Embed Size (px)
Citation preview
Reprinted from JOURNAL OF GEOLOGY, 1990, vol. 98, p. 135-155.
1
QUANTITATIVE FILLING MODEL FOR CONTINENTAL EXTENSIONAL BASINS WITHAPPLICATIONS TO EARLY MESOZOIC RIFTS OF EASTERN NORTH AMERICA
ROY W. SCHLISCHE AND PAUL E. OLSENDepartment of Geological Sciences, Lamont-Doherty Geological Observatory of Columbia
University, Palisades, New York 10964
ABSTRACT
In many half-graben, strata progressively onlap the hanging wall block of the basins, indicatingthat both the basins and their depositional surface areas were growing in size through time. Basedon these constraints, we have constructed a quantitative model for the stratigraphic evolution ofextensional basins with the simplifying assumptions of constant volume input of sediments andwater per unit time, as well as a uniform subsidence rate and a fixed outlet level. The modelpredicts (1) a transition from fluvial to lacustrine deposition, (2) systematically decreasingaccumulation rates in lacustrine strata, and (3) a rapid increase in lake depth after the onset oflacustrine deposition, followed by a systematic decrease. When parameterized for the earlyMesozoic basins of eastern North America, the model’s predictions match trends observed in LateTriassic-age rocks. Significant deviations from the model’s predictions occur in Early Jurassic-agestrata, in which markedly higher accumulation rates and greater lake depths point to an increasedextension rate that led to increased asymmetry in these half-graben. The model makes it possible toextract from the sedimentary record those events in the history of an extensional basin that are duesolely to the filling of a basin growing in size through time and those that are due to changes intectonics, climate, or sediment and water budgets.
INTRODUCTION
A fundamental problem in the study ofextensional basins is the quantitativedetermination of the first-order controls on basinstratigraphy. This problem is especially acute fornon-marine basins, which lack a sea-level datum.Traditionally, each major change in depositionalenvironment and sedimentation has beeninterpreted as a result of some "tectonic" event(such as increased boundary fault movement) ormajor climatic change, yet predictive,quantitative models for the stratigraphicdevelopment of extensional basins are notablylacking. This stands in contrast to the study ofpost-rift deposits, in which the fundamentalaspects of the stratigraphy are explained in termsof thermally-driven subsidence, sedimentloading, sediment compaction, and eustatic sea-level change (e.g., McKenzie 1978; Steckler and
Watts 1978; Bond and Kominz 1984; Steckler et al.1988).
Quantitative stratigraphic modeling of extensionalbasins has proven difficult because of the lack ofappreciation of the critical effects of basin geometryon the stratigraphic record. In this paper, we developa simple model of extensional basin filling thatappears to explain many of the elements of thestratigraphic record of early Mesozoic rift basins ofeastern North America. The model is based on thefilling of a basin growing in size through time. Evenunder conditions of uniform subsidence and constantvolume of sediment and water input per unit time,the model predicts transitions in major depositionalenvironments and trends in accumulation rates andlake depth. Furthermore, the model’s predictions canbe "subtracted" from the actual stratigraphic record,yielding the tectonic and climatic signal andallowing for the parameterization of sediment andwater budgets.
2
EXTENSIONAL BASIN GEOMETRY ANDSTRATIGRAPHIC ARCHITECTURE
In many extensional basins, younger strataprogressively onlap the "basement" rocks of thehanging wall block (fig. 1). Charles Lyell (1847)first noted this onlap geometry in the Blackheathregion of the hinge area of the TriassicRichmond basin of Virginia. Outcrop studies andcoal mine excavations indicate that thelowermost stratigraphic unit the Lower BarrenBeds is absent from this region and that thestratigraphically-higher Productive CoalMeasures rest directly on basement. Furthertoward the border fault of the half-graben,seismic reflection profiles and drill hole datarecord the presence and thickening of the LowerBarren Beds (B. Cornet, personalcommunication, 1989). At Tennycape, NovaScotia, strata of the Triassic Wolfville Formationclearly onlap the Carboniferous "basement"rocks in the hinge of the Fundy half-graben.Furthermore, the beds of the WolfvilleFormation dip at a shallower angle than theunconformity between the Triassic andCarboniferous rocks (Olsen et al. 1989).McLaughlin stated that the basal StocktonFormation onlapped the northeastern, southern,and southwestern margins of the Newark basin(McLaughlin 1945; McLaughlin and Willard1949; Willard et al. 1959; see fig. 1A). Seismicreflection profiles reveal this onlap geometry in(1) the Cenozoic basins of the Basin andRange the Dixie Valley basin, Nevada(Anderson et al. 1983, their fig. 4), the northernFallon basin of the Carson Desert, Nevada(Anderson et al. 1983, their fig. 5), the DiamondValley basin, Nevada (Anderson et al. 1983,their fig. 6), the Railroad Valley basin, Nevada(fig. 1C; Vreeland and Berrong 1979, their fig.8), and the Great Salt Lake basin, Utah (Smithand Bruhn 1984, their fig. 10) (2) LakeTanganyika of East Africa (Burgess et al. 1988,their fig. 35-12) (3) the North Viking graben ofthe North Sea (Badley et al. 1988, their figs. 6and 8) (4) the Hopedale and Saglek basins of theLabrador margin (Balkwill 1987, his figs. 9 and
11) and (5) the Mesozoic basins of the U.S. Atlanticpassive margin the Long Island, Nantucket, andAtlantis basins (fig. 1B; Hutchinson et al. 1986, theirfigs. 3, 7, 11, and 15).
Anderson et al. (1983) cited this onlap pattern asevidence that the extensional basins had grown insize through time. Leeder and Gawthorpe (1987)related the onlap pattern to the migration of thebasin’s hinge or fulcrum (reflecting the linebasinward of which the hanging wall hasexperienced subsidence) away from the border faultand have incorporated this concept into theirtectono-sedimentary facies models. We agree withthese authors’ interpretations and extend them onestep further: not only were the extensional basinsgrowing in size, but the depositional surface alsowas growing in area through time as a consequenceof the filling of the basin. This conclusion serves asthe foundation of the extensional basin filling model.
Continental extensional basins often display atripartite stratigraphic architecture (Lambiase andRodgers 1988) which, from bottom up, consists of:(1) a fluvial unit indicating through-going drainageand open-basin conditions; (2) a lacustrine unit withdeepest-water facies near its base, gradually shoalingupward, indicating predominantly closed-basinconditions; and (3) a fluvial unit reflecting onceagain through-going drainage. These major changesin depositional environments (on a basin-wide scale)have been explained in terms of changes in basinsubsidence (Lambiase and Rodgers 1988). Thefluvial unit is interpreted to have been depositedduring initial, slow basin subsidence, perhaps in anumber of linked, small sub-basins. The deep-waterlacustrine unit is interpreted to reflect a deepening ofthe basin resulting from increased subsidence andthe coalescence of sub-basins. The upward-shoalingis interpreted to reflect the gradual infilling of thebasin, with fluvial sedimentation returning after thebasin had filled to the lowest outlet of the basin.Taking a different approach, Olsen and Schlische(1988a, 1988b) have demonstrated how evolvingbasin geometry as a consequence of filling underconditions of uniform subsidence and uniform rateof input of sediments and water could produce muchthe same stratigraphic architecture and in addition
3
make specific predictions about the changes inaccumulation rates and times of transitions fromone environment to another.
EXTENSIONAL BASIN FILLING MODEL
The analytically simplest model consists of abasin, trapezoidal in cross section, bounded byplanar faults that dip at equal angles, andbounded along-strike by vertical surfaces (fig. 2).In this full-graben model, uniform extensionresults in uniform subsidence along the boundaryfaults (see fig. 3), and hence the depth (D) of thebasin increases linearly. To maintain simplicityin the model, the volumetric sedimentation rate(Vs, the volume of sediment added to the basinper unit time) and the available volume of water(Vw) are constant. The outlet of the basin also isassumed to be held fixed with respect to anexternal reference system.
If the basin subsides slowly enough or Vs islarge enough, the basin initially fills completelywith sediments, and excess sediment and waterleave the basin (fig. 3A.1). We characterize thismode of sedimentation as fluvial; the basin issedimentologically and hydrologically open.During this phase, accumulation rate (thicknessof sediment deposited per unit time) is constantand equal to the subsidence rate: with an excessof sediments available, the thickness of sedimentdeposited is governed solely by the space madeavailable through subsidence. As the grabencontinues to grow, the same volume of sedimentadded per unit time is spread out over a largerand larger depositional surface area. Ultimately,a point is reached when the sediments exactly fillthe basin (fig. 3A.2). Thereafter, the sedimentsno longer can completely fill the growing basin.The basin is sedimentologically closed (allsediments entering the basin cannot leave thebasin), and a lake can occupy that portion of thebasin between the sediment surface and thebasin’s outlet (fig. 3A.3). After the onset oflacustrine deposition, accumulation ratesdecrease systematically, driven by the increasing
size of the depositional surface and constant rate ofinput of sediments (fig. 3A.3-6).
Assuming an unlimited supply of water, lakedepth (W) would be equal to the difference betweenthe height of the basin’s outlet above the floor of thegraben (defined as the basement-sediment contact)and the cumulative sediment thickness (fig. 3A.3).However, the supply of water is not infinite, and apoint is reached when the graben has grown to sucha size that the finite volume of water just occupiesthe volume between the basin’s outlet and thedepositional surface (fig. 3A.4). Thereafter, the basinis hydrologically closed, and lake depth decreases asa function of the growing size of the basin and non-linear losses to evaporation from a lake whosesurface area is growing through time (fig. 3A.5-6).
If the subsidence rate is fast enough or Vs is smallenough, the basin never fills with sediments, andlacustrine deposition can occur from the outset (fig.3B.1). As the graben grows in size through time, therate of increase of cumulative sediment thicknessand the accumulation rate both decrease. Lake depthincreases quickly until the finite volume of waterjust fills the graben (fig. 3B.2) and thereafterdecreases (fig. 3B.3).
After the onset of lacustrine deposition,accumulation rate should decrease indefinitely.However, graben do not continue to subsideindefinitely. If the subsidence rate slows sufficientlyor stops and Vs remains constant, the accumulationrate and lake depth would continue to decrease untilthe depositional surface could reach the basin’soutlet, thereby heralding the return of fluvialsedimentation at a rate equal to the subsidence rate.
Quantitatively, under conditions of fluvialsedimentation, cumulative sediment thickness (Y)and the depth of the graben (D) are given by:
Y D S tR= = (1A)where SR is the subsidence rate and t is time.Accumulation rate (S ) is the derivative of thecumulative thickness curve:
S Yt SR= =d
d (2A)
After the onset of lacustrine sedimentation(sedimentologic closure), cumulative sedimentthickness and accumulation rate are given by:
4
Y Y
B YV
LB Y
f
fs
ft
= +
× + +
− +( )
( )
tan
cotcot
cot
α
α α α
2
24
22
12
1
(1B)
SY
t
V
LB Y
V
Lt
l
sf
sl= = +( ) +
−dd
242
12
cotcotα α
(2B)where Yf is the thickness of fluvial sediments, Bis the width of the base of the graben, L is thelength of the graben, α is the dip of the borderfaults, and tl is time after the onset of lacustrinedeposition. While the basin is hydrologicallyopen, water depth (W) is given by:
W = D - Y (3A)Equation (3A) also applies in the case of fluvialsedimentation and correctly predicts zero lakedepth since D = Y. The equation that describeslake depth after hydrologic closure is:
(3B)where Vw is the volume of water available.
The derivation of equations (1B) and (3B) areas follows, using the geometry depicted in fig. 2.After each increment of subsidence [equation(1A)], the capacity of the basin (the volume ofthe trapezoidal trough) is calculated. If thecapacity is less than the volume of sediment thatcan be supplied in that increment, the basin isfilled with sediments, and the thickness ofsediments deposited in that increment is equal tothe amount of subsidence for that increment. Ifthe capacity of the basin exceeds the volume ofsediment that can be supplied in that increment,the basin cannot completely fill. An integratedapproach can then be applied to derive theexpression for cumulative sediment thickness[equation (1B)]. From the geometry depicted infig. 2, the volume (V) of sediments is given byV V t x x L B Y x Ls l f= = ( )( )( ) + +( )( )( )cot cotα α2
(4)
where Vs is the volumetric sedimentation rate, tl isthe time after the onset of lacustrine sedimentation, αis the dip of the border faults, L is the length of thebasin, B is the initial fault spacing, Yf is the thicknessof fluvial sediments, and x is the next increment oflacustrine sediment thickness. Rearranging terms:
cot cotα α( ) + +( ) − =x B Y xV t
Lfs2 2 0 (5)
The positive root of x is obtained from the quadraticformula. Equation (1B) is then given by
Y = Yf + x (6)After the onset of lacustrine deposition, the capacityof the basin is again calculated knowing Y. If it isless than the available volume of water (Vw), waterdepth (W) is calculated as the difference between thetotal depth of the basin and Y. If the basin’s capacityexceeds Vw, lake depth is derived as follows, againbased on the geometry in fig. 2. V B Y W L W W Lw = +( )( )( ) + ( )( )( )2 cot cotα α (7)
L W BL Y W Vwcot cotα α( ) + +( ) − =2 2 0 (8)Solving for the positive root of W yields equation(3B).
In principle, the above analysis should hold forany concave-upward basin in which the depositionalsurface area grows through time under conditions ofuniform subsidence, constant rate of input ofsediments and water, and constant position of theoutlet fixed to an external reference system, althoughthe form of the equations will differ depending onthe evolving geometry of the basin. In a half-grabengrowing in volume through time (inferred fromonlap geometry) under conditions of uniformextension, it therefore seems reasonable to assumethat the changes in average accumulation rate andthe time of transition from fluvial to lacustrinesedimentation should follow a similar path as for afull-graben as long as the area of the depositionalsurface continues to grow. Although three-dimensional modeling of the filling of a half-grabenhas not yet been undertaken, preliminary two-dimensional modeling indicates that the aboveassumption may be valid. In the two-dimensionalapproach, a constant cross-sectional area (the two-dimensional analogue of the volumetricsedimentation rate) is iteratively fit to the observedgeometry of a half-graben (fig. 4). Throughout the
WL
BL YL LV
B Y
w= +( ) +[ ]− +
tancot cot
cotcot
α α α
αα
22 4
22
21
2
5
"filling" process, both the half-graben’s geometryand the stratal geometry are maintained. As seenin figures 4B and 4C, both the cumulativethickness and accumulation rate curves for both adrill hole through the deepest part of the basinand the average thickness of each stratal packageare quite similar in form to the trends producedin a full-graben (cf. figs. 4 and 9), implying thatfull- and half-graben fill similarly.
The evolution of half-graben, however, can bemore complicated than that proposed in themodel. The depositional surface area of a half-graben ceases to grow when the onlapping edgeof the basin fill passes inside the margin ofpreviously deposited sediments, i.e., when stratacease to onlap the "basement" rocks and begin toonlap previously-deposited strata. At that point,under conditions of uniform subsidence, theaccumulation rate would cease to change oractually decrease. Additional complications arisein half-graben bounded by domino-stylerotational faults. With continued extension, thenormal faults progressively rotate to shallowerdips (Wernicke and Burchfiel 1982). Therefore,even under conditions of uniform extension,subsidence will not be uniform. Furthermore,subsidence can be expected to decrease from amaximum along the boundary fault to aminimum at the hinged margin of the half-graben. We also expect significant along-strikechanges in subsidence, ranging from a maximumnear the center of the boundary fault to zero at itstips. This cumulative variation in subsidencewould then mimic the nature of subsidenceobserved following slip on normal faults, such asthe 1959 Hebgen Lake earthquake in Montana(Fraser et al. 1964).
The qualitative model presented abovepredicts major changes in depositionalenvironments as a consequence of the filling ofthe basin: fluvial sedimentation (basin issedimentologically and hydrologically open)giving way to hydrologically-open lacustrinesedimentation followed by hydrologically-closedlacustrine sedimentation. Furthermore, the modelcan be quantified to predict the timing of the
major transitions in depositional environments andtrends in accumulation rates and lake depth. Clearly,however, the model cannot address all aspects ofextensional basin stratigraphy. For example, during"lacustrine deposition," fluvial and fluvio-deltaicdeposits will ring the lacustrine deposits (see faciesmodels of Leeder and Gawthorpe 1987).Furthermore, near the main border faults of thebasin, lacustrine and alluvial fan deposits willinterfinger. Repeated motions along the border faultwill lead to the formation of tectonic cyclothems(Blair and Bilodeau 1988). In addition, the faciesdistribution will be influenced by climatic changes(Olsen 1984a, 1984b, 1986): the lakes may shrinkand swell, and the position of the marginal lacustrinefacies will migrate accordingly. The model thereforeis only intended as a guide to understanding thegross stratigraphic development of extensionalbasins, reflecting perhaps those processes operatingon a scale of 106 or 107 years.
The rationale of the assumptions used in the basinfilling model warrants additional discussion. Themodel assumes a constant volumetric sedimentationrate, yet as the basin fills, younger strataprogressively onlap potential source areas of newsediment. As a result, volumetric sedimentation ratemight decrease. However, in the East African rifts,most sediments are derived from the hanging wallblock and from rivers flowing along the axis of therift system (Lambiase, in press). If this axialcomponent is a general feature of most rift systems,then the effect of onlap onto source regions may notcritically affect sediment supply, and the assumptionof constant Vs is probably a good starting point.
The model also assumes that the outlet of thedepositional basin is fixed with respect to an externalreference system. Based again on the East Africanrifts, the lowest elevation outlet is commonly foundat the along-strike end of an individual half-graben(Lambiase, in press). As this is the region thatexperiences minimal hanging wall subsidence andfootwall uplift, then the outlet may indeed beassumed to be stationary.
As for subsidence and volume of water available,there are no reliable, long-term, modern data.Parsimony therefore dictates that we assume these
6
parameters to be constant. The model isspecifically designed to assess the intrinsiceffects of the filling of the basin. Only whenthese are understood can we begin to examineand decipher the causes of the extrinsic effects(changes in subsidence rate, climate, and rates ofsediment and water input). This approach isillustrated below with reference to theextensional basins containing the rocks of theearly Mesozoic-age Newark Supergroup,particularly the Newark basin.
APPLICATION OF THE BASIN FILLING MODEL
Geologic Overview of the Newark Super-group. Eastern North America forms part ofthe classic Atlantic-type passive continentalmargin (Bally 1981) created by rifting of thesupercontinent of Pangaea. All along the axis ofthe future Atlantic Ocean from Greenland toMexico, the Triassic initiation of the breakupwas marked by the formation of extended crust.Early Mesozoic rift basins of eastern NorthAmerican are exposed from South Carolina toNova Scotia; numerous others have beendiscovered or inferred by drilling, seismicreflection profiling, and gravity studies to beburied both by post-rift sediments on thecontinental shelf and below coastal plainsediments (fig. 5). Thousands of meters ofexclusively continental sediments, tholeiitic lavaflows, and diabase plutons filled the exposed riftbasins. The faulted, tilted, and eroded strata aretermed the Newark Supergroup (Van Houten1977; Olsen 1978; Froelich and Olsen 1984).
Milankovitch-Period Lacustrine Cycles. Thelacustrine rocks of Newark Supergroup basinsshow a common theme of repetitive cyclesnamed in honor of their discoverer, F.B. VanHouten, by Olsen (1986) . The fundamental VanHouten cycle consists of three lithologically-distinct divisions which are interpreted as (1)lake transgression, (2) high stand, and (3)regression plus low-stand facies (Olsen 1986).These cycles were apparently produced by the
basin-wide rise and fall of water level of very largelakes (Van Houten 1964; Olsen 1984a, 1984b, 1986;Olsen et al. 1989). Fourier analysis of stratigraphicsections has shown that the Van Houten cycles havea periodicity of approximately 21,000 years (Olsen1986). Furthermore, the 21,000-year-long cycles arehierarchically arranged in compound cycles of100,000 and 400,000 years. It is this hierarchy thatprovides the best evidence that lake levels wereclimatically driven by orbital variation according tothe Milankovitch theory (Olsen 1986). These cyclespermit well-constrained estimates of accumulationrate (thickness of cycle divided by its duration) inthe lacustrine intervals of these extensional basins,thereby providing a high-resolution test of thepredictions of the basin filling model.
Stratigraphic Pattern. Two markedly differentstratigraphic patterns are present within the basins ofthe Newark Supergroup and are exemplified by thestratigraphy of the Newark basin (Olsen 1980). Thefirst stratigraphic pattern is of mostly Late Triassicage and consists of (1) the basal red and brownfluvial strata of the Stockton Formation (depositedduring the first 3 M.yr. of basin subsidence),overlain by (2) the mostly gray and black lacustrinerocks of the Lockatong Formation, which is in turnoverlain by (3) the mostly red, lacustrine andmarginal lacustrine strata of the Passaic Formation(fig. 6). After the onset of lacustrine deposition,accumulation rates (as calibrated by the thickness of21,000-year-long Van Houten cycles) slowly andsystematically decreased through the remainder ofthe Triassic section (Lockatong and Passaicformations). Lake depths were controlled byorbitally-induced climatic changes, and the lakesdried out every 21,000 years (Olsen 1986).However, certain longer-term trends are alsoapparent. The first microlaminated black shalesformed about 800,000 years after the onset oflacustrine deposition and must have been depositedin at least 70-100 m of water, below wave base, atthe bottom of an anoxic, turbulently-stratified lake(Manspeizer and Olsen 1981; Olsen 1984a). Thebest-developed microlaminated units, suggesting thedeepest lakes, are found in strata deposited 1.4 M.yr.
7
after lacustrine deposition began. Higher in theLockatong Formation, red massive mudstonesbecome more and more abundant. With theexception of the Ukrainian Member of thePassaic Formation and the Early Jurassic-agesedimentary rocks, the last microlaminated unitis present near the base of the Passaic Formation.Gray and black shales are still present within therest of the Passaic Formation, but both theirfrequency and extent of development decreaseup-section (fig. 6). For ease in modeling, wehave defined "maximum" lake depth as the depthof the deepest lake produced by each 400,000year climate cycle; it is the bounding curve forthe entire fluctuating lake-level curve (see fig. 6)."Maximum" lake depth began shallow at theonset of lacustrine deposition, rapidly deepenedduring the deposition of the lower LockatongFormation, and then slowly and systematicallydecreased toward the Triassic-Jurassic boundary(fig. 6).
The second stratigraphic sequence of mostlyEarly Jurassic age consists of tholeiitic lavaflows interbedded with and succeeded by mostlylacustrine strata, which show marked changes inaccumulation rate and "maximum" lake depth.Approximately 30 m below the palynologically-determined Triassic-Jurassic boundary (Cornet1977), both the accumulation rate and"maximum" lake depth greatly increased (fig. 6).The highest accumulation rates and the deepestlake facies (well-developed microlaminatedunits) are associated with the thick extrusivebasalt flows. After the extrusion of the last basaltflow, accumulation rate and "maximum" lakedepth again decreased systematically.
Basin Geometry. Eastern North Americanrift basins were profoundly influenced by pre-existing structural control (Ratcliffe and Burton1985; Swanson 1986; Schlische and Olsen 1987,1988). Where the basins trended oblique to theregional extension direction, extensional strike-slip duplexes, flower structures, and smallsyndepositional horsts and basins weredeveloped in strike-slip zones (Schlische and
Olsen 1987; Olsen et al. 1989). In both volume ofsedimentary fill and surface area, however, thesestrike-slip regions are subordinate to thepredominantly dip-slip half-graben.
Where the early Mesozoic extension directionwas normal to the strike of pre-existing structures,Paleozoic thrust faults of the Appalachian orogenwere reactivated principally as dip-slip normalfaults, as in the Newark basin (Ratcliffe and Burton1985), forming half-graben in their collapsedhanging walls. Based on surface geology andseismic reflection profiles, the basin fill of thesehalf-graben is wedge-shaped (see fig. 1), indicatingthat normal faulting responsible for basin subsidencewas active during sedimentation. Furthermore,younger strata dip less steeply than older strata,indicating syndepositional rotation of the hangingwall block. As mentioned above, published seismicreflection profiles of offshore basins clearly showonlap onto the hanging wall, and outcrop and drillcore studies of the Newark, Fundy, and Richmondbasins lead to a similar conclusion.
The Newark basin of New York, New Jersey, andPennsylvania is a half-graben bounded on itsnorthwestern side by a southeast-dipping borderfault system (figs. 1, 7). The dip of the border faultranges from 60¡ to less than 25¡ (Ratcliffe andBurton 1985) with an average dip of 45¡. The basinis approximately 190 km long (excluding the narrowneck between the Newark and Gettysburg basins),presently 30-50 km wide, and contains cumulatively>7 km of sedimentary rocks, lava flows, and diabasesheets and dikes. Strata within the basin generallydip at less than 20¡ toward the border fault system,although they are locally more steeply dippingwithin transverse folds adjacent to the border andintrabasinal faults (see fig. 7). McLaughlin’s studiesof the Stockton Formation (McLaughlin 1945;McLaughlin and Willard 1949; Willard et al. 1959)and our mapping of the thickness of Van Houtencycles (Olsen et al. 1989; Olsen in press) clearlyshow that units of all ages thicken noticeably towardthe border fault side of the basin, indicating that thefaulting responsible for basin subsidence wasoccurring throughout the filling of the basin (contraFaill 1973). The basin is cut by a number of
8
intrabasinal faults, but these appear to havedeveloped late in the history of the basin, afterdeposition of most of the preserved sedimentaryrocks (Olsen 1985; Faill 1988; Schlische andOlsen 1988).
Comparison of Model Predictions withNewark Basin Stratigraphy. Before the full-graben model can be applied, the asymmetricNewark half-graben needs to be "transformed" toa symmetric basin. The approach is to determinea representative cross-sectional area of theNewark basin for a given time-slice and then byconserving cross-sectional area convert the basininto a full-graben with border faults dipping atequal angles. Thus the width of the basin afterdeposition of a certain stratigraphic horizon andthe depth of the basin for that time need to beknown. Unfortunately, the Newark basin hasbeen substantially eroded, and its presentboundaries do not reflect its limits duringdeposition. Variations in thickness of units withdistance from the border fault, however, doprovide important constraints. In the DelawareRiver valley, the Newark basin is cut up intothree fault blocks with repetition of the sameformations (fig. 8A). Outcrops in thenorthwestern fault block indicate that theStockton Formation is approximately 1800 mthick (Olsen 1980), whereas the Stockton is onlyabout 900 m thick where it outcrops in thesoutheastern fault block (Willard et al. 1959).Because the intrabasinal faulting completelypost-dates the preserved strata in these faultblocks (Olsen 1985; Faill 1988; Schlische andOlsen 1988), key horizons can bepalinspastically restored as shown in fig. 8A. Inthis restored basin, the base and top of theStockton Formation are constrained by knownthicknesses and by the average dip of the base ofthe Stockton in the northwestern fault block;where these two horizons converge is taken to bethe southeastern edge of the Newark basin afterdeposition of the Stockton Formation. A similarexercise can be performed on the PerkasieMember, where the constraints are (1) the
position of Perkasie outcrops in the northwesternfault block and (2) the height of the Perkasie abovethe top of the Stockton Formation in the middle faultblock. A projection of the line connecting these twopoints clearly extends further to the southeast thanthe limits of the Stockton Formation, again reflectingthe progressive onlap of younger strata ontobasement on the hanging wall margin of the basin.
In fig. 8B, the horizon representing the PerkasieMember has been rotated to paleohorizontal. Thebasin was approximately 6 km deep in its depocenter(corroborated by proprietary seismic lines) and >60km wide after deposition of the Perkasie Member.We have assumed that the border fault did not rotatealong with the hanging wall and hence the 30¡ dipshown in fig. 8B is the same as the present-day dip(Ratcliffe et al. 1986). The "full-graben" equivalentof this basin is shown in fig. 8C. By preservingcross-sectional area, choosing an initial border faultspacing of 5 km (so that the "full-graben" moreclosely approximates the triangular cross-sectionalgeometry of the palinspastically-restored Newarkhalf-graben), and by preserving the same 6 kmthickness of strata, the "border faults" of the "full-graben" must dip at approximately 13¡. Because ofthis transformation, the output of the model (alwaysgiven for the deepest part of the full-graben) willmost closely match the stratigraphic data from thedeepest portion of the Newark half-graben.
Additional parameters entered into the basinfilling model were determined as follows. Thesubsidence rate can be constrained because, underfluvial sedimentation during deposition of theStockton Formation, subsidence rate equaledaccumulation rate. The maximum thickness of theStockton Formation in the deepest part of the basinis estimated to be 2500 m (see fig. 8A), and itspaleontologically estimated duration is 3 M.yr.(Cornet and Olsen 1985), thus yielding an estimatedaccumulation rate of 0.833 mm/yr. A volumetricsedimentation rate (Vs = 4.15x106 m3/yr) was chosento yield a transition from fluvial to lacustrinedeposition 3 M.yr. after the start of subsidence. Themaximum volume of water available (Vw = 1.10x1012 m2) was chosen to yield the deepest lakes inthe lower Lockatong Formation.
9
It must be emphasized again that the Newarkbasin did not evolve as a full-graben. Thisexercise is only a heuristic procedure meant todemonstrate an application of the basin fillingmodel, not to precisely model the evolution ofthe Newark basin. It is clear that the approachhas severe limitations: namely that it cannotpredict transverse and along-strike changes inaccumulation rate and lake depth in anasymmetrically-subsiding basin. Nonetheless,the trends in accumulation rate and "maximum"lake depth generated by the model in manyinstances match the same trends observed in therock record.
The curves generated using equations (1-3)are shown in figure 9 with reference to actualstratigraphic data. Using the parameters outlinedabove, the model can generate a transition fromfluvial to lacustrine sedimentation at theappropriate time (fig. 9: b, b’, and b’’). It remainsto be seen whether the Vs needed to generate thistransition is realistic. A potential test would be tocalculate the volume of a key marker bed, suchas the Perkasie Member, over the entire basin;unfortunately, the trends in thickness of unitsalong-strike are not yet well enough defined. Themodel also predicts a systematic decrease inaccumulation rate in the Lockatong and Passaicformations (fig. 9: f). Actual stratigraphic datashow a similar trend, suggesting that the fillingof a basin growing in size through time mightindeed be responsible. Notice, however, that themodel’s accumulation rates are consistentlyhigher than those measured in the basin. If weassume that the sediment added to the modelbasin was fully compacted, then the discrepancyis most likely the result of the fact that themodel’s output applies to the deepest part of thebasin and that the actual stratigraphic data werenot collected from its depocenter. Additionalcomparisons and extrapolations are notwarranted at this time because the stratigraphicdata come from outcrops scattered throughout abasin in which appreciable transverse and along-strike changes occur.
The decreasing accumulation rates in theLockatong and Passaic formations should not benecessarily construed to reflect decreasingsubsidence rates. It can be argued that the lakecycles present in these formations represent fillingsequences: the basin subsided, formed a lake, whichsubsequently filled in, the basin subsided, etc. In thisscenario, the decompacted cycle thickness wouldapproximate the deepest lake and the amount ofsubsidence. This argument is flawed on twoaccounts: (1) The cycle thickness in the Lockatong isonly 5.5 m, decompacted to about 10 m, implying alake depth of 10 m, yet the facies analysis of Olsen(1984a) requires the lakes that depositedmicrolaminated sediments were minimally 70-100 mdeep. (2) Most of the Passaic Formation is notmicrolaminated, and therefore there are no goodconstraints on lake depth; it is then plausible thatsubsidence approximates decompacted cyclethickness and was decreasing through the depositionof the Passaic Formation. However, the UkrainianMember in the middle Passaic is microlaminated andwould require an instantaneous 70-100 m ofsubsidence, which had no effect on accumulationrate (see below). If subsidence was relativelyuniform as predicted by the basin filling model, thebasin would have been sediment-starved duringPassaic deposition, and the basin need not haveundergone a radical deepening to accommodate theUkrainian lake. All we can say for certain is that thesubsidence rate was greater than the accumulationrate.
The basin filling model also predicts the veryrapid rise in "maximum" lake depth inferred fromthe lower Lockatong Formation and its subsequentdecline (fig. 9: m). It is also plausible that longer-term changes in climate were responsible for thetrends in "maximum" lake depth. Clearly, climaticchanges played a large role in the basin’smicrostratigraphy (Olsen 1986). However, there isno evidence that the trends in accumulation ratewere governed by changes in climate. For example,in the upper Passaic Formation, the range of facies isidentical to those in the driest portions of theLockatong Formation (Olsen 1984a), yet theaccumulation rates in the Passaic Formation are
10
lower. Furthermore, in the rare instances wheredeep lakes were produced in the PassaicFormation (i.e., the Ukrainian Member), there isno appreciable change in accumulation rate fromthe surrounding dry facies (see fig. 9). Sincetrends in accumulation rate appear to be detachedfrom a climatic origin, it seems reasonable (andparsimonious) to assume that changes in"maximum" lake depth might be similarlydetached, although to a lesser extent. Note alsothat the model predicts that the lastmicrolaminated black shale should formconsiderably later than is indicated by the data(fig. 9: k). This most likely can be attributed tothe fact that the model did not incorporate theeffects of evaporation of water from a lakewhose surface area was continually increasing.Thus, actual lake depth decreased more rapidlythan predicted by the model.
Operating under its basic assumptions(uniform subsidence, constant rates of input ofsediment and water), the model cannot explainthe observed Early Jurassic increases in bothaccumulation rate and "maximum" lake depth(fig. 9: g, o), which imply that the Newark basinhad actually decreased in size. In full-graben,changes in subsidence have either no effect orthe wrong effect, for accumulation rate and"maximum" lake depth are governed by the areasof the depositional and lake surfacesrespectively. Increasing the subsidence rate onlydeepens the basin, but it cannot change thesurface areas. Decreasing the subsidence rateinitially has no effect until the basin fills to itslowest outlet, whereupon fluvial sedimentationreturns.
In half-graben, however, maximumsubsidence occurs adjacent to the border faultand decreases systematically toward the basin’shinge. An increase in the subsidence rate resultsin an increase in basin asymmetry. As a result,sediments and water shift to the deepest portionof the basin. A rapid increase in basin asymmetryin the earliest Jurassic of the Newark basintherefore could have resulted in (temporary)decreases in the areas of the depositional and
lake surfaces, causing increases in accumulation rateand "maximum" lake depth. Systematic decreases inaccumulation rate and "maximum" lake depthrecorded in the Boonton Formation (figs. 6 and 9 h)might then reflect the return to uniform subsidence,as the newly deposited sediments progressivelyonlapped the older sediments and "basement" rocksof the hanging wall.
Asymmetry may have been introduced by twofundamentally different processes (fig. 10): (1)accelerated rotation of the entire hanging wall blockand (2) the formation of antithetic faults between thebasin’s hinge and the border fault, thus generating anarrower sub-basin. Presently available geologicdata are not good enough to distinguish betweenthese two processes.
A dramatic increase in the volume of sedimentadded to the basin per unit time also would haveincreased the accumulation rate in the Jurassicsection. However, this added volume of sedimentwould have raised the depositional surface andthereby displaced the volume of water upward,perhaps even causing lake level to reach thehydrologic outlet. In a concave-upward basin, thiswould have had the effect of increasing the lake’ssurface area, thereby decreasing "maximum" lakedepth, not increasing it dramatically. An increase inthe volume of water (Vw) added to the basin duringthe wettest portions of the 400,000 year climatecycles would have increased "maximum" lake depthand presumably also would have brought in moresediments, increasing the accumulation rate.However, this would imply that the accumulationrate ought to be lower in the drier portions of the400,000 year cycles, which is not the case. Ittherefore appears likely that the observed EarlyJurassic increases in accumulation rate and"maximum" lake depth the resulted from an increasein basin asymmetry.
In terms of the approach taken in this paper, onlyone parameter subsidence rate needs to be alteredto explain the marked changes in the Jurassicsequence. In terms of the traditional approach, thechanges would be explained by increasing both thevolume of sediment and water added to the basin.For example, Manspeizer (1988) cites the deep
11
Jurassic lakes as evidence of a wetter Jurassicclimate. Our new approach indicates that climateneed not necessarily have been responsible.
Other Basins. The basin filling model pre-dicts a basic tripartite stratigraphic patternconsisting of (1) a fluvial base, which would notbe present if the subsidence rate is high and/orthe rate of sediment input is low), (2) a middlelacustrine interval showing decreasingaccumulation rates and evidence of deep lakesnear its base followed by gradual shoaling, and(3) possibly an upper fluvial interval, depositedduring waning subsidence. These predictionsprovide a convenient frame of reference againstwhich the observed stratigraphy of basins can becompared and their histories deciphered.
All of the Newark Supergroup basins shownin fig. 11 contain a basal fluvial unit (Reinemund1955; Hubert et al. 1978; Hubert and Forlenza1988; Ressetar and Taylor 1988; Olsen et al.1989; Smoot 1989). In the Fundy, Culpeper,Richmond, and Deep River basins, the fluvialunits are overlain by predominantly lacustrinedeposits (Reinemund 1955; Ressetar and Taylor1988; Smoot 1989) with the deepest waterintervals definitely occurring near the base of thelacustrine sequence in the Deep River,Richmond, and Fundy basins (Olsen et al. 1989).In the Blomidon Formation of the Fundy basin,studies of scattered outcrops support a decreasein accumulation rate up-section (Olsen et al.1989). In the Hartford basin, the basal NewHaven Arkose is fluvial until just prior to theeruption of the first basalt flow (Olsen et al.1989). It is possible that the basin was subsidingtoo slowly or the volumetric sedimentation ratewas too high to have permitted the onset oflacustrine deposition. It is also possible that thepreserved fluvial strata are marginal to lacustrinestrata. Since paleocurrents flowed from east towest at the time of New Haven deposition(Hubert et al. 1978) and the strata now dip to theeast, any lacustrine strata undoubtedly nowwould be eroded. In the upper Triassic portionsof the Culpeper and Deep River basins, a higher
content of coarse clastic material suggests thatfluvial sedimentation may have returned (Olsen et al.1989), signaling a decrease in the subsidence rate. Italso is possible that the coarse clastics reflect theproximity to the border fault or that they are themarginal facies to lacustrine strata deposited in adeeper portion of the basin. More study is needed toresolve this problem. The Otterdale Sandstone of theRichmond basin is largely fluvial (Ressetar andTaylor 1988) and contains evidence of Late Triassicpaleo-canyon systems apparently cut by rivers aftersubsidence had stopped or during uplift of the basin(B. Cornet, personal communication, 1989).
In the Fundy, Hartford, and Culpeper basins, thesyn- and post-extrusive formations record higheraccumulation rates and greater "maximum" lakedepths over the preceding Triassic portions (fig. 11;Olsen et al. 1989), possibly reflecting increasedasymmetry in each of these basins in the earliestJurassic. The uppermost Portland Formation of theHartford basin appears to be fluvial in origin (Olsenet al. 1989). If no lacustrine strata are locateddowndip of the exposed fluvial strata, then thisinterval recorded a slowing of basin subsidence,which allowed the Hartford basin to once again fillto its lowest outlet.
Lake Bogoria in the Gregory Rift and thehydrologically-open Lake Tanganyika represent twomodern examples (Lambiase and Rodgers 1988;Lambiase in press) of graben in the first two stages,respectively, of the model’s tripartite evolution. TheKeweenawan trough of Michigan and Wisconsin(Elmore and Engel 1988) also shows a tripartitestratigraphic architecture.
Jones (1988) presented a graph of cumulativesediment and lava flow thickness plotted againsttime for the Kenya Rift east of the Elgeyoescarpment. Although there is considerable scatter,the rate of increase of cumulative thicknessdecreases up-section, which may indicate that thebasin was increasing in size through time. Apaleomagnetically- and radiometrically-calibratedplot of cumulative sediment thickness against timefor the Ngorora Formation of the Kenya Rift showsthe same trend (Tauxe et al. 1985; their fig. 4).; Thehigher accumulation rates in the lower part of the
12
section were interpreted to reflect the rapidinfilling of a topographic depression and thesubsequent lower accumulation rates to reflectsedimentation keeping up with subsidence,implying that excess sediments were leaving thebasin. However, the stratigraphy of the uppertwo-thirds of the Ngorora Formation ispredominantly lacustrine (Bishop and Pickford1972), implying sedimentologic closure. Thus,the decrease in accumulation rate up-section maybe an indication of a basin growing in sizethrough time, fitting our model’s predictions.
Cumulative sediment thickness andaccumulation rate curves for the CenozoicVallecito-Fish Creek basin in the Imperial Valleyin California (Johnson et al. 1983; their figs. 4and 5) bear a strong resemblance to those in fig.9. As the sediments were deposited largely inshallow marine waters, implying thatsedimentation kept up with subsidence, Johnsonet al. (1983) interpreted the curves to reflectdecreasing rates of tectonic subsidence.Alternatively, the basin may have been subsidinguniformly and growing in size through time.
SUMMARY AND DISCUSSION
Stratal onlap geometry in extensional basinsindicates that the basins were growing in sizethrough time, and therefore the depositionalsurface area was also increasing through time.Under conditions of uniform subsidence andconstant rate of sediment and water input, asimple basin filling model predicts an initialperiod of fluvial sedimentation as the volume ofsediments available exceeds the capacity of thebasin, such that the accumulation rate equals thesubsidence rate. As the given volume ofsediment is spread over a larger and larger basinvolume, a point is reached when the sedimentsexactly fill the basin; subsequently, lacustrinedeposition occurs. Accumulation rate decreasesas the sediments are spread over a largerdepositional surface area. After the onset oflacustrine deposition, lake depth should rapidlydeepen to a maximum (as the finite volume of
water available just fills the basin between thehydrologic outlet and the depositional surface) andthen systematically decrease as the volume of wateris spread over a larger area. If the subsidence rate isfast enough or the sediment input rate slow enough,lacustrine deposition will occur from the outset.
The basin filling model provides a framework forunderstanding the stratigraphic development ofextensional basins, particularly the Newark basin ofeastern North America. Many of the observedchanges through the stratigraphic sequence can beexplained simply as the consequence of the filling ofa growing basin: the transition from fluvial tolacustrine deposition in the Triassic-age sequence,the decrease in accumulation rate (calibrated by thethickness of Milankovitch-period lacustrine cycles)after the onset of lacustrine deposition, and theinferred rapid rise and subsequent slower decline in"maximum" lake depth. Other observations requireadditional explanations. For example, figure 9: pshows a marked departure in "maximum" lake depthin the Ukrainian Member of the Passaic Formation.Prior to this type of analysis, the "excursion" wouldsimply have been interpreted as one of the manyclimatic changes thought to have influenced thebasin’s stratigraphic development. With the basinfilling model and Milankovitch-period modulationaccounting for most of the changes in thestratigraphic sequence, this anomaly stands out assomething unique a superwet climatic event.Accumulation rate was not affected, indicating thatsediment accumulation rates and climate are notcausally linked (fig. 9).
Analogous to thermal modeling of passivemargins, the usefulness of the basin filling modelrests in subtracting the stratigraphic effects duesimply to the filling of the basin (intrinsic effects)from the observed sedimentary record. The resultinganomalies reflect departures from the basin fillingmodel’s assumptions (e.g., constant volume ofsediment input per unit time, constant inflow rate ofwater, etc.) and point to important changes intectonics, climate, and sediment and water budgets(extrinsic effects).
Furthermore, two or more basins may becompared in order to separate local from regional
13
effects. For example, in each of the basinsillustrated in figure 11, the transition from fluvialto lacustrine deposition in the Triassic-agesequences occurred at different times (localeffect). This is to be expected: even if all thebasins had equal subsidence rates and begansubsiding simultaneously, their filling rates andthe timing of the fluvial-lacustrine transitionwould depend on the dimensions of the basins,their geometries, and the rates of sediment input.
On the other hand, the marked increases inaccumulation rate and "maximum" lake depthoccurred simultaneously in the earliest Jurassicin all of the basins containing this sequence(regional effect; fig. 11). This anomaly isattributed to a marked increase in basinasymmetry, causing sediment and water to pileup adjacent to the border fault (fig. 11). Thisasymmetry may have resulted from a suddenchange in border fault geometry. However, thisseems unlikely, as it would mean that all borderfaults experienced a similar change in geometryat the same time. More likely, the basins’asymmetry is attributable to a marked increase inthe extension rate, which was probably governedby deep-seated mantle processes.
The inferred regional increase in the extensionrate occurred just 150,000 years before thevolumetrically massive early Mesozoic extrusiveevent. The sudden increased extension may havethinned the crust sufficiently to allow melting inthe upper mantle. That the accelerated extensionwas a brief excursion is supported by the briefduration (<550,000 years) of the eastern NorthAmerican extrusive episode (Olsen and Fedosh1988). Furthermore, the loading of up to 700 mof basalt and coeval intrusives within the basinsmay have contributed to their asymmetry.
Both the filling of a growing extensional basinas a result of mechanical (fault-controlled)subsidence and subsidence induced by decay ofthermal anomalies on passive margins producemuch the same result on accumulation rates: adecrease in rate up-section. We do not imply thatour basin filling model is applicable to depositsresulting from thermal subsidence on known
passive margins. However, care must be taken ininterpreting accumulation rates in older (particularlyfossil-poor) rocks for which precise tectonic anddepositional settings (marine vs. continental) maynot be known and the cause of the subsidence(mechanical vs. thermal) may be confused. Theresolution of this problem may be aided somewhatby the observation that thermal subsidence tends tobe regionally persistent, whereas mechanicalsubsidence generally is restricted to smaller, discretebasins, where diagnostic coarse-grained facies maypoint to normal faulting as the cause of subsidence.
In basins in which fault-controlled subsidence isthought to be the dominant mechanism of basinevolution, as in the basins of the NewarkSupergroup, the minimum subsidence rate can beconstrained, for the subsidence rate is greater thanthe filling rate during lacustrine deposition and equalto the filling rate during fluvial deposition.Ultimately, by combining the results of the basinfilling model’s subsidence analysis with knowledgeon how extensional basins grow and deform throughtime (e.g., Wernicke and Burchfiel 1982; Gibbs1984), it may be possible to estimate how theextension rate varied throughout the history of EarlyMesozoic extension, provided the basins’ geometriesare also defined through time.
Before the extensional basin filling model can beused to its full potential, several importantimprovements need to be made. Foremost, three-dimensional quantitative models for the filling ofhalf-graben need to be developed, models that takeinto account not only variations in subsidence withposition in the basin but also build in the along-strike growth of the basin. The effects of differentialrelief and the erosion of uplifted blocks on the fillingof the basins has yet to be addressed. The non-linearlosses of water volume to evaporation still need tobe quantified as well as the effects of sedimentcompaction. Future models also need to incorporatedifferent filling sequences for structurally-distinctyet sedimentologically-linked half-graben, such asthe Newark-Gettysburg-Culpeper system in easternNorth America (Olsen et al. 1989) and the Ruzizi-Kigoma-Southern depositional basins of LakeTanganyika (Burgess et al. 1988; Lambiase in press).
14
In addition, the existing geometry of theextensional basins needs to be more tightlyconstrained, ideally through a network ofintersecting transverse and along-strike seismiclines.
Despite the shortcomings of the extensionalbasin filling model presented in this report, thepreliminary results are encouraging: even underthe simplest conditions, continental facies inextensional basins are expected to exhibitpredictable changes through time as aconsequence of the changing geometry of thebasin and not necessarily as a result of changingtectonic or climatic parameters. Indeed, in orderto recognize changes in the extrinsic parameters,the changes intrinsic to the basin must first besubtracted from the stratigraphic record.
ACKNOWLEDGMENTS. We thank Gerard Bond,William Bosworth, Roger Buck, Al Froelich, JosephLambiase, Fred Mandelbaum, and Joseph Smoot forvaluable discussions. This paper was greatlyimproved by reviews from Mark Anders, NicholasChristie-Blick, Bernie Coakley, Bruce Cornet,Michelle Kominz, Teresa Jordan, Marjorie Levy,and Warren Manspeizer. We gratefully acknowledgesupport from the Donors of the American ChemicalSociety administered by the Petroleum ResearchFund and the National Science Foundation (grantsBSR-87-17707 to PEO and EAR-87-21103 to G.Karner, M. Steckler, and PEO). This paper isLamont-Doherty Geological ObservatoryContribution Number 4575.
REFERENCES CITED
ANDERSON, R. E.; ZOBACK, M. L.; and THOMPSON, G. A.,
1983, Implications of selected subsurface data on thestructural form and evolution of some basins in the
northern Basin and Range province, Nevada and Utah:
Geol. Soc. America Bull., v. 94, p. 1055-1072.
BADLEY, M. E.; PRICE, J. D.; DAHL, C. R.; and AGDESTEIN,
T., 1988, The structural evolution of the northern Vikinggraben and its bearing upon extensional modes of basin
formation: Jour. Geol. Soc. London, v. 145, p. 455-472.
BALKWILL, H. R., 1987, Labrador basin: Structural and
stratigraphic style, in BEAUMONT, C., and TANKARD, A.
J., eds., Sedimentary Basins and Basin-Forming
Mechanisms: Canadian Soc. Petr. Geol. Memoir 12, p.17-43.
BALLY, A. W., 1981, Atlantic-type margins, in BALLY,
A.W., ed., Geology of Passive Continental Margins:
History, Structure, and Sedimentologic Record (with
Emphasis on the Atlantic Margin): AAPG EducationCourse Note Series 19, p. 1-1 to 1-48.
BISHOP, W. W., and PICKFORD, M. H. L., 1975, Geology,
fauna and paleoenvironments of the Ngorora Formation,
Kenya Rift Valley: Nature, v. 254, p. 185-192.
BLAIR, T. C., and BILODEAU, W. L., 1988, Development of
tectonic cyclothems in rift, pull-apart, and forelandbasins: Sedimentary response to episodic tectonism:
Geology, v. 16, p. 517-520.
BOND, G. C., and KOMINZ, M. A., 1984, Construction of tectonic
subsidence curves for the early Paleozoic miogeocline,southern Canadian Rocky Mountains: Implications for
subsidence mechanisms, age of breakup, and crustal thinning:
Geol. Soc. America Bull., v. 95, p. 155-173.
BURGESS, C. F.; ROSENDAHL, B. R.; SANDER, S.; BURGESS, C. A.;
LAMBIASE, J.; DERKSEN, S.; and MEADER, N., 1988, Thestructural and stratigraphic evolution of Lake Tanganyika: A
case study of continental rifting, in MANSPEIZER, W., ed.,
Triassic-Jurassic Rifting, Continental Breakup, and the Origin
of the Atlantic Ocean and Passive Margins: New York,
Elsevier, p. 859-881.
CORNET, B., 1977, The palynostratigraphy and age of the NewarkSupergroup: Unpub. Ph. D. dissertation, Pennsylvania State
University, State College, PA, 505 p.
———, and OLSEN, P. E., 1985, A summary of the
biostratigraphy of the Newark Supergroup of eastern North
America, with comments on early Mesozoic provinciality, inWE B E R , R., ed., III Congresso Latinoamericano de
Paleontologia. Mexico., Simposio Sobre Floras del Triasico
Tardio, su Fitografia y Paleoecologia., Memoria., p. 67-81.
ELMORE, R. D., and ENGEL, M. H., 1988, The Precambrian
Nonesuch Formation of the North American Mid-Continent
rift, sedimentology and organic geochemical aspects oflacustrine deposition, in Lacustrine Exploration: Case Studies
and Modern Analogues: AAPG Research Conference,
Snowbird, Utah, September 7-9, 1988.
15
FAILL, R. T., 1973, Tectonic development of the Triassic
Newark-Gettysburg basin in Pennsylvania: Geol. Soc.America Bull., v. 84, p. 725-740.
———, 1988, Mesozoic tectonics of the Newark basin, as
viewed from the Delaware River, in HUSCH, J.M., and
HOZIK, M.J., eds., Geology of the Central Newark Basin:
Fifth Meeting of the Geological Association of New
Jersey, Rider College, Lawrenceville, p. 19-41.FEDOSH, M. S., and SMOOT, J. P., 1988, A cored stratigraphic
section through the northern Newark basin, in FROELICH,
A. J., and ROBINSON, G. R., Jr., eds., Studies of the Early
Mesozoic Basins of the Eastern United States: U.S. Geol.
Survey Bull. 1776, p. 19-24.FRASER, G. D.; WITKIND, I. J.; and NELSON, W. H., 1964, A
geological interpretation of the epicentral area—the dual-
basin concept: U.S. Geol. Survey Prof. Paper 435, p. 99-
106.
FROELICH, A. J., and OLSEN, P. E., 1984, Newark
Supergroup, a revision of the Newark Group in easternNorth America: U.S. Geol. Survey Bull. 1537A, p. A55-
A58.
GIBBS, A. D., 1984, Structural evolution of extensional basin
margins: Jour. Geol. Soc. London, v. 141, p. 609-620.
HUBERT, J. F., and FORLENZA, M. F., 1988, Sedimentologyof braided river deposits in Upper Triassic Wolfville
redbeds, southern shore of Cobequid Bay, Nova Scotia,
Canada, in MANSPEIZER, W., ed., Triassic-Jurassic
Rifting, Continental Breakup, and the Origin of the
Atlantic Ocean and Passive Margins: New York,
Elsevier, p. 231-247.———; REED, A. A.; DOWDALL, W. L.; and GILCHRIST, J.
M., 1978, Guide to the red beds of Central Connecticut:
1978 Field Trip, Eastern Section, Society of Economic
Paleontologists and Mineralogists, University of
Massachusetts, Amherst, Department of Geology andGeography, Contribution No. 32, 129 p.
HUTCHINSON, D. R.; KLITGORD, K. D.; and DETRICK, R. S.,
1986, Rift basins of the Long Island Platform: Geol. Soc.
America Bull., v. 97, p. 688-702.
JOHNSON, N. M.; OFFICER, C. B.; OPDYKE, N. D.; WOODARD,
G. D.; ZEITLER, P. K.; and LINDSAY, E. H., 1983, Ratesof Cenozoic tectonism in the Vallecito-Fish Creek basin,
western Imperial Valley, California: Geology, v. 11, p.
664-667.
JONES, W. B., 1988, Listric growth faults in the Kenya Rift
Valley: Jour. Struc. Geol., v. 10, p. 661-672.LAMBIASE, J., in press, A model for tectonic control of
lacustrine stratigraphic sequences in continental rift
basins, in KATZ, B., and ROSENDAHL, B., eds., Lacustrine
Exploration: Case Studies and Modern Analogues:
AAPG Memoir.
———, and RODGERS, M. R., 1988, A model for tectonic control
of lacustrine stratigraphic sequences in continental rift basins,in Lacustrine Exploration: Case Studies and Modern
Analogues: AAPG Research Conference, Snowbird, Utah,
September 7-9, 1988.
LE E D E R, M. R., and GAWTHORPE, R. L., 1987, Sedimentary
models for extensional tilt-block/half-graben basins, in
Coward, M. P.; Dewey, J. F.; and Hancock, P. L., eds.,Continental Extensional Tectonics: Geol. Soc. London Spec.
Publ. 28, p. 139-152.
LYELL, C., 1847, On the structure and probable age of the coal
field of the James River, near Richmond, Virginia: Quarterly
Jour. Geol. Soc. London, v. 3, p. 261-280.MANSPEIZER, W., 1988, Triassic-Jurassic rifting and opening of
the Atlantic: An overview, in MANSPEIZER, W., ed., Triassic-
Jurassic Rifting, Continental Breakup, and the Origin of the
Atlantic Ocean and Passive Margins: New York, Elsevier, p.
41-79.
———, and OLSEN, P. E., 1981, Rift basins of the passivemargin: Tectonics, organic-rich lacustrine sediments, basin
analyses, in HOBBS, G. W., III, ed., Field Guide to the
Geology of the Paleozoic, Mesozoic, and Tertiary Rocks of
New Jersey and the Central Hudson Valley: New York,
Petroleum Exploration Society of New York, p. 25-105.MCKENZIE, D. P., 1978, Some remarks on the development of
sedimentary basins: Earth Planet. Sci. Lett., v. 40, p. 25-32.
MCLAUGHLIN, D. B., 1945, Type sections of the Stockton and
Lockatong Formations: Pennsylvania Acad. Sci. Proceedings,
v. 19, p. 102-113.
———, and WILLARD, B., 1949, Triassic facies in the DelawareValley: Pennsylvania Acad. Sci. Proceedings, v. 23, p. 34-44.
OLSEN, P. E., 1978, On the use of the term Newark for Triassic
and Early Jurassic rocks of eastern North America: Newsl.
Strat., v. 7, p. 90-95.
———, 1980, The latest Triassic and Early Jurassic formations ofthe Newark basin (eastern North America, Newark
Supergroup): stratigraphy, structure, and correlation: N.J.
Acad. Sci. Bull., v. 25, p. 25-51.
———, 1984 a, Comparative paleolimnology of the Newark
Supergroup: A study of ecosystem evolution: Unpub. Ph. D.
dissertation, Yale University, New Haven, CT, 726 p.———, 1984 b, Periodicity of lake-level cycles in the Late
Triassic Lockatong Formation of the Newark basin (Newark
Supergroup, New Jersey and Pennsylvania), in BERGER, A.;
IMBRIE, J.; HAYS, J.; KUKLA, G.; and SALTZMAN, B.; eds.,
Milankovitch and climate, pt. 1: Dordrecht, D. ReidelPublishing Co., p. 129-146.
———, 1985, Significance of great lateral extent of thin units in
the Newark Supergroup (Lower Mesozoic, eastern North
America): AAPG Bull., v. 69, p. 1444.
———, 1986, A 40-million-year lake record of early Mesozoic
climatic forcing: Science, v. 234, p. 842-848.
16
———, in press, Stratigraphy, facies, depositional
environments, and paleontology of the NewarkSupergroup, in HATCHER, R. H., and VIELE, G., eds., The
Appalachians and Oachitas, Geol. Soc. America, The
Geology of North America.
———, and FEDOSH, M., 1988, Duration of Early Mesozoic
extrusive igneous episode in eastern North America
determined by use of Milankovitch-type lake cycles:Geol. Soc. America Abst. with Prog., v. 20, p. 59.
———, and SCHLISCHE, R. W., 1988 a, Quantitative rift
basin evolution: Application of extensional basin filling
model to early Mesozoic rifts: Geol. Soc. America Abst.
with Prog., v. 20, p. 59.———, and ———, 1988 b, Unraveling the rules of rifts:
Lamont-Doherty Geological Observatory /1988
[Yearbook], p. 26-31.
———; ———; and GORE, P. J. W., eds., 1989, Tectonic,
Depositional, and Paleoecological History of Early
Mesozoic Rift Basins, Eastern North America:International Geological Congress Field Trip T-351,
Washington, D.C., American Geophysical Union, 174 p.
RA T C L I F F E , N. M., and BURTON, W. C., 1985, Fault
reactivation models for the origin of the Newark basin
and studies related to U.S. eastern seismicity: U.S. Geol.Survey Circular 946, p. 36-45.
———; ———; D'ANGELO, R. M.; and COSTAIN, J. K.,
1986, Low-angle extensional faulting, reactivated
mylonites, and seismic reflection geometry of the
Newark basin margin in eastern Pennsylvania: Geology,
v. 14, p. 766-770.REINEMUND, J. A., 1955, Geology of the Deep River coal
field, North Carolina: U.S. Geol. Survey Prof. Paper 246,
159 p.
RESSETAR, R., and TAYLOR, G. K., 1988, Late Triassic
depositional history of the Richmond and Taylorsvillebasins, eastern Virginia, in MANSPEIZER, W., ed.,
Triassic-Jurassic Rifting, Continental Breakup, and the
Origin of the Atlantic Ocean and Passive Margins: New
York, Elsevier, p. 423-443.
SCHLISCHE, R. W., and OLSEN, P. E., 1987, Comparison of
growth structures in dip-slip vs. strike-slip dominatedrifts: Eastern North America: Geol. Soc. America Abst.
with Prog., v. 19, p. 833.
———, and ———, 1988, Structural evolution of the
Newark basin, in HUSCH, J. M, and HOZIK, M. J., eds.,
Geology of the Central Newark Basin: 5th Annual Meeting of
the New Jersey Geological Association, p. 43-65.SMITH, R. B., and BRUHN, R. L., 1984, Intraplate extensional
tectonics of the eastern Basin-Range: Influences on structural
style from seismic reflection data, regional tectonics, and
thermal-mechanical models of brittle-ductile deformation:
Jour. Geophys. Res., v. 89, p. 5733-5762.
SMOOT, J. P., 1989, Fluvial and lacustrine facies of the earlyMesozoic Culpeper basin, Virginia and Maryland:
International Geological Congress Field Trip T-213,
Washington, D.C., American Geophysical Union, 15 p.
STECKLER, M. S., and WATTS, A. B., 1978, Subsidence of the
Atlantic-type continental margin off New York: Earth Planet.Sci. Lett., v. 41, p. 1-13.
———, WATTS, A. B.; and THORNE, J. A., 1988, Subsidence and
basin modeling at the U.S. Atlantic passive margin, in
Sheridan, R. E., and Grow, J. A., eds., The Atlantic
Continental Margin: U.S.: Geol. Soc. America, The Geology
of North America, v. I-2, p. 399-416.SWANSON, M. T., 1986, Preexisting fault control for Mesozoic
basin formation in eastern North America: Geology, v. 14, p.
419-422.
TAUXE, L.; MONAGHAN, M.; DRAKE, R.; CURTIS, G.; and
STAUDIGEL, H., 1985, Paleomagnetism of Miocene EastAfrican Rift sediments and the calibration of the geomagnetic
reversal time scale: Jour. Geophys. Res., v. 90, p. 4639-4646.
VAN HOUTEN, F. B., 1964, Cyclic lacustrine sedimentation, Upper
Triassic Lockatong Formation, central New Jersey and
adjacent Pennsylvania: Geol. Survey Kansas Bull., v. 169, p.
497-531.———, 1977, Triassic-Liassic deposits of Morocco and eastern
North America: Comparison: AAPG Bull., v. 61, p. 79-94.
VREELAND, J. H., and BERRONG, B. H., 1979, Seismic exploration
in Railroad Valley, Nevada, in NEWMAN, G. W., and GOODE,
H. D., eds., Basin and Range Symposium and Great BasinField Conference: Denver, Rocky Mountain Association of
Geologists, p. 557-569.
WERNICKE, B., and BURCHFIEL, B. C., 1982, Modes of extensional
tectonics: Jour. Struc. Geol., v. 4, p. 105-115.
WILLARD, B.; FREEDMAN, J.; MCLAUGHLIN, D. B.; RYAN, J. D.;
WHERRY, E. T.; PELTIER, L. C.; and GAULT, H. R., 1959,Geology and Mineral Resources of Bucks County,
Pennsylvania: Pennsylvania Geol. Survey, 4th Series, Bull.
C9, 243 p.
.
0
2
4
6
85 km
TW
TT
, s
SNPost-rift sediments
Onlap
5 km
Onlap
LLS
NW SE0
2
4
6
8
H T Pr F OB
P
Pd
Onlap5 kmA
B
C
W Ekm
FIG. 1—Cross-section and seismic lines of typical half-graben, showing the wedge-shaped geometry and theprogressive onlap of younger strata onto the hanging wall basement block on the hinged side of the (A) Newark(see figs. 5 and 7 for location), (B) Atlantis (see fig. 5 for location), and (C) Railroad Valley basins.Abbreviations for stratigraphic units in the Newark basin are: B, Boonton Formation; F, Feltville Formation; H,Hook Mountain Basalt; L, Lockatong Formation; O, Orange Mountain Basalt; P, Passaic Formation; Pd,Palisades diabase; Pr, Preakness Basalt; S, Stockton Formation; and T, Towaco Formation. TWTT is two-waytravel time. Seismic profiles were traced from figures presented by Hutchinson et al. (1986) and Vreeland andBerrong (1979).
Yf
x
W
B
Y
D
L
W cot
x cot
FIG. 2—Full-graben model, showing meaning of symbols used in the equations in the text.
.
Ae = 33%1
e = 67%2
e = 100%3
e = 133%4
e = 167%5
e = 200%6
e = 33%B
1
e = 67%
2
e = 100%3
Fluvial Sediments
Deep-water
Shallow-water
Lacu
strin
eS
edim
ents
FIG. 3—Diagrammaticrepresentation of the salientfeatures of the filling of a full-graben under conditions ofuniform subsidence andconstant volume input ofsediment per unit time; e equalspercent extension. See text forfurther explanation.
.
Drill Hole
A2 km
Time (millions of years)
0
1
2
3
0 1 2 3 4 5
BDrill Hole Data
Basin Average
Cu
mu
lati
ve
Basin Average
Drill Hole Data
C
0
1
2
0
Acc
um
ula
tio
n R
ate
1 2 3 4 5Time (millions of years)
Th
ickn
ess
(km
)(m
m/y
r)
FIG. 4—Half-graben filling model. (A) Cross section of half-grabenloosely based on the geometry of the Atlantis basin (fig. 1), in whichstratal surfaces, corresponding to time slices of 550,000 years,delineate areas of 1.5 x 106 m2 which were iteratively fit to the basin.Cumulative sediment thickness (B) and accumulation rate (C) showsimilar trends as in a full-graben (cf. fig. 9).
.
N
0 100 200 300 400 km
Newark
Pomperaug
Hartford
Deerfield
Gettysburg
Culpeper
Dan River/Danville
Deep River
Richmond
Taylorsville/Queen Anne
Fundy
Atlantis
Long Island
New York Bight
Nantucket
Farmville
Scotsburg
DavieCounty
Mesozoic
ExposedInferred
Rift Basins
Sanfordsub-basin
FIG. 5—Early Mesozoic rift basins of eastern North America. Names of basins mentioned in thetext are bold-faced. Modified from Olsen et al. (1989).
.
(Milankovitch Cycles)Lake Depth
Low High
21,0
00 y
r10
0,00
0 yr
400,
000-
year
cyc
le
Lake DepthLow High
"Maximum"
0.50 (g)
1.20 (f)
0.90 (e)
0.18 (d)
0.22 (c)
0.24 (b)
lation Rate(mm/yr)
km
StocktonFormation
Lithostratigraphy
0
1
2
3
4
5
6
7
Jura
ssic
Tri
assi
c
200
205
210
215
220
225
230
Ma
Chronostratigraphy
400,000yr. cycle
cycl
e c
ycle
Accumu-
ExtrusiveInterval
Boonton Fm.
PassaicFormation
LockatongFormation
0.36-0.51(a)
FIG. 6—Stratigraphic pattern of the Newark basin. Black lines in the stratigraphic column represent the most prominent gray andblack units for which their proper stratigraphic position is known. Accumulation rates were calculated using the thicknesses of the21,000-year-long Van Houten cycles. Letters refer to specific measured sections: a, Lower Lockatong Formation, Lumberville, PA(Olsen unpubl. data); b, Middle Lockatong Formation along NJ 29, near Byram, NJ (Olsen 1986); c, Perkasie Member of thePassaic Formation, US 202, Dilts Corner, NJ (Olsen et al. 1989); d, Ukrainian Member of the Passaic Formation, UkrainianVillage, NJ (Olsen unpubl. data); e, f, and g, Feltville, Towaco, and Boonton formations, respectively, from Army Corps ofEngineers drill cores, Pompton Plains to Clifton, NJ (Fedosh and Smoot 1988; Olsen and Fedosh 1988; Olsen et al. 1989)."Maximum" lake depth is the enveloping surface of the Milankovitch-period fluctuations in lake level. The extrusive intervalconsists of the following formations (in stratigraphic order): Orange Mountain Basalt, Feltville Formation, Preakness Basalt,Towaco Formation, and Hook Mountain Basalt.
.
FO
T
B
TH
T
Pr
F
Pr
O
HH
Pr O
Pr
P
S
Pd L
P
LP
kP
S
SLL
PO
S
S
LL
LLP
P
Pk
Pk
P
S
FJ
P
I
3
45
678
onla
p
onla
p
N
05
1015
2025
km
LL1
onla
p
75°0
0'
74°3
0'
75°3
0'
75°3
0'
74°0
0'
74°0
0'
40°3
0'
40°3
0'40
°45'
40°4
5'
41°0
0'
41°0
0'
74°0
0'
2
A
B
B'
FIG
. 7—
Geo
logi
c m
ap o
f th
e N
ewar
k ba
sin
of P
enns
ylva
nia,
New
Jer
sey,
and
New
Yor
k. D
ark
shad
ing
repr
esen
ts d
iaba
se in
trus
ions
, lig
ht s
hadi
ng r
epre
sent
s la
va f
low
s.D
otte
d lin
es a
re f
orm
line
s of
bed
ding
, and
thin
bla
ck li
nes
are
gray
and
bla
ck u
nits
in th
e Pa
ssai
c Fo
rmat
ion.
Bal
ls a
re o
n do
wnt
hrow
n si
de o
f no
rmal
fau
lts. A
ispo
sitio
n of
cro
ss s
ectio
n sh
own
in f
igur
e 1;
cro
ss s
ectio
n B
-B' s
how
n in
fig
. 8A
. Abb
revi
atio
ns a
re: B
, Boo
nton
For
mat
ion;
F, F
eltv
ille
Form
atio
n; H
, Hoo
k M
ount
ain
Bas
alt;
I, b
asem
ent i
nlie
rs; J
, Jac
kson
wal
d B
asal
t; L
, Loc
kato
ng F
orm
atio
n; O
, Ora
nge
Mou
ntai
n B
asal
t; P
, Pas
saic
For
mat
ion;
Pd,
Pal
isad
es d
iaba
se; P
k, P
erka
sie
Mem
ber
of P
assa
ic F
orm
atio
n; P
r, P
reak
ness
Bas
alt;
S, S
tock
ton
Form
atio
n; a
nd T
, Tow
aco
Form
atio
n. N
umbe
rs r
efer
to s
peci
fic
sect
ions
dis
cuss
ed in
text
and
fig
. 9: 1
,St
ockt
on, N
J; 2
, Lum
berv
ille,
PA
; 3, B
yram
, NJ;
4, D
ilts
Cor
ner,
NJ;
5, U
krai
nian
Vill
age,
NJ;
6, 7
, and
8, A
rmy
Cor
ps o
f E
ngin
eers
cor
es, P
ompt
on L
akes
to C
lifto
n,N
J. M
odif
ied
from
Sch
lisch
e an
d O
lsen
(19
88).
6 km30° 6.5°Perkasie Member
64 km
Stockton thickness: 1800 mStocktonthickness: 900 m
Height of Perkasie above top ofStockton: 1.9 km
Outcrop of Perkasie Member
Erosion Surface
Border fault
NW
6 km
13° 13°
Onlap
A
B
C5 km
2 km
2 km
B B'
FIG. 8—(A) Cross section A-A' of the Delaware River valley (see fig. 7 for location) before and after palinspasticrestoration of slip on the intrabasinal faults. (B) Cross-sectional geometry of palinspastically-restored basin justafter the deposition of the Perkasie Member. (C) Full-graben equivalent of basin shown in (B). The areas of theshaded regions in (B) and (C) are equal.
.
0
2
4
6
8
b
cd
a
0 5 10 15 20 25 30Time (millions of years)
Dep
th/T
hic
knes
s (k
m)
Subsidence(D)
Y+W
CumulativeSediment
Thickness (Y)
e
10
12
0.2
0.4
0.6
1.2A
ccu
mu
lati
on
Rat
e (m
m/y
r)
0 5 10 15 20 25 30
f h
g
1.0
0.8
0 5 10 15 20 25 30
"Max
" L
ake
Dep
th (
m) i
j k
m
n
o
0
40
80
120
160
Sto
ckto
n F
m.
Lock
aton
g F
m.
Pas
saic
Fm
.
Ext
rusi
ve Z
one
Boo
nton
Fm
.
a'b'
h
b"
d'
l
c'
FIG. 9—Predictions of the full-graben model (solidlines), based on the geometry depicted in figs. 7 and8, compared to data from the Newark basin (dottedlines and crosses). The Y + W curve represents theheight of the lake above the floor of the graben.Abbreviations are: a, filling equals subsidenceresulting in fluvial sedimentation, with accumulationrate (a') equal to subsidence rate; b, b', and b", onsetof lacustrine deposition and sedimentologic closure;c and c', rapid increase in lake depth as depositionalsurface subsides below hydrologic outlet; d and d',deepest lake predicted by model and onset ofhydrologic closure; e, major deviation fromcumulative thickness curve in Early Jurassic; f,hyperbolic decrease in accumulation rate predictedby model; solid cross represents data onaccumulation rates from cyclical lacustrine stratafrom scattered outcrops in the basin (see fig. 7);shaded cross represents estimate of accumulationrate in fluvial Stockton [estimated thickness(calculated from outcrops near location 1, fig. 7)divided by estimated duration]; g, anomalously highaccumulation rate in extrusive zone, a deviation fromthe model's predictions; h, decreasing accumulationrate in the Boonton Formation; i, 100-m-water depthline, the minimum depth for the formation ofmicrolaminated sediments (Olsen 1984a); j, actualestimated first occurrence of microlaminatedsediments; k, actual last occurrence ofmicrolaminated sediments in lower PassaicFormation; l, predicted last occurrence ofmicrolaminated sediments; m, slow decrease in lakedepth; n, anomalous "superwet" climatic anomaly ofthe Ukrainian Member; o, anomalously deep Jurassiclakes.
A
B
Present-day erosionsurface after additionaltilting
C
FIG. 10—Effects of asymmetric basin subsidence. (A) Normal filling sequence underconditions of uniform subsidence. (B) Accelerated tilting causes sediment and water toshift toward the border fault side of the basin and results in the onlap of youngest strataonto previously-deposited strata. (C) Antithetic faulting creates a smaller sub-basin.Fault need not be located within present extent of the basin. Darkest shading in (B) and(C) corresponds to Jurassic deposits of the Newark basin.
.
Flu
vial
and
Allu
vial
Sed
imen
ts
T R I A S S I CJ U R A S S I C
205
215
225
195
235Ma
HA
RT
FO
RD
BA
SIN 0.
20
≈0.5
0
0.25
Por
tland
Fm
New
Hav
enA
rkos
e
CU
LP
EP
ER
BA
SIN
Tib
bsto
wn/
Cat
harp
inC
reek
Fm
s
Bal
ls B
luff
Silt
ston
e
Man
assa
sS
ands
tone
0.100.19
0.16
≈1.2
6
FU
ND
Y B
AS
IN
0.03
0.02
0.01
0.04
Wol
fvill
e F
m
Blo
mid
on
McC
oy
Sco
ts B
ay B
rook
Fm
Fm Fm
Acc
umul
atio
n ra
te(m
m/y
r) d
eter
min
ed b
yth
ickn
ess
of fi
xed-
perio
dla
cust
rine
cycl
es
divi
ding
form
atio
n's
thic
knes
s
Acc
umul
atio
n ra
te (
mm
/yr)
est
imat
ed b
y
by it
s pa
leon
tolo
gica
llyes
timat
ed d
urat
ionLo
ckat
on
g
NE
WA
RK
BA
SIN 0.
60
0.40
0.24
0.22
0.18
≈1.0
00.
50B
oont
on F
m
Pas
saic
Fm
Fm
230
220
210
200
190
"Dee
p-W
ater
" La
cust
rine
Sed
imen
ts
Lacu
strin
e S
hale
s an
d S
iltst
ones
Lava
Flo
ws
and
Inte
rcal
ated
Sed
imen
ts
San
ford
Fm
Pek
in F
mC
um
no
ck
0.25
0.26
0.50
San
ford
Su
b-b
asin
DE
EP
RIV
ER
BA
SIN
RIC
HM
ON
D B
AS
IN 0.41
0.63
0.91
Low
er B
arre
n B
eds
Vin
itaO
tterd
ale
Ss
Bed
sS
tock
ton
Fm
FIG
.11—
Chr
onos
trat
igra
phic
sec
tions
of
seve
ral N
ewar
k Su
perg
roup
bas
ins,
sho
win
g tr
ends
in a
ccum
ulat
ion
rate
. In
all o
f th
e ba
sins
, the
tran
sitio
n fr
om f
luvi
al to
lacu
stri
ne d
epos
ition
occ
urre
d at
dif
fere
nt ti
mes
. The
mar
ked
incr
ease
s in
acc
umul
atio
n ra
te a
nd "
max
imum
" la
ke d
epth
ass
ocia
ted
with
the
extr
usiv
e ep
isod
e oc
curr
edsi
mul
tane
ousl
y in
all
basi
ns th
at c
onta
in a
Jur
assi
c se
ctio
n. M
odif
ied
from
Ols
en e
t al.
(198
9).
