basalt flow-top volume and ≈1 km3 of potential pore volume
boundary porosity. In subocean locations, flow tops would be sea- could be available for CO2 storage in this basin alone. This
water filled and capped by dense basalt as well as fine-grained and amounts to a total volume of pore space for injection of up to

Fig. 5. Potential onshore and offshore sites for pilot drilling studies in CAMP basalt along eastern US coastal regions. Black squares indicate initial study areas.
Maps refer to inset box locations in Fig. 1. (A) Sites located in buried Mesozoic rift basins on the Long Island platform (shaded), modified from ref. 33 and nearby
onshore basins (cross-hatched): a, Newark basin; b, Hartford basin; c, Sandy Hook basin; d–e, New York Bight basin; f, Nantucket basin; g–h, Atlantis basin. (B)
Sites located across the extent of buried basalt (hachured) in the South Georgia Rift basin (shaded), modified from ref. 38: i, location proximal to Clubhouse
Crossroads basalt cores; j–k, sites more proximal to seaward-dipping reflectors.

1330 ∣ www.pnas.org/cgi/doi/10.1073/pnas.0913721107 Goldberg et al.

 900 Mt of CO2 , the equivalent of emissions from three or four resolved by drilling and experimental investigation in the area.
1-GW coal-fired power plants for 40 years. The existence of Drilling studies at onshore locations may be a cost-effective
buried basalt flows in the Sandy Hook basin, and in the other means to explore and evaluate these reservoirs prior to evaluating
Mesozoic offshore basins, can only be confirmed by high- deeper offshore targets.
resolution survey mapping followed by drilling in and around
these locations. Conclusions
The injection of CO2 in CAMP basalt formations offers critical
South Georgia Rift Basin advantages for sequestration that warrant pressing investigation.
At the southern margin of the eastern United States, the South The flow-top boundary zones in the CAMP basalt provide ample
Georgia Rift basin underlies the coastal plain of South Carolina pore space for injection and storage of CO2 , and the overlying
and Georgia, extending more the 40; 000 km2 and equal in size to basalt interiors and conformable as well as overlapping sediment
the other eastern US rift basins combined (Fig. 5B). With much cover will act as impermeable geological caps for long periods of
less erosion than its northern counterparts, the stratigraphy of the time. Offshore storage locations provide a further advantage of
South Georgia basin appears to primarily contain thick sequences eventually circulating displaced seawater from the reservoir into
of red beds (37), intrusive diabase sills, and basalt flows (38). The the ocean with benign effect after completing the injection. Along
Clubhouse Crossroads basalt flows have been identified on seis- the eastern US metropolitan coastline, the South Georgia basin,
mic transects and sampled, although incompletely, in three drill the New York Bight basin, and the Sandy Hook basin offer
cores in South Carolina (39, 40). Olsen et al. (19) summarized
potential basalt-hosted reservoirs both onshore and near shore,
radiometric, geochemical, and paleomagnetic dating evidence
with considerable potential for long-term CO2 injection and
from these cores and concluded that the basalt flows were syn-
sequestration. Further research for CO2 storage and fixation in
chronous with other CAMP basalts (≈202 Ma). This large basalt
unit also has been associated with massive offshore basalts iden- basalt formations and for characterization of potential sites is
tified seismically as “seaward-dipping reflectors” that were em- essential: initially, to drill and evaluate the geological properties
placed during the early opening phases of the Atlantic Ocean in characteristic basalt formations at cost-effective onshore loca-
tions, and second, to conduct pilot injection studies in potential

SUSTAINABILITY
(41). These features may also be offshore basalts associated with
CAMP basalt reservoirs. Offshore areas would require high-

SCIENCE
the CAMP, but they have never been drilled or dated.
Based on the US Geological Survey cores in South Carolina, resolution survey mapping to identify, and avoid, neotectonic
the Clubhouse Crossroads basalt is a 250-m thick series of basalt faults that could be reactivated above basalt reservoirs. Several
flows that are chemically similar to CAMP basalts, underlain by suggested initial study locations, both onshore and offshore, in
rift basin sediments, and overlain by coastal plain deposits and, potential CAMP basalt reservoirs are indicated on Fig. 5.
like other CAMP flows, consist of multiple eruptive events (42); Because of their proximity to major metropolitan centers, and
similar properties of the basalt interiors and interflow zones can thus to large industrial sources of CO2 , these CAMP targets could
be anticipated at these sites. Given that the South Georgia Rift ultimately provide safe and secure sequestration of carbon at
basin far exceeds the known area of the other eastern US coastal several practical locations.
Downloaded from https://www.pnas.org by 173.219.167.170 on June 3, 2023 from IP address 173.219.167.170.

basins and is buried at greater depths and with thicker sedimen-

tary cover, it has been suggested previously as a potential geo- ACKNOWLEDGMENTS. New England Research, Inc. measured the Helium
logical sink for CO2 storage (9, 43). The Clubhouse Crossroads pyncnometer densities shown in Fig. 4. We thank C. Broglia and C. Brenner
basalt, in particular, may add enormous potential as a secure res- for assistance with the preparation of Figs. 1, 4, and 5. We also thank
D. Hutchinson and G. Gohn for reviewing the manuscript and B. P. McGrail
ervoir for permanent CO2 sequestration in the South Georgia for helpful discussions. This work was supported by the Lamont-Doherty
Rift basin. Remaining questions about the timing, structure, Earth Observatory (LDEO) of Columbia University. LDEO Contribution
and extent of the Clubhouse Crossroads basalt must first be No. 7314.

1. Bachu S, Gunter WD, Perkins EH (1994) Aquifer disposal of CO2 hydrodynamic and 15. Assayag N, et al. (2009) Water–rock interactions during a CO2 injection field test:
mineral trapping. Energ Convers Manage, 35:269–279. Implications on host rock dissolution and alteration effects. Chem Geol, 265:227–235.
2. Bergman PD, Winter EM (1995) Disposal of carbon dioxide in aquifers in the US. Energ 16. House K, Schrag D, Harvey C, Lackner K (2006) Permanent carbon dioxide storage in
Convers Manage, 36:523–526. deep-sea sediments. Proc Natl Acad Sci USA, 103:12291–12295.
3. Hitchon B (1996) Aquifer Disposal of Carbon Dioxide: Hydrodynamic and Mineral 17. Herzog H, Eliasson B, Kaarstad O (2000) Capturing greenhouse gases. Sci Am 72–79.
trapping—Proof of Concept (Geoscience Publishing Ltd., Alberta), pp 1–165. 18. Marzoli A, et al. (1999) Extensive 200-million year old continental flood basalts of the
4. Holloway S (2001) Storage of fossil fuel-derived carbon dioxide beneath the surface of Central Atlantic Magmatic province. Science, 284:616–618.
the Earth. Annu Rev Energ Env, 26:145–166. 19. Olsen PE, et al. (2003) Cyclo-, magneto-, and bio-stratigraphic constrants on the
5. Jessen K, Kovscek AR, Orr FM Jr (2005) Increasing CO2 storage in oil recovery. Energ duration of the CAMP event and its relationship to the Triasssic-Jurassic boundary.
Convers Manage, 46:293–311. Am Geophy Un Monograph, 136:7–32.
6. Oelkers EH, Gislason SR, Matter J (2008) Mineral carbonation of CO2 . Elements, 20. Takahashi T, Goldberg D, Mutter JC (2000) Secure, long-term sequestration of CO2 in
4:333–337 doi: 10.2113/gselements.4.5.333. deep saline aquifers associated with oceanic and continental basaltic rocks. Proc of the
7. Matter JM, et al. (2009) Permanent carbon dioxide storage into basalt: The CarbFix SRI Int Sympos, Deep Sea & CO2 (The Ship Research Inst, Mitaka, Japan).
pilot project, Iceland. Energy Procedia, 1:3641–3646 doi: 10.1016/j.egypro.2009. 21. Schaef HT, McGrail BP, Owen AT (2009) Basalt-CO2 -H2 O interactions and variability in
02.160. carbonate mineralization rates. Energy Procedia, 1:4899–4906 doi: 10.1016/j.egypro.
8. Goldberg D, Takahashi T, Slagle A (2008) Carbon dioxide sequestration in deep-sea 2009.02.320.
basalt. Proc Natl Acad Sci USA, 105:9920–9925 doi: 10.1073/pnas.0804397105. 22. Matter JM, et al. (2006) Fracture permeability in igneous and metasedimentary rocks:
9. McGrail BP, et al. (2006) Potential for carbon dioxide sequestration in flood basalts. J New results from hydraulic testing and geophysical logging in the Newark Rift Basin.
Geophys Res, 111:1–13 doi: 10.1029/2005JB004169. Hydrogeology, 14:689–699 doi: 10.1007/s10040-005-0456-3.
10. Hawkins DG (2004) No exit: Thinking about leakage from geologic carbon storage 23. Matter JM, Takahashi T, Goldberg D (2007) Experimental evaluation of in situ CO2 -
sites. Energy, 29:1571–1578. water-rock reactions during CO2 injection in basaltic rocks: Implications for
11. Rochelle CA, Czernichowski-Lauriol I, Milodowski AE (2004) The impact of chemical geological CO2 sequestration. Geochem Geophys Geosy, 8:Q02001 doi: 10.1029/
reactions on CO2 storage in geological formations: A brief review. Geological Storage 2006GC001427.
of Carbon Dioxide, (Geological Society of London, London), Special Publ, 233, pp 24. May PR (1971) Pattern of Triassic-Jurassic diabase dikes around the North Atlantic in
87–106. the context of the predrift configuration of the continents. Geol Soc Am Bull,
12. Seifritz W (1990) CO2 disposal by means of silicates. Nature, 345:486 (lett). 82:1285–1292.
13. Matter JM, Kelemen PB (2009) Permanent storage of carbon dioxide in geologic re- 25. Hames W, McHone JG, Renne P, Ruppel C (2003) Introduction. Am Geophy Un Mono-
servoirs by mineral carbonation. Nat Geosci www.nature.com/naturegeoscience graph, 136:16–32.
[online] doi: 10.1038/NGE0683. 26. Puffer JH, Philpotts AR (1988) Eastern North American quartz tholeiites: Geochemistry
14. Gunter WD, Bachu S, Benson S (2004) The role of hydrogeological and geochemical and petrology, in Triassic-Jurassic rifting; continental breakup and the origin of the
trapping in sedimentary basins for secure geological storage of carbon dioxide. Atlantic Ocean and passive margins. Dev Geotecton, 22B:579–605.
Geological Storage of Carbon Dioxide, (Geological Society of London, London), Spe- 27. Olsen PE, Smoot JP, LeTourneau P (2007) Field guide to Earth’s largest continental
cial Publ, 233, pp 129–145. flood basalt province, the CAMP, as expressed in the Culpeper basin, VA. A fieldtrip

Goldberg et al. PNAS ∣ January 26, 2010 ∣ vol. 107 ∣ no. 4 ∣ 1331
 for the 11th annual continental scientific drilling workshop, Arlington, VA (Drilling, 36. Folger DW, Hathaway JC, Chrisopher RA, Valentine PC, Poag CS (1978) Stratigraphic
Observation, and Sampling of the Earth’s Continental Crust, Salt Lake City), pp 1–45. test well, Nantucket Island, Massachusetts. US Geol Survey Circ, 773:1–28.
28. Whiteside JH, et al. (2007) Synchrony between the Central Atlantic Magmatic province 37. Olsen PE, et al. (1991) Rift basins of early Mesozoic age. Geology of the Carolinas, ed
and the Triassic-Jurrassic mass extinction event?. Palaeogeogr Palaeocl, 244:3345–367. Horton W 142–170.
29. Kent DV, Olsen PE (2008) Early Jurassic magnetostratigraphy and paleolatitudes from
38. McBride JH, et al. (1989) Evidences and implications of an extensive early Mesozoic rift
the Hartford continental rift basin (eastern North America): Testing for polarity bias
basin and basalt/diabase sequence beneath the southeast coastal plain. Geol Soc Am
and abrupt polar wander in association with the Central Atlantic Magmatic Province.
J Geophys Res, 113:1–24 doi: 10.1029/2007JB005407. Bull, 101:512–520.
30. Schoene B, et al. (2006) Reassessing the uranium decay constants for geochronology 39. Rankin DW (1977) Studies related to the Charleston SC earthquake of 1886: A prelim-
using ID-TIMS U-Pb data. Geochim Cosmochim Acta, 70:426–445. inary report. US Geol Survey Prof Paper, 1028:1–204.
31. Puffer JH, Student JJ (1992) Volcanic structures, eruptive style, and posteruptive 40. Gohn GS, Houser BH, Schneider RR (1983) Geology of the lower Mesozoic sedimentary
deformation and chemical alteration of the Watchung flood basalts, New Jersey. Geol rocks in the Clubhouse Crossroads test hole #3, near Charleston SC. US Geol Survey Prof
S Am S, 268:261–277. Paper, 1313:D1–D17.
32. Goldberg DS, et al. (1984) Well logging results from the Newark Rift Basin Coring 41. Holbrook WS, Kelemen PB (1993) Large igneous province on the US Atlantic margin
Project. Sci Drilling, 4:267–281.
and implications for magmatism during continental breakup. Nature, 364:433–436.
33. Hutchinson DR, Klitgord KD, Detrick RS (1986) Rift basins of the Long Island platform.
42. Gottfried D, Annell CS, Byerly GS (1983) Geochemistry and tectonic significance of
Geol Soc Am Bull, 97:688–702.
34. Hutchinson DR, Grow JA (1985) New York Bight fault. Geol Soc Am Bull, 96:975–989. subsurface basalts from Charleston, South Carolina: Clubhouse Crossroads test holes
35. Maguire TJ, Sheridan RE, Volkert RA (2004) Geophysical modeling of the northern #2 and #3. US Geol Survey Prof Paper, 1313:A1–A10.
Apppalachian Brompton-Cameron, Central Maine, and Avalon terranes under the 43. Smyth RC, et al. (2007) Potential sinks for geologic storage of carbon dioxide
New Jersey Coastal Plain. Geodynamics, 37:457–485. generated in the Carolinas. US Bureau of Econ Geol, summary report 1–14.
Downloaded from https://www.pnas.org by 173.219.167.170 on June 3, 2023 from IP address 173.219.167.170.

1332 ∣ www.pnas.org/cgi/doi/10.1073/pnas.0913721107 Goldberg et al.