The Late Jurassic Tithonian, a greenhouse phase in the ...myslu.stlawu.edu/~ahusinec/Publications/1 - Papers... · The Late Jurassic Tithonian, a greenhouse phase in the Middle Jurassic–Early

The Late Jurassic Tithonian, a greenhouse phase in the MiddleJurassic–Early Cretaceous ‘cool’ mode: evidence from the cyclicAdriatic Platform, Croatia

ANTUN HUSINEC* and J. FRED READ�*Croatian Geological Survey, Sachsova 2, HR-10000 Zagreb, Croatia�Department of Geosciences, Virginia Tech, 4044 Derring Hall, Blacksburg, VA 24061, USA (E-mail:[email protected])

ABSTRACT

Well-exposed Mesozoic sections of the Bahama-like Adriatic Platform along

the Dalmatian coast (southern Croatia) reveal the detailed stacking patterns of

cyclic facies within the rapidly subsiding Late Jurassic (Tithonian) shallow

platform-interior (over 750 m thick, ca 5–6 Myr duration). Facies within

parasequences include dasyclad-oncoid mudstone-wackestone-floatstone and

skeletal-peloid wackestone-packstone (shallow lagoon), intraclast-peloid

packstone and grainstone (shoal), radial-ooid grainstone (hypersaline

shallow subtidal/intertidal shoals and ponds), lime mudstone (restricted

lagoon), fenestral carbonates and microbial laminites (tidal flat).

Parasequences in the overall transgressive Lower Tithonian sections are 1–

4Æ5 m thick, and dominated by subtidal facies, some of which are capped by

very shallow-water grainstone-packstone or restricted lime mudstone;

laminated tidal caps become common only towards the interior of the

platform. Parasequences in the regressive Upper Tithonian are dominated by

peritidal facies with distinctive basal oolite units and well-developed laminate

caps. Maximum water depths of facies within parasequences (estimated from

stratigraphic distance of the facies to the base of the tidal flat units capping

parasequences) were generally <4 m, and facies show strongly overlapping

depth ranges suggesting facies mosaics. Parasequences were formed by

precessional (20 kyr) orbital forcing and form parasequence sets of 100 and

400 kyr eccentricity bundles. Parasequences are arranged in third-order

sequences that lack significant bounding disconformities, and are evident on

accommodation (Fischer) plots of cumulative departure from average cycle

thickness plotted against cycle number or stratigraphic position. Modelling

suggests that precessional sea-level changes were small (several metres) as

were eccentricity sea-level changes (or precessional sea-level changes

modulated by eccentricity), supporting a global, hot greenhouse climate for

the Late Jurassic (Tithonian) within the overall ‘cool’ mode of the Middle

Jurassic to Early Cretaceous.

Keywords Adriatic Platform, Croatia, greenhouse, Late Jurassic, precession.

INTRODUCTION

The Late Jurassic–Early Cretaceous has beenconsidered to lie in the Mesozoic ‘cool’ mode byFrakes et al. (1992). Interpretation of this ‘cool’mode is supported by oxygen isotope signals

(Veizer et al., 2000) and by the pCO2 plotsGEOCARB III of Berner & Kothavala (2001). Atthe long-term scale, these climate proxies are inagreement, but at shorter time scales, there issome conflict both in degree of warming orcooling and timing of the events (Veizer et al.,

Sedimentology (2007) 54, 317–337 doi: 10.1111/j.1365-3091.2006.00837.x

� 2006 The Authors. Journal compilation � 2006 International Association of Sedimentologists 317

2000; Royer et al., 2004). The cyclic Late Jurassicrecord of the shallow Adriatic Carbonate Plat-form, Croatia, is used in this paper to providehigher resolution palaeoclimate data to helpresolve this.

Stacking patterns of parasequences have beenused to document precessional and eccentricityforcing in Tithonian-Berriasian units in France(Strasser, 1994) and England (Anderson, 2004a).These areas differ markedly from the AdriaticPlatform in that they were much more slowlysubsiding and more humid. The high subsidencerates of the Adriatic Platform should potentiallyyield a relatively complete record of eustaticsignals that affected the platform, given that ittended to be aggraded to near sea-level, andprobably had fewer missing beats than moreslowly subsiding areas elsewhere.

Because shallow-water carbonate platforms lieclose to sea-level, they are sensitive tohigh-frequency global sea-level changes. Conse-quently, sea-level changes in a purely green-house, relatively ice-free world should leave adifferent record within the platform stratigraphythan sea-level changes driven by small tomoderate amounts of ice in a cooler, but non-ice-house world. Shallow interiors (<10 m) ofsuch platforms if formed under relatively ice-free greenhouse conditions are likely to containmany small parasequences formed by preces-sion-eccentricity driven or even sub-Milanko-vitch sea-level changes (Goldhammer et al.,1990; Read, 1995; Zuhlke et al., 2003) or ifdeposited in a few tens of metres, poorly cyclicshallow subtidal successions. In contrast, plat-forms that develop under moderate amounts ofglobal ice associated with a cooler (but noticehouse) Earth (transitional eustasy of Read,1995, and ‘doubthouse’ of Miller et al., 1998)appear to have distinctly different parasequencestacking, probably driven by obliquity andshort-term and long-term eccentricity, and alsoshow periodic emergence and palaeosol forma-tion on the platform. In these platforms, only afew of the precessional beats flood the platformand are preserved in the platform record, theremainder not being preserved on the platformtop. These are end-member scenarios, and theactual stratigraphic record may be a morecomplicated mix of the end-members.

Stacking patterns of the platform-interior para-sequences of the Adriatic Platform of Croatia,which lie in the precessional band, suggest thatthe Late Jurassic Tithonian stage was hot green-house within the overall Middle Jurassic–Early

Cretaceous ‘cool’ mode (Frakes et al., 1992).Additionally, the parasequence stacking patternson the platform record third-order relative sea-level changes that can be evaluated against theglobal (Haq et al., 1987) curve and the Arabiansea-level curve (Haq & Al-Qahtani, 2005).

GEOLOGIC AND TECTONIC SETTING OFADRIATIC PLATFORM

The Adriatic Platform (Fig. 1) was part of aregional Permian to Early Triassic mega-platformthat was attached to Gondwana along the south-ern margin of Tethys. Following Middle Triassicrifting, this mega-platform broke into large plat-forms that included the Adria microplate. TheAdria microplate then underwent subsequentEarly Jurassic rifting to form the Adriatic andthe Italian Apulian and Apennine platforms (e.g.Pamic et al., 1998; Bosellini, 2002; Jelaska, 2003;Vlahovic et al., 2005). The Mesozoic Adriaticplatform consists of an immense (up to 6 kmthick) pile of shallow-water carbonates, punctu-ated by periods of subaerial exposure, palaeokarstand bauxites and by several pelagic (incipientdrowning) episodes (Fig. 2). Intra-platformtroughs were buried beneath syn-orogenic sedi-ments and were deformed during the CenozoicAlpine orogeny, when the Adriatic Platformbecame part of the peri-Adriatic mountain belt.Subsidence history curves (Husinec & Read, 2005;Husinec & Jelaska, 2006; Fig. 3) show that subsi-dence rates on the Late Jurassic Adriatic platformwere rapid, up to 15 cm kyr)1, slowing into theAptian-Albian. Subsidence rates then increasedagain into the Late Cretaceous with onset ofcollision. The relatively high subsidence rates (upto three times as fast as the Appalachian Cambro-Ordovician passive margin, USA; Koerschner &Read, 1989) favour a fairly complete stratigraphicsuccession being preserved on the platform forthe Tithonian.

Biostratigraphic control

The restricted shallow waters of the Late JurassicAdriatic Platform interior precluded the occur-rence of fully marine organisms, and carbonateproducers were limited predominantly to benthicforaminifera and calcareous green algae. Conse-quently, there was not an opportunity to recoverstratigraphic records with high-resolution chro-nology provided by open-marine organismssuch as ammonites, radiolarians or calpionellids.

318 A. Husinec and J. F. Read

� 2006 The Authors. Journal compilation � 2006 International Association of Sedimentologists, Sedimentology, 54, 317–337

In addition, the long-distance sequence strati-graphic correlation with coeval Tethyan succes-sions is hampered because the platform-interiorsuccession does not contain sediments depositedin a spectrum of settings, but instead are mainlyshallow peritidal carbonates. Biostratigraphiccontrol for Upper Jurassic Tithonian platform-interior strata consists of two dasycladaceanbiozones (Early Tithonian Clypeina jurassicas.str., and Late Tithonian Clypeina jurassica andCampbelliella milesi), which have been describedin detail by Velic (1977) and Velic & Sokac(1978). Numerous references on Late Jurassicbiostratigraphy of both benthic foraminifera andcalcareous algae from this region show that thereis a reasonable correlation with other Mediterra-nean localities where benthic biozones are calib-rated with the established ammonite andcalpionellid schemes (De Castro, 1993; and refer-ences therein). The determined local stratigraph-ical ranges of individual benthic species, ifstudied in the context of the stratigraphic rangesof the entire microfossil association, seem toremain constant at a regional, platform-widescale. The strata underlying the studied Titho-nian succession are characterized by the scarcealga Clypeina jurassica [its first appearancedatum (FAD) is registered just above the stratawith last recorded occurrence of foraminiferaPraekurnubia crusei], associated with foramini-fera of longer stratigraphic ranges, including

Kurnubia palastiniensis, Pseudocyclammina lit-uus and Valvulina lugeoni. The Late Kimmerid-gian age is in agreement with the data from theJura Mountains, where FAD of C. jurassica isreported in the ammonite euxodus Zone (Bernier,1984). The beginning of Tithonian is character-ized by more frequent findings of C. jurassica inassociation with Pseudoclypeina distomensis, analga whose FAD in the Dinarides is at theKimmeridgian–Tithonian boundary. The upperpart of Tithonian is characterized by abundantfindings of the alga Campbelliella. Following thedisappearance of Campbelliella and several for-aminiferan species that are restricted to LateJurassic (e.g. Parurgonina caelinensis), the begin-ning of Berriasian is characterized by FADs ofseveral index species of calcareous algae in thearea investigated, including Otternstella lemmen-sis, Clypeina isabelae, Clypeina parasolkani,Clypeina catinula, Humiella sardiniensis, Humi-ella catenaeformis and Salpingoporella katzeri(Husinec & Sokac, 2006). Magnetostratigraphic orstrontium-isotope chemostratigraphic datingtechniques are needed in the future to improvethe chronostratigraphic correlation.

Tisljar et al. (2002) suggested that the begin-ning of Berriasian is characterized by a regionalregression, as evidenced by peritidal limestones,black-pebble breccia, emersion breccia, residualclays, swamp deposits and palaeosols, sporad-ically even with bauxites. In the study area,

Fig. 1. (A) General location mapshowing outline of the MesozoicAdriatic Carbonate Platform withinCroatia and environs (modified fromVelic et al., 2002). Black rectangleoutlines study area shown enlargedin (B). (B) Detailed map of studyarea showing location of sectionsLSL (Lastovo Island) and MSV(Mljet Island).

Late Jurassic greenhouse 319


several emersion breccia and/or residual clayhorizons have been reported from the basalBerriasian of Lastovo (Husinec, 2002). In thewider Tethyan realm, the overall regressioncharacterized the Tithonian, and it was markedby continuous-exposure conditions on the peri-Tethys margins, and by the associated long-termforced regression and gravitational deposits inbasinal areas (Jacquin et al., 1998, and refer-ences therein).

METHODS

Data in this paper are based on detailed sectionsmeasured by Husinec et al. at the CroatianGeological Survey (see Husinec, 2002). The sec-tions used are detailed bed-by-bed logs recordingstratigraphic position, sedimentary structures,textures, Dunham rock types and fossils inclu-ding age-diagnostic microfossils. These werereformatted to provide colour-coded computer-

Fig. 2. (A) Generalized strati-graphic column of MesozoicAdriatic Platform showing majorunits and geological events (Jelaska,2003). The Late Jurassic (Tithonian)interval of study is shown by blackbar alongside column. (B) Schematiccolumns of sections used in thestudy showing regional faciesrelations. Correlation is inferredbased on facies stacking patternsand unconformity surfaces (basalBerriasian). The MSV section islocated ca 40 km inboard from theplatform margin. Heavy blackvertical bars mark measuredsections.



drafted columns emphasizing Dunham rock typesand dominant grain types. This simplified thepicking of parasequences and parasequence sets.Accommodation (Fischer) plots were generatedby graphing cumulative departure from meancycle thickness vs. either cycle number or strati-graphic position in the section (Fischer, 1964;Read & Goldhammer, 1988; Sadler et al., 1993;Day, 1997). These accommodation plots showperiods of increased accommodation as risinglimbs and decreased accommodation as fallinglimbs on the graphs. Note that plotting cumul-ative departure vs. stratigraphic position does notaffect the relative amplitudes of the plots, butonly the steepness of the rising or falling limbs(Day, 1997). Fourth-order (0Æ1–0Æ5 Myr) and third-order (0Æ5–5 Myr) relative sea-level cycles werepicked from the accommodation plots. Effects ofcompaction were neglected because much of thedewatering in carbonates is assumed to be earlyand occurring during deposition of a cycle; sub-sequent burial compaction is difficult to accessgiven the variable timing of cementation in car-bonates. It should be stressed that the Fischerplots only provide a relative measure of sea-level

changes. This is because not all parasequencesshallow to each highstand of sea-level, thus notall accommodation was filled during shallowing(Boss & Rasmussen, 1995). The estimates ofamplitudes of relative sea-level cycles are likelyto be minimum estimates. Modelling was doneusing Phil� 1Æ5 developed by Scott Bowman(Petrodynamics, Houston, TX, USA). For themodelling, precessional and obliquity sea-levelcycles were convolved with eccentricity sea-levelcycles, as in the deep sea Pleistocene isotoperecord. Another way to model these, would be tohave precessional sea-level amplitudes modula-ted by eccentricity. The main difference betweenthe approaches would be in the effect on thedeeper water platform facies during sea-level lowstands. As we are modelling the shallow-waterstratigraphy, it is probable that either approachwould yield a similar stacking of cyclic facies.

FACIES AND ENVIRONMENTS

Facies within the interior of the Late Jurassicportion of the Adriatic Platform are summarized

Fig. 3. Accommodation history plot using base of Jurassic as reference horizon, for Jurassic-Cretaceous interval,Adriatic Platform (from Husinec & Jelaska, 2006). The Jurassic to Late Cretaceous is the passive margin stage of theAdria Microplate. The Tithonian interval has the highest subsidence rate (12–15 cm kyr)1) of any of the stages,including the Late Cretaceous collisional event.



in Table 1; an idealized parasequence is shownin Fig. 4, and the facies are illustrated in Fig. 5.The detailed stacking patterns are shown inFigs 6 and 7. Minimum estimates of waterdepths of each facies were determined bymeasuring the stratigraphic distance from thebase of each facies to the base of the cappingtidal flat facies. The estimate is too small ifcompaction is significant, or if sea-level wasfalling as the parasequence shallowed (Koersch-ner & Read, 1989). Estimates of water depthswould be too high if there was significantload-induced subsidence during deposition ofeach parasequence. These estimates of waterdepths using fenestral/microbial carbonate as asea-level datum (Fig. 8) are thus relative meas-ures, and show that the subtidal facies hadoverlapping depth ranges, implying mosaic-likefacies patterns. The facies identified in thisstudy are outlined below (from deepest toshallowest).

Dasyclad-oncoid mudstone-wackestone-floatstone (‘deeper’ lagoon)

These facies (described in Table 1) are lightbrown, massive to thick-bedded mudstone-wackestone-floatstone (Fig. 5A) composed ofvariable amounts of dasyclads and oncoids,and small amounts of benthic foraminiferansand molluscs. They formed in quiet, slightlydeeper lagoon settings (down to depths of 3 m,and rarely as much as 8 m; Fig. 8) below thezone of frequent wave reworking. Given theisolated platform setting, this low-energy envir-onment may have resulted from the shelteringeffect of reefs, islands and tidal flats along theeastern, windward margin of the platform (Velicet al., 2002). Without this sheltering effect,bottom sediments at such shallow lagoonaldepths probably would have been winnowedduring storms, as in the present-day Bahamagrapestone lithotopes, and the fines transportedoff-platform or into sheltered peritidal settingsin the lee of islands (Illing, 1954; Imbrie &Purdy, 1962). Lack of open-marine foraminifer-ans, echinoderms and cephalopods, indicatesthat the platform-interior was relatively re-stricted compared with the open ocean, andpossibly had near-normal to slightly elevatedsalinities for these facies. The abundant micro-bial oncoids suggest that waters were relativelysupersaturated with respect to calcium carbo-nate, compared with the present-day conditions(Riding, 1992). Intense bioturbation homo-

genized these sediments obliterating any sedi-mentary structures.

Skeletal-peloid wackestone-packstone(shallow lagoon)

These facies (described in Table 1) are lightbrown, thin-bedded to thick-bedded wackestone-packstone (Fig. 5B) composed of peloids andskeletal fragments of dasyclad algae, benthic fora-minifera and hydrozoans (?), oncoids, and sparseooids and intraclasts. They formed in water depthsfrom 0 to 3 m (Fig. 8) in low-energy settings.Waters were moderately restricted and poss-ibly metahaline, and hence lack open-marineoceanic assemblages, although the presence ofhydrozoans (?) suggests near-normal salinities.

Intraclast-peloid packstone and grainstone(shoal-water)

These facies (described in Table 1) are lightbrown thin-bedded to thick-bedded packstone-grainstone (Fig. 5C) composed of intraclasts,peloids, and locally, minor ooids; in a few beds,there are common dasyclad algae, benthic fora-minifera, hydrozoan fragments and less commononcoids and molluscs. These formed in shallow-wave and current-agitated settings by accumu-lation of resident dasyclads and foraminifera alongwith peloids and intraclasts derived fromreworked muds or micritization of grains. Waterdepths probably ranged from 0 to 2 m (Fig. 8) andoverlapped those of the oolitic facies, but waterswere not sufficiently supersaturated to form ooids.Grainstones may have developed in areas of mobilesubstrates whereas packstones may have devel-oped beneath stable bottoms, perhaps beneathsubtidal mats or vegetation (cf. Bathurst, 1973).

Radial-ooid grainstone (hypersaline shallowsubtidal/intertidal shoals and ponds)

These facies (described in Table 1) are thin tothick, horizontally bedded grainstones (less com-monly packstones) that lack high-energy sedi-mentary structures. They include light grey unitsdominated by fine-grained ooids and darker,brown to grey units composed of poorly sorted,sand-sized to granule-sized whole, fragmentedand recoated ooids (‘vadoids’ of Tisljar, 1985).Oolitic units (Fig. 5D) also contain some grape-stone-like aggregates (mainly of ooids) as well aslesser amounts of peloids, gastropods and algae.The ooids appear to be primary radial calcite



Table

1.

Tit

hon

ian

pla

tform

inte

rior

facie

s.

Facie

sT

idal

flat

Rest

ricte

dla

goon

Hyp

ers

ali

ne

shall

ow

subti

dal/

inte

rtid

al

shoals

an

dp

on

ds

Sh

oal-

wate

rM

od

era

tely

shall

ow

lagoon

‘Deep

er’

lagoon

Lit

holo

gy

Fen

est

ral

lim

em

ud

ston

e-w

ackest

on

e;

Mic

robia

lla

min

ite;

Pla

nar

lam

init

e

Un

foss

ilif

ero

us

lim

em

ud

ston

es

Ooid

gra

inst

on

e-

packst

on

eIn

tracla

st-p

elo

idp

ackst

on

ean

dgra

inst

on

e,

locall

ysk

ele

tal

Skele

tal-

pelo

idw

ackest

on

e-p

ackst

on

eL

ime

mu

dst

on

eto

on

coid

-dasy

cla

dw

ackest

on

e-fl

oats

ton

e

Colo

rL

igh

tbro

wn

Lig

ht

bro

wn

Dark

bro

wn

togra

y(c

oars

er

ooli

tes)

;li

gh

tgra

y(fi

ner

ooli

tes)

Lig

ht

bro

wn

Lig

ht

bro

wn

Lig

ht

bro

wn

Sed

imen

tary

stru

ctu

res

Wavy

lam

inati

on

,com

mon

LL

H-s

trom

ato

lite

s;p

oly

gon

al

dess

icati

on

cra

cks;

fen

est

ral

lam

inati

on

;m

illi

metr

ela

min

ae

of

lim

em

ud

ston

ean

dp

elo

id-s

kele

tal

packst

on

e

Th

in-b

ed

ded

,h

om

ogen

ou

sm

icri

te;

rare

bu

rrow

fen

est

rae

Th

into

thic

kbed

ded

;sc

arc

eero

sion

al

con

tacts

wit

hu

nd

erl

yin

gfe

nest

ral

carb

on

ate

s

Th

into

thic

kbed

ded

Th

into

med

ium

bed

ded

Th

ick-b

ed

ded

,com

mon

lyh

eavil

ybio

turb

ate

d

Fabri

cP

oorl

yso

rted

,m

ud

-to

gra

nu

le-s

ize

Dom

inan

tly

mu

d-s

ize

Poor-

tow

ell

-sort

ed

,fi

ne-s

an

dto

gra

nu

le-s

ize

Ran

ge

from

well

-so

rted

top

oorl

yso

rted

,fi

ne

san

dto

gra

nu

le-s

ize

Poorl

yso

rted

,m

ud

matr

ixw

ith

san

d-

tora

regra

nu

le-s

ize

gra

ins

Poorl

yso

rted

mu

ds

wit

hsa

nd

-to

gra

nu

le-s

ize

gra

ins

Gra

inty

pes

Mic

riti

zed

pelo

ids,

intr

acla

sts

Rare

pelo

ids

Ooid

s(c

om

mon

lybro

ken

an

dre

heale

d),

pelo

ids,

gra

pest

on

e-l

ike

aggre

gate

s

Intr

acla

sts,

pelo

ids,

scarc

eon

coid

san

dooid

s

On

coid

s,p

elo

ids,

scarc

eooid

san

din

tracla

sts

On

coid

s,h

yd

rozoan

s,ra

rem

icri

tein

tracla

sts

Bio

taC

alc

are

ou

salg

ae,

small

ben

thic

fora

min

ifera

,an

dcalc

ified

cyan

obacte

ria

(Cayeu

xia

?)

Foss

ils

abse

nt

tosp

ars

e;

rare

lym

ay

have

gast

rop

od

san

dost

racod

s

Calc

are

ou

salg

ae,

gast

rop

od

sC

alc

are

ou

salg

ae

an

dben

thic

fora

min

ifera

an

dm

oll

usc

scom

mon

ina

few

bed

s

Com

mon

calc

areou

salg

ae

an

dben

thic

fora

min

ifera

,biv

alv

es,

gast

rop

od

s

Com

mon

calc

are

ou

salg

ae

an

dben

thic

fora

min

ifera

,le

sscom

mon

ost

racod

s,m

oll

usc

s,gast

rop

od

s,biv

alv

es,

en

cru

sters

,h

yd

rozoan

s(?)

En

vir

on

men

tw

ate

rd

ep

thT

idal

zon

e0–2

m0–2

m0–3

m0–4

m0–6

m;

max.

8m



ooids that formed from seawater with low Mg/Caratios compatible with Late Jurassic seawater(Stanley & Hardie, 1998), although Strasser

(1986) suggested that slightly younger Purbeckianooids were originally high-Mg calcite. The finer-grained oolitic units are shallow marine ooidshoal facies. Where there was more restriction, asin hypersaline ponds and restricted lagoons, theooids grew to relatively large sizes (cf. Loreau &Purser, 1973). Periodic exposure caused grainbreakage and regrowth of ooid cortices withsubmergence (Tisljar, 1985). Many of these ooliticunits are transgressive and located at bases ofparasequences, whereas thin oolitic units in theupper parts of parasequences are regressive(Husinec & Read, 2006; Fig. 5F).

Lime mudstone (restricted lagoon)

These facies (described in Table 1) are lightbrown, thin-bedded homogenous lime mudstonewith very rare peloids, gastropods and ostracods.They formed in nearshore very shallow (mainly<1 m), low-energy restricted settings seaward oftidal flats. Waters may have been hypersaline assuggested by the very restricted biotas.

Fenestral carbonates and microbial laminites(tidal flat)

These facies (described in Table 1) are lightbrown, and include planar (mechanical) lami-nites, wavy or crinkly microbial laminites andfenestral laminites. They contain peloids, smallerintraclasts, benthic foraminifera, calcareous algaeand calcified cyanobacterial tufts (Cayeuxia?),and less common ostracods. The microbial lam-inites are common in the Lower Tithonian plat-form-interior and formed in intertidal settingswith well-developed microbial mat covers andlow to moderate sediment influx; this formedalternating mat and sub-centimetre sediment-richlayers with only a few fenestrae (cf. Logan et al.,1974). The laminoid fenestral carbonates (Fig. 5Gand H) are common in the oolitic Upper Titho-

Fig. 4. Idealized succession of facies within a Titho-nian parasequence. Oolitic units at the base of para-sequences are generally absent from the LowerTithonian. In the Upper Tithonian, the basal ooliticunits are mainly transgressive (upward narrowingtriangle). Although facies are shown as a shallowingupward succession (upward widening triangle), theyhave overlapping depth ranges implying facies mosa-ics. Symbols are explained in Fig. 7.

Fig. 5. Thin-section photographs of typical facies. Scale bar is 2 mm. (A) Oncoid floatstone with foraminifera Par-urgonina caelinensis. (B) Skeletal wackestone-packstone with numerous fragments of dasyclad Clypeina jurassica.(C) Skeletal-intraclast-peloid grainstone with numerous fragments of dasyclad Campbelliella (curved sparry grainswith micritic rinds). (D) Poorly sorted ooid grainstone composed of whole and fragmented-and-recoated radial ooids,some with cerebroid outlines (arrows). (E) Transgressive indurated contact between fenestral, tidal flat carbonatewith dispersed ooids and grapestone-like aggregates of ooids and peloids (bottom), and the overlying ooid packstone(top). (F) Contact between meniscus micrite-cemented ooid grainstone (bottom) and the overlying skeletal (Camp-belliella) wackestone-packstone (top). Aragonitic grains have been leached; lack of leaching of ooids indicates pri-mary radial calcite mineralogy. (G) Fenestral limestone composed of intraclasts, peloids and lumps of calcifiedfilamentous cyanobacteria(?), some with radiating habit. Some larger fenestrae are lined with micritic cement andfine turbid bladed calcite. (H) Fenestral peloid-ooid grainstone composed of very poorly sorted peloids and com-posite grains, many of which are micritized and superficially coated. Oolitic coats are composed of thin concentricenvelopes of yellowish radial calcite (arrows).



A B

C D

E F

G H



nian, and formed in well-drained organic-poor,intertidal to supratidal flats with a thin mat cover,high sediment influx and incipient cementation(Logan et al., 1974). Teepees generally are absent.Hypersalinity is suggested by lack of burrowing inthese facies and lack of charophytes, which arefreshwater algae (cf. Hardie & Ginsburg, 1974).

PARASEQUENCES AND STACKINGPATTERNS

We use the following definitions in the sense ofVan Wagoner et al. (1988). A parasequence is ashallowing-upward, conformable succession ofgenetically related strata bounded by marine

Fig. 6. Stratigraphic columns of the Tithonian sections showing facies stacking, parasequences, and parasequencesets. Location of sections shown in Fig. 1B. Heavy black vertical bars mark portions of sections enlarged in Fig. 7. (A)Lower Tithonian section MSV; section lies in the more interior location than section LSL. (B) Lower Tithoniansection LSL. (C) Upper Tithonian section LSL.



flooding surfaces. A parasequence set is a succes-sion of genetically related parasequences whichform a distinctive stacking pattern that may bebounded by major flooding surfaces, sequenceboundaries, or boundaries of systems tracts. Asystems tract of a depositional sequence maycontain one or more parasequence sets. TheTithonian succession consists of a lower unit ofdominantly subtidal parasequences, and anupper unit dominated by peritidal and oolite-bearing parasequences (Fig. 6). Moreover, theupper part is characterized by the presence ofthe dasyclad alga Campbelliella. However, thereneed not necessarily be coincidence between

lithologically defined lower and upper Tithonianunits and the Early and Late Tithonian agedefined by the fossils. An idealized shallowing-upward facies succession is shown in Fig. 4,although the full complement of facies is seldomobserved within a single parasequence. The lackof a high-resolution geochronology on the plat-form (and indeed most platforms) makes assign-ing fifth-order and sixth-order cycle nomenclature(e.g. Anderson, 2004a,b) potentially risky becauseof the possibility of sub-Milankovitch cyclesbeing confused with precessional cycles, theproblem of autocycles, and the problem of differ-entiating, with surety, between precessional,obliquity and short-term eccentricity cycles(Zuhlke et al., 2003). Thus we have opted to usethe descriptive terms ‘parasequence’ and ‘parase-quence sets’, which do not have a time connotation.

Lower Tithonian parasequences andparasequence sets

The Lower Tithonian parasequences commonlyare 0Æ7–4Æ5 m thick, but a few are up to 8 m thickand probably are compound parasequences. Theyare dominated by subtidal facies (Figs 6A and7A), typically skeletal-oncoidal wackestone/floatstone and mudstone, grading up into intra-clast-peloid packstone-grainstone. Some parase-quences (36%) are capped by restricted,unfossiliferous lime mudstone, a few of whichcontain fenestrae suggestive of spar-filled bur-rows. Intraclast-peloid packstone-grainstone andskeletal-peloid wackestone-packstone also arecommon as parasequence caps (31% and 24%,respectively), while <10% of parasequences arecapped by oncoid floatstone (Fig. 8A). Furtherinto the platform-interior (Mljet Island section;Figs 6B and 7B), the Early Tithonian parase-quences are still dominated by muddy subtidalfacies, but grainstones are less common, andfenestral, planar and microbial laminites capmost parasequences (55%; Fig. 8B). Oncoid float-stone and restricted lime mudstone cap someparasequences (18% and 13%, respectively),whereas intraclast-peloid packstone and skel-etal-peloid wackestone-packstone cap few(<10%) parasequences.

The Lower Tithonian parasequences are ar-ranged in sets commonly 10 m thick, but rangingfrom 8 to 15 m and composed of three to sixparasequences (Fig. 6A and B). Parasequenceswithin each set tend to be characterized by pro-gressively shallower lithofacies upward within theset (cf. Anderson, 2004a,b). That is, parasequences

Fig. 6. Continued.



Fig. 7. Enlarged portions of the stratigraphic columns of Fig. 6. (A) Subtidal parasequences typical of the LowerTithonian at section LSL (Fig. 1B). (B) Peritidal cycles typical of the upper part of the Lower Tithonian section MSV(Fig. 1B) in more interior location relative to section LSL. (C) Typical Upper Tithonian parasequences with thickbasal ooid grainstone units and laminate caps in section LSL (Fig. 1B).



in the lower part of a set consist dominantly ofoncoid-skeletal wackestone-mudstone and skel-etal-intraclastic subtidal facies, whereas theuppermostparasequence(s) are capped by restric-ted lime mudstones some of which are fenestral;caps in the more interior part of the platform aremicrobial and planar laminites. Some of thesefacies-defined parasequence sets are evident on theFischer plots as small rises and falls, but many arenot.

Upper Tithonian parasequences andparasequence sets

The Upper Tithonian parasequences are 1–3 mthick and are rarely more than 4 m thick. Thebasal portion of the section is dominated byparasequences consisting of intraclast-peloidpackstone-grainstone capped by fenestral-lamin-ated carbonates. The bulk of the Upper Tithonianis oolitic (Figs 6C and 7C). These parasequences(Fig. 4) consist of basal transgressive oolites(Husinec & Read, 2006), overlain by skeletalmudstone-wackestone (rare), intraclast-peloidpackstone-grainstone (common), regressive oolite(rare), unfossiliferous mudstone, and most (57%)are capped by fenestral grainy and muddy car-bonate (Figs 8C and 9A,B). Restricted lime mud-

stones and skeletal-peloid wackestone-packstone,are less common as parasequence caps (Fig. 9A).

The Upper Tithonian parasequence sets arevariably developed and are commonly 8–12 mthick, ranging from 5 to 20 m. They consist of threeto six parasequences per set (Figs 6C and 7C). Thinunits of the deeper subtidal facies occur in thelower parts of some sets, as do thicker ooliticcarbonates. Peritidal fenestral carbonates cap para-sequences in upper parts of some sets; in others,fenestral carbonates occur throughout the set.Some parasequence sets coincide with small risesand falls on the Fischer plots, but many do not.

INTERPRETATION

Climate

The Adriatic Platform during the Late Jurassicoccupied a low-latitude tropical region (below22�N; e.g. Stampfli & Mosar, 1999; Neugebaueret al., 2001), within the area of high evaporationas inferred from coeval evaporite deposits else-where and uniformitarian principles (Ziegleret al., 2003), and associated with the Pangeanmegamonsoon system (Parrish et al., 1982; Hal-lam, 1984). The low-latitude regions during theLate Jurassic were either desert or seasonallysummer or winter wet, and lacked a tropical ever-wet biome (Rees et al., 2004). A seasonal wet/dryclimate for the Adriatic Platform is supported bylack of arid climate indicators such as widespreadplatform-interior evaporites and lack of humidindicators such as coal and extensive bauxites.Evaporites in the platform succession are limitedto rare probable gypsum pseudomorphs in some

Fig. 8. Bar graphs showing range of stratigraphic dis-tances of facies below base of laminite cap or top ofrestricted lime mudstone, used as a datum to estimateminimum water depths of facies. In all the plots, thefacies are arranged from shallowest to deepest, but theyhave markedly overlapping depth ranges. (A) Plot ofdepth ranges of facies in subtidal parasequences inLower Tithonian using top of restricted mudstones asdatum, section LSL. (B) Plot of depth ranges of facies inperitidal parasequences in Lower Tithonian plottedbelow laminate caps, section MSV. (C) Plot of depthranges of facies at section LSL in peritidal, ooliticparasequences, using base of laminate caps as datum;transgressive oolites not included.

Fig. 9. Histograms showing: (A) frequency of faciesforming caps to parasequences, (B) number of ooidgrainstone-packstone occurrences that are transgressivevs. regressive.



of the subtidal facies, which could have formedfrom saline groundwaters during shallow restric-ted platform phases. Limited burrowing in thetidal flat facies is compatible with hypersalinetidal waters.

The global seawater Sr-isotope curve is markedby a pronounced minimum in the Callovian-Oxfordian followed by a transition to 87Sr-enriched values through the Late Jurassic andinto the Early Cretaceous (Jones et al., 1994;Grocke et al., 2003). This is consistent withincreasing weathering rates throughout the LateJurassic and Early Cretaceous (Weissert & Mohr,1996). An increased continental erosion andrunoff is further suggested by depleted Nd-isotope values of Tethyan sediments, with themost depleted values towards the end of theJurassic and in the Early and mid-Cretaceous(Stille & Chaudhuri, 1993). This eutrophicationcaused widespread carbonate platform growthcrises and repeated carbonate platform drowning(Hallock & Schlager, 1986; Masse, 1989; Weissert& Erba, 2004) under increasingly zonal climateconditions. The impact of accelerated weatheringand nutrition poisoning was far less dramatic forthe Late Jurassic carbonate platforms, which maybe a consequence of a monsoon-controlled rain-fall pattern with rainfall belts shifted to mid-latitudes (Weissert & Mohr, 1996).

Weissert (1989) has shown how the LateJurassic C-isotope curve shifts from relativelypositive d13C-values near +3Æ0& in the Kimme-ridgian-Early Tithonian to values near +1Æ30& inthe Late Tithonian-Berriasian. Recently, Grockeet al. (2003) reported a positive shift from theEarly to Middle Tithonian to the Tithonian-Berriasian boundary. The excessive burial oforganic carbon at times of high Tithonian carbon-ate accumulation rates (and times of high sea-level) was initiated by a greenhouse climateassociated with a monsoonal rainfall patterncontrolled by the disintegration of Pangea duringthe Late Jurassic (Sheridan, 1983; Ziegler, 1988;Parrish, 1993). Weissert & Mohr (1996) arguedthat the Middle and Late Tithonian drop inC-isotope values is possibly related to a reorgani-zation of the global climate system.

The Tethyan pelagic bulk O-isotope curve(Weissert & Erba, 2004) and palynological datafrom Northern Europe (Abbink et al., 2001) pointto cool Oxfordian and Early Kimmeridgian cli-mates, followed by a long-term warming trendlasting from the Kimmeridgian into the earliestCretaceous. The models and proxy records pre-sent conflicting evidence for warmer and cooler

modes. The long-term trends of atmospheric CO2

models (e.g. Geocarb III of Berner & Kothavala,2001) are relatively robust but they become moreuncertain at the finer scale (few tens of millions ofyears), presenting conflicts with climate proxydata (cf. Royer et al., 2004). These discrepanciesbetween pCO2 and palaeotemperatures (Veizeret al., 2000) may be due to a low-resolutionpalaeoclimate record masking the high-amplitudeclimate fluctuations (Weissert & Erba, 2004).Model results of a warmer Kimmeridgian-Titho-nian world with an elevated greenhouse effect ofMoore et al. (1991), show that the greatest warm-ing occurs over the higher-latitude oceans andleast over the equatorial and subtropical regions.It also indicates that the trade winds bring heavyrainfall to eastern Gondwana and the Tethys Seamargin, and a strong monsoon system overSoutheast Asia.

Sea surface temperatures probably increasedfrom the Middle into the Late Jurassic marking awarm phase in the overall cool mode of the MiddleJurassic to Early Cretaceous (Frakes et al., 1992).This Late Jurassic warm, relatively dry climate,probably promoted the development of hyper-salinity on the Adriatic Platform, and restrictionof the platform-interior. This hypersalinity pro-moted widespread ooid formation over wide areasof the platform-interior during shallow-waterphases (Husinec & Read, 2006), but was notsufficiently arid to generate evaporite units. Theintense dolomitization further inboard makes itdifficult to determine how far into the platform theoolitic facies extends (Husinec & Read, 2006).

Accommodation rates and parasequencedurations

Accommodation rates for the Tithonian were ca12–15 cm kyr)1 based on a duration of the inter-val of 5Æ3–6Æ5 Myr (Gradstein et al., 1994, 2004)and a thickness of up to 750 m. These arefar higher than the accommodation rates(1Æ4–3Æ3 cm kyr)1) of the slightly younger, Englishand Spanish Purbeck (Berriasian) sections des-cribed by Anderson (2004a,b). The Adriatic Plat-form underwent the highest subsidence rates inthe Tithonian compared with any time before orafter, even during the Late Cretaceous collisionalevent (Fig. 3). This rapid subsidence followedKimmeridgian platform differentiation and localdrowning within deeper water intraplatformtroughs. These high subsidence rates favouredthe platform record capturing most of the high-frequency sea-level oscillations. Approximately



320 parasequences developed on the platformduring the Tithonian which, given its duration(5Æ3–6Æ5 Myr; Gradstein et al., 1994, 2004), sug-gests that the parasequences have average dura-tions of 16–20 kyr. Given the uncertainty on theboundary ages of stages, this is close to theprobable duration of precession in the Jurassicwhich is only slightly less than 19 and 23 kyr(Berger et al., 1989). This supports a relativelycomplete record of precessional beats for theplatform, perhaps coupled with additional para-sequences being formed because of autocyclicprocesses (Wilkinson et al., 1997) or sub-Milan-kovitch sea-level changes (cf. Zuhlke et al., 2003).

Relative sea-level changes

The Fischer plots (Fischer, 1964; Read & Gold-hammer, 1988) for the Tithonian interval, LSLsection (Fig. 10) show a long-term relative sea-level rise and fall. Estimates of the duration arebased on the estimated 16–20 kyr age range of theparasequences making up the third-order se-quences, and could be erroneous if the estimatedparasequence duration is wrong. On this long-term rise and fall, the following third-ordersea-level cycles are superimposed: (1) a 1 Myrthird-order relative sea-level cycle starting at thebase of Tithonian composed of rise and fall up to

Fig. 10. Fischer plots of cumulativedeparture from mean cycle thick-ness versus stratigraphic position, ofsections LSL (and MSV, inset) plot-ted beneath stratigraphic column ofsection LSL. Parasequences andparasequence sets shown on strati-graphic column by upward widen-ing triangles. Third-order, relativesea-level cycles shown by numberson Fischer plots. Note that theinclined lines at the bases of thetriangles on the Fischer plots usingstratigraphic position (rather thancycle duration) no longer representsubsidence, as they did in Fischerplots originally. Subtidal para-sequences dominate sections on thelong-term rise, while peritidal andoolitic peritidal cycles dominate thelong-term fall. Facies legend is thesame as in Fig. 6.



parasequence 46, followed by (2) a shorter-term(<0Æ8 Myr) rise and fall up to parasequence 77.This is followed by (3) a ca 3 Myr relativesea-level cycle comprised of a rapid rise up toparasequence 105, a covered interval, then arelatively stable relative sea-level to parasequence178, and then a long-term fall to parasequence243. This is followed by (4) a ca 1Æ5 Myr relativesea-level cycle comprising a relatively low riseand a longer-term fall to the Tithonian-Berriasianboundary. In the basal Berriasian, this is mani-fested by the development of an emergencehorizon of clayey breccia. Thus the large-scalestacks of parasequences appear to be third-ordersequences. However, these are subtle in the field,not delimited by significant bounding discon-formities, or clear maximum flooding surfaces,and are indicated by the thickness trends of theparasequences. These trends relate to thicker-than-average parasequences in the transgressivesystems tracts and thinner-than-average para-sequences in the highstand tracts, which theFischer plots bring out. Such subtle third-ordersequence development is evident in the green-house Cambrian platform of the USA (Koerschner& Read, 1989; Montanez & Osleger, 1993).

The Fischer accommodation plot is similar tothe Arabian Plate Tithonian sea-level curve (Haq& Al-Qahtani, 2005) which is characterized by asmall relative sea-level cycle (our relative sea-level cycle 1, and perhaps 2) followed by a longer,larger, relative sea-level cycle (our cycle 3, andperhaps 4; Fig. 10). It is similar to the Haq et al.(1987) global curve which shows four sea-levelcycles, but differs from it in that the Fischer plotshows that cycle 3 is of longer term and higheramplitude than the others. When corrected forsediment and water loading (Read & Goldham-mer, 1988), any long-term (�5 Myr) sea-levelchange might have been <25 m, unless theFischer plots have grossly underestimated ampli-tudes. It might have been even less than this, ifthe long-term rise and fall on the Fischer plotwere related to increased, then decreased Titho-nian subsidence. These low amplitudes of sea-level change are compatible with relatively littleice and greenhouse conditions.

The approximate average duration of para-sequences (16–20 kyr) is strongly suggestive of aprecessional cyclicity driving high-frequency sea-level changes of relatively low magnitude.Although the facies stacking suggests bundlingof three to six parasequences into parasequencesets 10–15 m thick, such bundling of para-sequences into sets is poorly shown on the

Fischer plots. This suggests that short-term eccen-tricity (100 kyr) sea-level changes (or eccentricitymodulation of precessional sea-level changes;Anderson, 2004a) were relatively small (a fewmetres). Longer-term Fischer plot bundles at thescale of 40–60 m are very poorly developedsuggesting that the 400 kyr signal (long-termeccentricity, perhaps modulating precessionallydriven sea levels) was also very small. That theshort-term and long-term eccentricity signal is sopoorly represented in the stratigraphy is compat-ible with a greenhouse climate and a relativelyice-free Earth at this time. Dominant eccentricityforcing of sea levels appears to be typical of timesof significant ice buildup at the poles. At suchtimes, precessional sea-level changes modulatedby large eccentricity forcing, only infrequentlyflood the platform tops, thus few precessionalcycles are preserved on the peritidal platforms.This is because the platform fills to the highstandposition of one or two large precessional cycles,preventing smaller precessional cycles fromflooding the platform, thus resulting in missedprecessional beats (Goldhammer et al., 1990).

Formation of parasequences

The average duration of the parasequences sug-gests that they mainly formed by precessional sea-level changes. In the way of additional support,Strasser (1994) and Strasser et al. (2004) sugges-ted that parasequences (called elementary se-quences) in the Late Tithonian of the FrenchJura were also the product of precessional forcing,as were those in the Lower Cretaceous Berriasianof England (the sixth-order cycles of Anderson,2004a). The very shallow water facies on theAdriatic Platform and the scarcity of palaeosolformation at tops of parasequences suggest thatthe sea-level changes were of low amplitude. Thesubtidal parasequences typical of the LowerTithonian section formed during the overalllong-term sea-level rise associated with third-order sea-level cycles 1 and 2 (Fig. 10). Whencoupled with the high platform subsidence ratesin the Tithonian, this generated much accommo-dation (roughly 15–17 cm kyr)1) on the platform.Long-term platform sedimentation rates had to bein excess of long-term accommodation to keep theplatform in water depths of no more than a fewmetres. At a minimum sedimentation rate of20 cm kyr)1, a 2Æ7 m cycle (the average subtidalcycle thickness) could form in about 15 kyr. Thiswould result in little time being lost betweendeposition of successive subtidal cycles, which is



supported by the lack of tidal or supratidal facies(except in the more interior part of the platform;MSV section). The high rates of accommodationduring the long-term rise would also suppress theeffects of precessional relative sea-level falls, thusinhibiting the exposure of the subtidal facies (cf.Koerschner & Read, 1989; Goldhammer et al.,1990). Hence the shallowest water facies in LowerTithonian subtidal parasequences are the relat-ively sparse occurrences of barren lime mudstone(section LSL), and further platformward, tidal flatlaminite caps in upper parts of parasequence setsin section MSV.

On the falling limbs of the latter half of theFischer plot for the Upper Tithonian, sea-levelfall reduced accommodation by at least 50 m over2Æ5 Myr (excluding the duration of the short-termthird-order rises). This would have reduced totalaccommodation to between 9 and 12 cm kyr)1, a30% decrease in accommodation for the UpperTithonian section. The decreased accommodationduring deposition of each cycle caused smallerrelative sea-level rises, and enhanced relative sea-level falls compared with the transgressive EarlyTithonian. This decreased accommodation, cou-pled with the higher sedimentation rates of theshallower water settings, would cause more rapidshallowing to sea-level. Thus, an Upper Titho-nian parasequence (average thickness of 2 m)would shallow to sea-level in a minimum of10 kyr at a minimum sedimentation rate of20 cm kyr)1 leaving at least 5–10 kyr in which acycle would have been at sea-level, with depos-ition of intertidal microbial laminites culminatingin supratidal surfaces. Hence, the Upper Titho-nian shows a dominance of tidal flat-cappedcycles, compared with the Lower Tithonian, inwhich the tidal flat-capped cycles are located inthe more interior part of the platform. These wereonly able to prograde out towards the marginduring the long-term fall of the Upper Tithonian.

The abundance of oolitic peritidal cycles in theUpper Tithonian, along with the evidence forexposure and vadose diagenesis (Tisljar, 1979,1985) must reflect this marked decrease inaccommodation. Oolitic units are absent fromthe Lower Tithonian even though intermittentshallow-water conditions were present in themore interior parts of the platform. It cannot bejust shallow-water depths that limited the develop-ment of ooids to the Upper Tithonian section.Perhaps the increased, parasequence-scale flood-ing events in the Lower Tithonian resulted inopen-marine, near-normal salinity waters on theplatform which inhibited high supersaturation

states of the waters. Once the platform became thesite of predominantly very shallow-water depos-ition in the Upper Tithonian, the platform watersbecame more restricted (evident in the biotas) andhighly supersaturated with respect to calcite.This favoured precipitation of calcite ooidsaround skeletal and non-skeletal nuclei in moreplatform-interior, shallow-water settings (Husi-nec & Read, 2006).

Modelling and significance of parasequencesets

For modelling, we opted to have the precessionalsea-level signal ride on the obliquity and eccen-tricity signal. However, in a greenhouse setting itwould be just as reasonable to have precessionalsea-levels modulated by eccentricity, rather thanriding on the eccentricity signal. This does notmake a great deal of difference to the develop-ment of the platform top stratigraphy. However itwould have a significant effect if we weremodelling deeper subtidal facies, which is notthe concern of this paper. Given the long-termaccommodation rates of 12–15 cm kyr)1, theweak bundling of parasequences at the 10–15 mscale and at the 40–60 m scale is likely to havebeen a response to small, short-term (100 kyr) andlong-term (400 kyr) eccentricity sea-level changes(Fig. 11) or variation of precessionally driven sealevels within the short-term and long-term eccen-tricity bands. Short-term and long-term eccentri-city bundling also is evident in the Late Tithonianof the French Jura and the Early Berriasian ofEngland (Strasser, 1994; Anderson, 2004a). Thepoor development of those signals in the Titho-nian cyclic record supports the idea that therewas little ice at the poles at this time, as thestrength of the eccentricity signal in the LatePleistocene record is considered to be a responseto significant ice sheets (Raymo & Ruddiman,1992; Hinnov & Park, 1999). This supports theidea that the Tithonian was a time of globalgreenhouse conditions with little polar ice.Development of high-frequency sea-level cycleson such an ice-poor Earth has been ascribed to acombination of locking up of water in high-latitude mountain glaciers, draining and fillingof rift basin lakes, variation in aquifer waterlevels, and heating and cooling of the oceanicwater column (Jacobs & Sahagian, 1993).

The modelling (Fig. 11) supports the idea ofsmall sea-level changes driven by precessionalforcing with only small eccentricity signals oreccentricity modulation of precession. With



driving subsidence of about 4–6 cm kyr)1, maxi-mum platform sedimentation rates of 30 cm kyr)1

decreasing with depth, small precessional sea-level changes (5 m or so), along with a 100 kyrsignal of 5 m, and a very small (2 m) 400 kyrsignal, stacking patterns similar to those observedin the Tithonian are produced. Peritidal para-sequences are produced, with bundling of threeto five parasequences into parasequence sets.Increasing the magnitude of the precessionalsignal tends to cause periodic flooding of theplatform to greater depths than observed in thedata. Increasing the magnitude of the 100 kyreccentricity signal tends to cause periodic flood-ing of the platform to significantly greater depthsthan observed and, more importantly, causesincreased number of missing precessional beatsfrom the platform record (see Hardie & Shinn,1986; Goldhammer et al., 1990). Increasing themagnitude of the 400 kyr sea-level changes cau-ses numerous missed beats in the upper para-sequence sets of a 400 kyr bundle which is notobserved. The effect would be similar had weused a sea-level curve consisting of precessionalsea levels modulated by short-term and long-termeccentricity (cf. Anderson, 2004a,b).

The low-amplitude precessional signal, and theevidence for suppressed 100 and 400 kyr sea-level changes, all are compatible with greenhouseconditions. The small eccentricity signal also is

manifested by the weak (to absent) bundling ofparasequences into 100 and 400 kyr sets on theFischer plots. Such bundling, where present,results from thicker parasequences occurringlow in a set, and thinner parasequences higherin the set, which is evident in the model output(Fig. 11; cf. Goldhammer et al., 1990). Thisresults from the increased accommodation asso-ciated with the small eccentricity signal.

Thus in summary, the shallow character of thefacies within parasequences, the presence of tidalflat caps on parasequences in the platform-interiorthat progressively extend out towards the marginin the Upper Tithonian highstand, the apparentprecessional durations, the need for suppressedeccentricity sea-level changes, the poor develop-ment of eccentricity bundling on Fischer plots, andthe lack of significant emergence horizons (palaeo-sols), all point to the Tithonian as being a hot,greenhouse, relatively ice-free world within theoverall Middle Jurassic–Early Cretaceous ‘cool’mode of Frakes et al. (1992).

CONCLUSIONS

The exposed Upper Jurassic (Tithonian), carbon-ates of the Adriatic Platform, Croatia, have well-developed parasequences dominated by veryshallow water facies. These include dasyclad-

Fig. 11. Computer model run for 400 kyr duration showing synthetic stratigraphy generated by small sea-levelchanges typical of a greenhouse world with Milankovitch climate forcing and little ice (inset shows periods andmagnitudes of sea-level changes used in model run; the 19 kyr and the 100 kyr signals were asymmetric with a rapidrise and gradual fall). Maximum platform sedimentation rate was 30 cm kyr)1, peaking at 1 m water depth anddecreasing with water depth. Driving subsidence ranged from 5 cm kyr)1 on the inner platform to 7Æ5 cm kyr)1 in thebasin. The small 100 kyr signal generates the 10–15 m bundling. Having small 100 and 400 kyr sea-level changes iskey to having relatively complete preservation of precessional cycles, because with larger sea-level changes, there arenumerous missing beats. Having a larger 400 kyr signal results in numerous missed beats in the upper two bundles.



oncoid mudstone-wackestone-floatstone (shallowlagoon water depths, typically <4 m depth),skeletal-peloid wackestone-packstone (very shal-low lagoon), intraclast-peloid packstone andgrainstone (shoal), radial-ooid grainstone (hyper-saline shallow subtidal/intertidal shoals andponds), lime mudstone (restricted lagoon), fenes-tral carbonates and microbial laminites (tidal flat).

Long-term accommodation was high, 12–15 cm kyr)1. Accommodation (Fischer) plotsshow four third-order relative sea-level cyclessuperimposed on a 5–6 Myr long-term rise andfall. These sea-level cycles have some similaritieswith the published sea-level cycles of the ArabianPlate and the global curve.

The Tithonian parasequences are of roughly20 kyr average duration and were probablyformed by precessionally driven, small sea-levelchanges. During the long-term, 2Æ5 Myr sea-levelrise, predominantly subtidal cycles were depos-ited, whereas peritidal cycles developed on thelong-term (2Æ5 Myr) sea-level fall. Evidence of aneccentricity signal in the parasequence bundlingis relatively weak and produced successivelymore restricted facies upward in parasequencesets but little clear signal on the Fischer plots.Palaeosols are conspicuously absent from theTithonian, suggesting that the relative sea-levelfalls driven by precession were suppressed by thehigh accommodation rates. The parasequencestacking patterns strongly suggest that the Titho-nian was probably a time of global greenhouseconditions during the Middle Jurassic to EarlyCretaceous ‘cool’ mode.

ACKNOWLEDGEMENTS

Support for this work was provided by Fulbrightgrant no. 68428172 to A. Husinec and NSF grantEAR-0341753 to J.F. Read. We thank L. Fucek, I.Vlahovic, D. Palenik, T. Grgasovic, D. Maticec, andN. Ostric for assistance in field work, which wassupported by the Ministry of Science, Educationand Sport of Croatia. We also thank B. Sokac forhelp in calcareous algae determination.

Constructive reviews and suggestions by E.J.Anderson and A. Strasser, and the editorI. Montanez are acknowledged with thanks.

REFERENCES

Abbink, O., Targarona, J., Brinkhuis, H. and Visscher, H.(2001) Late Jurassic to earliest Cretaceous paleoclimatic

evolution of the southern North Sea. Global Planet. Change,

30, 231–256.

Anderson, E.J. (2004a) Facies patterns that define orbitally

forced third-, fourth-, and fifth-order sequences and sixth-

order cycles and their relationship to ostracod fauni-

cycles: the Purbeckian (Berriasian) of Dorset, England. In:

Cyclostratigraphy – An Essay of Approaches and Case

Histories (Eds B. D’Argenio, A.G. Fischer, I.S. Premoli

Silva, H. Weissert and V. Ferreri), SEPM Spec. Publ., 81,245–260.

Anderson, E.J. (2004b) The cyclic hierarchy of the ‘Purbeck-

ian’ Sierra del Pozo Section, Lower Cretaceous (Berriasian),

southern Spain. Sedimentology, 51, 455–477.

Bathurst, R.G.C. (1973) Carbonate Sediments and their Dia-

genesis. Elsevier, Amsterdam, 620 pp.

Berger, A., Loutre, M.F. and Dehant, V. (1989) Influence of the

changing lunar orbit on the astronomical frequencies of pre-

Quaternary insolation patterns. Paleoceanography, 4, 555–

564.

Berner, R.A. and Kothavala, Z. (2001) GEOCARB III: A revised

model of atmospheric CO2 over Phanerozoic time. Am.

J. Sci., 301, 182–204.

Bernier, P. (1984) Les formations carbonatees du Kimmerd-

giene et du Portlandiene dans le Jura meridional. Strati-

graphie, micropaleontologie, sedimentologie. Doc. Lab.

Geol. Fac. Sci. Lyon, 92, 803 pp.

Bosellini, A. (2002) Dinosaurs «re-write» the geodynamics of

the eastern Mediterranean and the paleogeography of the

Apulia Platform. Earth-Sci. Rev., 59, 211–234.

Boss, S. and Rasmussen, K.A. (1995) Misuse of Fischer plots

as sea-level curves. Geology, 23, 221–224.

Day, P.I. (1997) The Fischer diagram in the depth domain;

atool for sequence stratigraphy. J. Sed. Res., 67, 982–984.

De Castro, P. (1993) Observations on Campbelliella Radoicic,

1959 and Neoteutloporella Bassoullet et al., 1978 (green

algae, Dasycladales). In: Studies on Fossil Benthic Algae

(Eds F. Barattolo, P. De Castro and M. Parente), Boll. Soc.Paleont. Spec. Vol., 1, 121–184.

Fischer, A.G. (1964) The Lofer cyclothems of the Alpine Tri-

assic. Geol. Surv. Kansas Bull., 169, 107–149.

Frakes, L.A., Francis, J.E. and Syktus, J.I. (1992) ClimateModes of the Phanerozoic. The History of the Earth’s Cli-

mate over the Past 600 Million Years. Cambridge University

Press, Cambridge, 274 pp.

Goldhammer, R.K., Dunn, P.A. and Hardie, L. (1990) Depo-

sitional cycles, composite sea-level changes, cycle stacking

patterns, and the hierarchy of stratigraphic forcing: exam-

ples from Alpine Triassic platform carbonates. GSA Bull.,102, 535–562.

Gradstein, F.M., Agterberg, F.P., Ogg, J.G., Hardenbol, J., VanVeen, P., Thierry, J. and Huang, Z. (1994) A Mesozoic time

scale. J. Geophys. Res., 99, 24051–24074.

Gradstein, F.M., Ogg, J.G. and Smith, A.G. (2004) A Geologic

Time Scale 2004. Cambridge University Press, Cambridge,

610 pp.

Grocke, D.R., Price, G.D., Ruffell, A.H., Mutterlose, J. and

Baraboshkin, E. (2003) Isotopic evidence for Late Jurassic-

Early Cretaceous climate change. Palaeogeogr. Palaeo-

climatol. Palaeoecol., 202, 97–118.

Hallam, A. (1984) Continental humid and arid zones during

the Jurassic and Cretaceous. Palaeogeogr. Palaeoclimatol.

Palaeoecol., 47, 195–233.

Hallock, P. and Schlager, W. (1986) Nutrient excess and the

demise of coral reefs and carbonate platforms. Palaios, 1,389–398.



Haq, B.U. and Al-Qahtani, A.M. (2005) Phanerozoic cycles of

sea-level change on the Arabian Platform. GeoArabia, 10,127–160.

Haq, B.U., Hardenbol, J. and Vail, P.R. (1987) Chronology of

fluctuating sea levels since the Triassic. Science, 235, 1156–

1167.

Hardie, L.A. and Ginsburg, R.N. (1974) Layering: the origin

and environmental significance of lamination and thin

bedding. In: Sedimentation on the Modern Carbonate Tidal

Flats of Northwest Andros Island, Bahamas (Ed. L.A.

Hardie), Johns Hopkins Univ. Stud. Geol., 22, 50–123.

Hardie, L.A. and Shinn, E.A. (1986) Carbonate depositional

environments, modern and ancient. Part 3 – Tidal flats.

Q. J. Colorado Sch. Min., 80, 74 pp.

Hinnov, L.A. and Park, J. (1999) Strategies for assessing Early-

Middle (Pliensbachian-Aalenian) Jurassic cyclochrono-

logies. Phil. Trans. R. Soc., Lond., A, 357, 1831–1859.

Husinec, A. (2002) Mesozoic stratigraphy of the island of Mljet

within the geodynamic evolution of the southern part of theAdriatic carbonate platform. PhD Thesis, University of

Zagreb, Zagreb, Croatia, 300 pp.

Husinec, A. and Jelaska, V. (2006) Relative sea-level changes

recorded on an isolated platform: Tithonian to Cenomanian

succession, southern Croatia. J. Sed. Res., 76, in press.

Husinec, A. and Read, J.F. (2005) Facies stacking patterns in a

Late Jurassic Bahama-type platform interior, Adriatic Plat-

form, Croatia (abstract). In: AAPG 2005 Annual Convention,

Calgary, Alberta, Canada, 14, p. A65.

Husinec, A. and Read, J.F. (2006) Transgressive oversized

radial ooid facies in the Late Jurassic Adriatic Platform

interior: low-energy precipitates from highly supersaturated

hypersaline waters. GSA Bull., 118, 550–556.

Husinec, A. and Sokac, B. (2006) Early Cretaceous benthic

associations (foraminifera and calcareous algae) of a shallow

tropical-water platform environment (Mljet Island, southern

Croatia). Cret. Res., 27, 418–441.

Illing, L.V. (1954) Bahaman calcareous sands. AAPG Bull., 38,1–95.

Imbrie, J. and Purdy, E.G. (1962) Classification of modern

Bahaman carbonate sediments. In: Classification of

Carbonate Rocks (Ed. W.E. Ham), AAPG Mem., 1, 253–

272.

Jacobs, D.K. and Sahagian, D.L. (1993) Climate-induced fluc-

tuations in sea-level during non-glacial times. Nature, 361,710–712.

Jacquin, T., Dardeau, G., Durlet, C., de Graciansky, P.-C. and

Hantzpergue, P. (1998) The North Sea cycle: an overview of

2nd-order transgressive/regressive facies cycles in western

Europe. In: Mesozoic and Cenozoic Sequence Stratigraphy

of European Basins (Eds P.-C. de Graciansky, J. Hardenbol,

T. Jacquin and P.R. Vail), SEPM Spec. Publ., 60, 445–466.

Jelaska, V. (2003) Carbonate platforms of the External Dina-

rides. In: Evolution of Depositional Environments from the

Palaeozoic to the Quaternary in the Karst Dinarides and the

Pannonian Basin. 22nd IAS Meeting of Sedimentology,

Field-trip Guidebook (Eds I. Vlahovic and J. Tisljar), pp. 67–

71, Institute of Geology, Zagreb, Croatia, Zagreb.

Jones, C.E., Jenkyns, H.C., Coe, A.L. and Hesselbo, S.P. (1994)

Strontium isotopic variations in Jurassic and Cretaceous

seawater. Geochim. Cosmochim. Acta, 58, 3061–3074.

Koerschner, W.F. and Read, J.F. (1989) Field and modeling

studies of Cambrian carbonate cycles, Virginia Appalachi-

ans. J. Sed. Petrol., 59, 654–687.

Logan, B.W., Read, J.F., Hagan, G.M., Hoffman, P., Brown,R.G., Woods, P.J. and Gebelein, C.D. (1974) Evolution and

Diagenesis of Quaternary Carbonate Sequences, Shark Bay,

Western Australia. AAPG Mem., 22, 358 pp.

Loreau, J.P. and Purser, B.H. (1973) Distribution and ultra-

structure of Holocene ooids in the Persian Gulf. In: ThePersian Gulf: Holocene Carbonate Sedimentation and Dia-

genesis in a Shallow Epicontinental Sea (Ed. B.H. Purser),

pp. 279–328, Springer-Verlag, New York.

Masse, J.-P. (1989) Relations entre modifications biologiques

et phenomenes geologiques sur les plates-formes carbonat-

ees du domaine perimediterraneen au passage Bedoulien-

Gargasien. Geobios Mem. Spec., 11, 279–294.

Miller, K.G., Mountain, G.S., Browning, J.S., Kominz, M.,Sugarman, P.J., Christie-Blick, N., Katz, M.E. and Wright,J.D. (1998) Cenozoic global sea-level, sequences, and the

New Jersey transect: results from coastal plain and contin-

ental slope drilling. Rev. Geophys., 36, 569–601.

Montanez, I.P. and Osleger, D.A. (1993) Parasequence stacking

patterns, third order accommodation events, and sequence

stratigraphy of Middle to Upper Camrbrian platform car-

bonates, Bonanza King Formation, southern Great Basin. In:

Carbonate Sequence Stratigraphy: Recent Developments

and Applications (Eds R.G. Loucks and J.F. Sarg), AAPGMem., 57, 305–326.

Moore, G.T., Hayashida, D.N., Ross, C.A. and Jacobson, S.R.(1991) Paleoclimate of the Kimmeridgian/Tithonian (Late

Jurassic) world: I. Results using a general circulation model.

Palaeogeogr. Palaeoclimatol. Palaeoecol., 93, 113–150.

Neugebauer, J., Greiner, B. and Appel, E. (2001) Kinematics of

the Alpine-West Carpathian orogen and palaeoclimatic

implications. J. Geol. Soc. London, 158, 97–110.

Pamic, J., Gusic, I. and Jelaska, V. (1998) Geodynamic evo-

lution of the Central Dinarides. Tectonophysics, 297, 251–

268.

Parrish, J.T. (1993) Climate of the supercontinent Pangea.

J. Geol., 101, 215–233.

Parrish, J.T., Ziegler, A.M. and Scotese, C.R. (1982) Rainfall

patterns and the distribution of coals and evaporates in the

Mesozoic and Cenozoic. Palaeogeogr. Palaeoclimatol. Pal-

aeoecol., 40, 67–101.

Raymo, M.E. and Ruddiman, W.F. (1992) Tectonic forcing of

late Cenozoic climate. Nature, 359, 117–122.

Read, J.F. (1995) Overview of carbonate platform sequences,

cycle stratigraphy and reservoirs in greenhouse and icehouse

worlds. In: Milankovitch Sea-level Changes, Cycles and

Reservoirs on Carbonate Platforms in Greenhouse and Ice-house Worlds (Eds J.F. Read, C. Kerans, L.J. Weber, J.F. Sarg

and F.W. Wright), SEPM Short Course Notes, 35, 1–102.

Read, J.F. and Goldhammer, R.K. (1988) Use of Fischer plots

to define third-order sea-level curves in Ordovician peri-

tidal cyclic carbonates, Appalachians. Geology, 16, 895–

899.

Rees, P.M., Noto, C.R., Parrish, J.M. and Parrish, J.A. (2004)

Late Jurassic climates, vegetation, and dinosaur distribu-

tions. J. Geol., 112, 643–653.

Riding, R. (1992) Temporal variation in calcification in marine

cyanobacteria. J. Geol. Soc. London, 149, 979–989.

Royer, D.L., Berner, R.A., Montanez, I.P., Tabor, N.J. and

Beerling, D.J. (2004) CO2 as primary driver of Phanerozoic

climate. GSA Today, 14, 4–10.

Sadler, P.M., Osleger, D.A. and Montanez, I.P. (1993) On the

labeling, length, and objective basis of Fischer plots. J. Sed.

Res., 63, 360–368.

Sheridan, R.E. (1983) Phenomena of pulsation tectonics rela-

ted to the breakup of the Eastern North American contin-

ental margin. Tectonophysics, 94, 169–185.



Stampfli, G.M. and Mosar, J. (1999) The making and becoming

of Apulia. In: 3rd Workshop on Alpine Geological Studies

(Eds G. Gosso, F. Jadoul, M. Sella and M.I. Spalla), Sci. Geol.

Mem., 51, 141–154.

Stanley, S.M. and Hardie, L.A. (1998) Secular oscillations in

the carbonate mineralogy of reef-building and sediment-

producing organisms driven by tectonically forced shifts in

seawater chemistry. Palaeogeogr. Palaeoclimatol. Palaeo-ecol., 144, 3–19.

Stille, P. and Chaudhuri, S. (1993) Neodymium, strontium and

hydrogen isotope compositions of waters in present and

past oceans: a review. Lect. Notes Earth Sci., 43, 389–410.

Strasser, A. (1986) Ooids in Purbeck limestones (lowermost

Cretaceous) of the Swiss and French Jura. Sedimentology,

33, 711–727.

Strasser, A. (1994) Milankovitch cyclicity and high-resolution

sequence stratigraphy in lagoonal-peritidal carbonates (up-

per Tithonian-lower Berriasian, French Jura Mountains). In:

Orbital Forcing and Cyclic Sequences (Eds P.L. de Boer and

D.G. Smith), IAS Spec. Publ., 19, 285–301.

Strasser, A., Hillgartner, H. and Pasquier, J.-B. (2004) Cyclo-

stratigraphic timing of sedimentary processes: an example

from the Berriasian of the Swiss and French Jura Mountains.

In: Cyclystratigraphy: Approaches and Case Histories (Eds

B. D’Argenio, A.G. Fischer, I. Premoli Silva, H. Weissert and

V. Ferreri), SEPM Spec. Publ., 81, 135–151.

Tisljar, J. (1979) Fenestral and vadose early-diagenetic fabrics

of the Tithonian and Berriasian peritidal limestones near

Dubrovnik. Krs Jugosl. (Carsus Iugosl.), 10, 77–86 [in Croa-

tian with English summary]

Tisljar, J. (1985) Structural type and depositional environ-

ments of Jurassic coated grain limestones in the southern

Adriatic region. Krs Jugosl. (Carsus Iugosl.), 11, 71–99 [in

Croatian with English summary]

Tisljar, J., Vlahovic, I., Velic, I. and Sokac, B. (2002) Car-

bonate platform megafacies of the Jurassic and Cretaceous

deposits of the Karst Dinarides. Geol. Croat., 55, 139–170.

Van Wagoner, J.C., Posamentier, H.W., Mitchum, R.M., Vail,P.R., Sarg, J.F., Loutit, T.S. and Hardenbol, J. (1988) An

overview of the fundamentals of sequence stratigraphy and

key definitions. In: Sea Level Changes – An Integrated Ap-proach (Eds C.K. Wilgus, B.S. Hastings, C.G.St.C. Kendall,

H.W. Posamentier, C.A. Ross and J.C. Van Wagoner), SEPM

Spec. Publ., 42, 39–45.

Veizer, J., Godderis, Y. and Francois, L.M. (2000) Evidence for

decoupling of atmospheric CO2 and global climate during

the Phanerozoic eon. Nature, 408, 698–701.

Velic, I. (1977) Jurassic and Lower Cretaceous assemblage-

zones in Mt. Velika Kapela, central Croatia. Acta Geol., 9,15–37.

Velic, I. and Sokac, B. (1978) Biostratigraphic analysis of the

Jurassic and Lower Cretaceous in the wider region of Ogu-

lin, central Croatia. Geol. Vjesn., 30, 309–337 [In Croatian,

English summary].

Velic, I., Vlahovic, I. and Maticec, D. (2002) Depositional se-

quences and palaeogeography of the Adriatic Carbonate

Platform. Soc. Geol. Ital. Mem., 57, 141–151.

Vlahovic, I., Velic, I., Tisljar, J. and Maticec, D. (2005) Evo-

lution of the Adriatic Carbonate Platform: Palaeogeography,

main events and depositional dynamics: Palaeogeogr. Pal-

aeoclimatol. Palaeoecol., 220, 333–360.

Weissert, H. (1989) C-isotope stratigraphy, a monitor of

paleoenvironmental change; a case study from the Early

Cretaceous. Surv. Geophys., 10, 1–61.

Weissert, H. and Erba, E. (2004) Volcanism, CO2 and palaeo-

climate: a Late Jurassic-Early Cretaceous carbon and oxygen

isotope record. J. Geol. Soc. London, 161, 695–702.

Weissert, H. and Mohr, H. (1996) Late Jurassic climate and its

impact on carbon cycling. Palaeogeogr. Palaeoclimatol.

Palaeoecol., 122, 27–43.

Wilkinson, B.H., Drummond, C.N., Rothman, E.D. and Died-rich, N.W. (1997) Stratal order peritidal carbonate se-

quences. J. Sed. Res., 67, 1068–1078.

Ziegler, P.A. (1988) Evolution of the Arctic-North Atlantic and

the western Tethys. AAPG Mem., 43, 1–198.

Ziegler, A.M., Eshel, G., Rees, P.M., Rothfus, T.A., Rowley,D.B. and Sunderlin, D. (2003) Tracing the tropics across

land and sea: Permian to present. Lethaia, 36, 227–254.

Zuhlke, R., Bechstadt, T. and Mundil, R. (2003) Sub-Milan-

kovitch and Milankovitch forcing on a model Mesozoic

carbonate platform – the Latemar (Middle Triassic, Italy).

Terra Nova, 15, 69–80.

Manuscript received 1 July 2005; revision accepted 8August 2006



Documents

The Late Jurassic Tithonian, a greenhouse phase in the ...myslu.stlawu.edu/~ahusinec/Publications/1 - Papers... · The Late Jurassic Tithonian, a greenhouse phase in the Middle Jurassic–Early