Basin Research (2004) 16, 451–465, doi: 10.1111/j.1365-2117.2004.00242.x Late Cretaceous and Cenozoic sea-level estimates: backstripping analysis of borehole data, onshore New Jersey William A. Van Sickel,n1 Michelle A. Kominz,n Kenneth G. Millerw and James V. Browningw nDepartment of Geosciences,Western Michigan University, Kalamazoo, MI, USA wDepartment of Geological Sciences, Rutgers University,Piscataway, NJ, USA ABSTRACT Backstripping analysis of the Bass River and Ancora boreholes from the New Jersey coastal plain (Ocean Drilling Project Leg174AX) provides new Late Cretaceous sea-level estimates and corroborates previously published Cenozoic sea-level estimates. Compaction histories of all coastal plain boreholes were updated using porosity^depth relationships estimated from New Jersey coastal plain electric logs.The new porosity estimates are considerably lower than those previously calculated at the o¡shore Cost B-2 well.Amplitudes and durations of sea-levelvariations are comparable in sequences that are represented at multiple boreholes, suggesting that the resultant curves are an approximation of regional sea level. Both the amplitudes and durations of third-order (0.5^5 Myr) cycles tend to decrease from the Late Cretaceous to the late Miocene.Third-order sea-level amplitudes in excess of 60 m are not observed. Long-term (108^107 years) sea level was approximately constant at 30^80 m in the Late Cretaceous, rose to a maximum early Eocene value of approximately 100^140 m, and then fell through the Eocene and Oligocene. INTRODUCTION Estimates for the Late Cretaceous to the Miocene based on changing ocean volumes (e.g., Harrison, 1990) are One of the more complex problems involved in unravelling somewhat higher than estimates based on continental Earth history is determination of the record of global hypsometry (Bond, 1979) and on backstripping (Watts & sea-level change (eustasy). Any estimate derived from con- Steckler, 1979). However, recent analysis of the age distri- tinental or continental margin data is necessarily a combi- bution of ocean £oor suggests that there may have been nation of both eustatic and tectonic changes, and is also no change in spreading rates for the last180 Myr, and thus, impacted by local or regional sedimentary processes and no related sea-level change (Rowley, 2002). Although this their climatic controls (e.g., Sloss, 1963; Bond, 1979; Watts does not eliminate other contributions to ocean volume & Steckler, 1979; Posamentier et al., 1988). Proxies, such as change, estimates by Harrison (1990) suggest that the pri- oxygen isotopes, can be used to estimate the volume of mary control on long-term eustasy was mid-ocean ridge water sequestered as glacial ice (Miller et al., 1987), but this volumes. Backstripping of the Cenozoic section from signal is overprinted by variations in temperature and, to a New Jersey (Kominz etal.,1998) indicates a relatively small, lesser extent, salinity (Miller, 2002). Estimates of eustasy but signi¢cant, long-term Cenozoic sea-level fall. Because from variations in the changing volume of the oceans, in- this record did not include the entireCretaceous sea-level cluding ridge volumes, sediment volumes, large igneous rise, Kominz et al. (1998) suggested that the magnitude of provinces, continent^continent collision and continental long-term sea-level fall was underestimated. The two ad- extension (e.g., Harrison, 1990; Larson, 1991) are compli- ditional boreholes analysed here were drilled into theLate cated by the fact that subduction removes increasingly Cretaceous in part to determine whether backstripping large portions of the data as estimates range towards ear- would support the larger- or smaller-magnitude estimates lier times (e.g., Kominz,1984). of long-term sea-level change. Sea-level change occurs in a cyclic, but generally not Backstripping of the New Jersey coastal plain sediments periodic, manner on various time scales (e.g., Vail et al., also allows us to estimate the magnitude of the equally 1977). The magnitude of long-term (50^200-Myr controversial third-order (about a half to three millionyear duration) variations has recently become controversial. duration) sea-level changes. Sea-level estimates based on sequence stratigraphic data (e.g., Haq et al., 1987) are con- 1 Present address: Polaris Energy Inc., 180 W.Michigan Avenue, troversial both because of the proprietary nature of the Suite 508, Jackson, MI 49201, USA. Correspondence: Michelle A. Kominz, Deptartment of Geos- data on which the estimates were based and because of ciences, Western Michigan University, Kalamazoo, MI 49008, limitations of the method (e.g., Miall, 1986, 1992; Chris- USA. E-mail: [email protected] tie-Blick & Driscoll, 1995). Finally, sequence stratigraphic r 2004 Blackwell Publishing Ltd 451 W. A. Van Sic kel et al. data suggest that some very large magnitude third-order su⁄cient £exural rigidity for o¡shore sediments that ¢lled sea-level changes of 100 m or more occurred (Haq et al., the thermally subsiding margin to generate accommoda- 1987).The onlyknown mechanism for such large, rapid eu- tion landward of the actively stretched crust (Watts, 1981; static variations is continental glaciation (e.g., Pitman & Steckler et al., 1999). The resulting coastal plain consists Golovchenko,1991). As such, these events are particularly of Lower Cretaceous to Holocene strata that dip gently problematic when theyoccurduring greenhouse climates, seaward and thicken down-dip (Olsson et al., 1988). Low- such as the Late Cretaceous. Thus, this is an important stand systems tracts are generally absent on the coastal time period in which to determine the magnitude of sea- plain so that sequences represent stacked transgressive level change. Backstripping of Cenozoic New Jersey coast- and highstand systems tracts (Miller et al., 1998b). al plain strata (Kominz et al., 1998) previously indicated that the magnitude of sea-level change was roughly half that estimated by the Haq et al. (1987) method. Testing for BACKSTRIPPING METHOD a eustatic origin requires that estimates of the sea-level magnitude be obtained at multiple locations.The New Jer- Backstripping is a quantitative method of estimating tec- sey boreholes provide overlapping middle Eocene tonic subsidence, which is de¢ned as the vertical move- ( 45 Ma) through late Miocene ( 7 Ma) data. Two new ment of basement in the absence of both sediment boreholes drilled at Ancora and Bass River, New Jersey loading and sea-level change (Watts & Ryan, 1976). We provide overlapping data and extend from the LateCretac- apply a one-dimensional analysis and assume an Airy iso- eous through middle Miocene. If third-order sea-level static response to loads (e.g., Bond et al., 1989).The calcu- variations are consistentwithin this data set, local tecton- lation of basement subsidence inwater,R1(also called ¢rst ism can be eliminated as a mechanism for relative sea-level reduction or accommodation), is a modi¢cation of the change. general backstripping equation of Steckler & Watts (1978): r R1 ¼ TS þ DSL a r À r REGIONAL SETTING a w à ra À rSà Two new boreholes, Bass River and Ancora (Fig. 1), ¼ S þ WD ð1Þ r À r were drilled by the Ocean Drilling Program (ODP) a w as Leg 174AX (Miller et al., 1998a, 1999a). These sites where TS is tectonic subsidence, DSL is eustatic sea-level targeted Cretaceous sections beneath the New Jersey change, Sn is decompacted sediment thickness, WD is coastal plain. paleodepth and ra and rw are the density of the asthenso- The New Jersey coastal plain forms the western margin sphere(3.18 g cmÀ 3 at1300 1C) and seawater (1.03 g cm À 3), of the Baltimore CanyonTrough, a large sedimentary basin respectively.The ¢rst reduction (R1) removes the e¡ects of that underlies the continental shelf along the middle sediment loading and compaction, producing curves that Atlantic, east coast of the United States (Grow & Sheridan, are an approximation of the subsidence that would have 1988). This basin formed as a result of LateTriassic^Early occurred if the basin had subsided in water without Jurassic rifting between North America and Africa (Grow sediment deposition (Bond et al., 1989). R1 includes & Sheridan, 1988). By Cretaceous time, the crust attained both tectonic subsidence and eustatic sea-level change. 76 75 74 73 72 W Fig. 1. Location map of New Jersey 41 N coastal plain boreholes and wells used in this study,including Ocean Drilling Pre-Cretaceous Program Leg150X (Island Beach, outcrop N Atlantic City,and Cape May) and Leg Onshore ODP sites 174AX (Bass River and Ancora) sites.The Cretaceous from legs 150X and USGS wells (Butler Place and Island outcrop 174AX Beach #33-01031) were used in conjunction with the Island Beach Cenozoic USGS wells outcrop 40 borehole and the o¡shore Cost B-2 well to Butler Place Island Beach #33-01031 estimate compaction of sand and mud. The USGS Clayton and Island Beach BassRiver Island Beach Clayton Ancora Atlantic Ocean wells were used to estimate thicknesses and lithologies beneath the Bass River and Ancora boreholes.The Baltimore Atlantic City CanyonTrough (grey,o¡-shore shading) is Cost B-2 Baltimore a Mesozoic^Cenozoic sedimentary basin, Canyon 39 which continues to the northeast and Cape May Trough 0 50 km to the south beyond the boundaries of this map. 452 r 2004 Blackwell Publishing Ltd, Basin Research, 16,451^465 Late Cretaceous and Cenozoic sea-level estimates Estimation of Sn requires the assumption of compaction Porosity (%) rates. Ideally, compaction is based on porosity estimates 0 20 40 60 80 100 from the wells or boreholes in the region of the study, as 0 is the case here. Seaward of the coastal plain, thermal subsidence and se- diment loading dominated accommodation (Steckleretal., 1 1988). The £exural response of the adjacent, unstretched crust generated su⁄cient accommodation onshore for se- diment accumulation, thus forming the coastal plain.