Perturbation of a Tethyan coastal environment during the ... · calcareous taxa (Flaplophragmoides) and rare planktic foraminifera, indicating a slightly hypersaline eutro- phic inner

Palaeogeography, Palaeoclimatology, Palaeoecology 317-318 (2012) 66-92

ELSEVIER

C o n ten ts lis ts ava ilab le a t S c iV erse S c ienceD irec t

Palaeogeography, Palaeoclimatology, Palaeoecology

jo u r n a l h o m e p a g e : w w w . e l s e v ie r . c o m / lo c a t e / p a la e o

a;PALAEO S 3

©©$

Perturbation of a Tethyan coastal environment during the Paleocene-Eocene thermal maximum in Tunisia (Sidi Nasseur and Wadi Mezaz)

Peter Stassen a *, Christian Dupuis b, Etienne Steurbauta,c, Johan Yans d, Robert P. Speijer aa Department o f Earth and Environmental Sciences, K.U.Leuven, Celestijnenlaan 200E, B-3001 Leuven, Belgium b Faculté Polytechnique de Mons, Université de Mons, Rue de Houdain 9, B-7000 Mons, Belgium c Department o f Paleontology, Royal Belgian Institute o f Natural Sciences, Vautierstraat 29, B-1000, Brussels, Belgium ä Department o f Geology, FUNDP, UCL-Namur, Rue de Bruxelles 61, B-5000 Namur, Belgium

A R T I C L E I N F O A B S T R A C T

Article history:Received 11 April 2011Received in revised form 1 December 2011Accepted 12 December 2011Available online 21 December 2011

Keywords:Benthic foraminifera TunisiaPaleocene-Eocene thermal maximumPaleoenvironmentSea levelDysoxia

Despite the large num ber o f studies on the Paleocene-E ocene therm al m axim um (PETM), the know ledge of environm ental and biotic responses in shallow m arine environm ents rem ains quite poor. Benthic foraminiferal assem blages o f the Sidi Nasseur and W adi M ezaz sections in Tunisia w ere studied quantitatively and the paleoecologic interpretations provide n ew insights into the com plex relationship betw een PETM global w arm ing and perturbations o f sha llow m arine settings. These section s exp ose upper Paleocene to low er Eocene shales and m arls o f th e El Haria Form ation up to th e ph osph ate layers o f th e Chouabine Formation un derlying the El Garia lim eston es. The Sidi N asseur section contains a m ore com p lete and expand ed P aleocen e-E ocen e boundary interval com pared to W adi M ezaz, although being truncated at th e top. The W adi M ezaz section contains a m ore com p lete post-PETM interval. The studied interval can be subdivided into a seq uence o f 4 biofacies, rep resen ting respectively a la test P aleocene biofacies, tw o PETM biofacies and on e post-PETM E ocene biofacies.The la test Paleocene b iofacies 1 consists o f num erous calcareous benth ic foram inifera (e .g .A nom alino ides m idw ayensis, Frondicularia aff. phosphatica and various Bulim ina and Lenticulina sp ecies), abundant non - calcareous taxa (Flaplophragmoides) and rare planktic foram inifera, ind icating a slightly hypersaline eutrophic inner neritic to coastal environm ent, regularly interrupted by oxygen defic iency (m oderate dysoxia). During the la test Paleocene, th is h igh ly productive en v ironm ent sh a llow ed as ind icated by the increasing abundances o f A. m idw ayensis. The variable dom inance o f non-calcareous agglutinated taxa in biofacies 1 ind icates p ost-m ortem d isso lu tion effects. The TOC S13Corg record reveals a sharp negative excursion, m arking the base o f th e Eocene. In general, the absence o f litho log ie changes, an increasing sed im entation rate and absence o f rew orking indicate th at th e initial part o f the PETM is com p lete and exp an d ed in the Sidi N asseur section . A sharp faunal turnover co incides w ith th is negative S13Corg excu rsion and is characterized by th e d isappearance or d im inution o f com m on P aleocene taxa in th is area. During th e PETM, b en th ic foram inifera are less abundant and con sist o f opp ortunistic non -calcareous taxa togeth er w ith deeper d w ellin g (m idd le neritic) lagenids and bu lim inids (b iofacies 2 and 3). Planktic foram inifera, d om inated by flat-spired Acarin ina (m ain ly A. m u ltica m era ta ) , becom e m ore abundant, as observed in m any open m arine seq u en ces w orldw ide. All th ese faunal param eters su ggest m ore stressed probably severe dysox ic sea floor conditions w ith in a transgressive phase during th e o n se t o f the PETM. An estim ation o f th e total duration o f the Sidi N asseur PETM interval is difficult to establish , y e t th e lack o f recovery carbon iso top e values suggests that the preserved PETM interval reflects on ly a part o f th e CIE “core". The top o f th e PETM interval is truncated due to local (?) erosion during th e early Eocene. The Eocene recovery fauna is m ain ly com posed o f Lenticulina and Stain forth ia sp ecies (b iofacies 4 ) , ind icating restricted coastal to hyp osalin e lagoonal eutrophic conditions, d istin ctly different from earlier environm ental conditions.

© 2011 Elsevier B.V. All rights reserved.

* Corresponding author at: Biogeology Research Group, D epartm ent of Earth and Environmental Sciences, K.U. Leuven, Celestijnenlaan 200E, B-3001 Leuven, Belgium. Tel.: + 32 16 32 64 52; fax: +32 16 32 29 80.

E-mail addresses: [email protected] (P. Stassen), [email protected] (C. Dupuis), [email protected] (E. Steurbaut), [email protected] (J. Yans), [email protected] (R.P. Speijer).

0031-0182/$ - see front m atter © 2011 Elsevier B.V. All rights reserved, doi: 10.1016/j.palaeo.2011.12.011

1. Introduction

The Paleocene-Eocene therm al maximum (PETM), -55 .8 Ma ago, was a geologically brief (-1 7 0 kyr; Röhl et al., 2007) episode of globally elevated tem peratures, superimposed on the long-term late Paleocene and early Eocene w arming trend tha t culm inated in the highest ocean tem peratures of the Cenozoic (EECO, Early Eocene

http://www.elsevier.com/locate/palaeo

mailto:[email protected]





P. Stassen et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 317-318 (2012) 66-92 67

climatic optimum ; Kennett and Stott, 1991; Zachos e t al., 2001,2008). The PETM is characterized by a global 5 -8 °C warming of Earth's surface as well as the deep oceans and the onset of the PETM is marked by a worldw ide simultaneous negative excursion in ô13C values (C1E — carbon isotope excursion; Kennett and Stott, 1991 ). Although the ultim ate cause and trigger of the C1E is uncertain (for an overview see Sluijs et al„ 2007), the dissociation of m ethane hydrates along continental margins is a plausible hypothesis to account for the injection of large amounts of 13C-depleted carbon into the oceanic and atm ospheric reservoirs (Dickens et al.,1997). Simultaneously w ith the C1E, major global biotic changes are recorded, such as a major extinction of deep-sea benthic foraminifera (e.g. Tjalsma and Lohmann, 1983; Thomas, 1998), blooms of the tropical-subtropical planktic foraminiferal genus Acarinina (e.g. Kelly et al„ 1996), an acme of the dinoflagellate Apectodinium a t middle and high latitudes (e.g. Crouch et al„ 2001), distinctive assemblages of calcareous nannoplankton (e.g. Bralower, 2002), a turnover of larger foraminifera (e.g. Orue-Etxebarria et al„ 2001; Scheibner et al„ 2005), the disappearance of coral reefs (Scheibner and Speijer, 2008), changes in ostracod assemblages (e.g. Steineck and Thomas, 1996; Speijer and Morsi, 2002) and rapid dispersion of modern orders of mammals (e.g. Gingerich, 2006).

Although the geologically brief episode of global warming during the PETM has been intensively studied since its breakthrough in 1991 (Kennett and Stott, 1991), debates on the onset, total duration and recovery phases still continue until today. In order to establish the exact pathways of biotic responses to climate change during the PETM, detailed analyses of faunal and floral changes are needed from a wide spectrum of different environments, from the deep ocean, through shallow basins to terrestrial settings. Despite the increasing number of studies on the PETM in oceanic environments, the knowledge of marine processes and biotic responses in neritic environments remains quite meager. Sediments deposited along the borders of the Tethys Ocean, w hat was once an extensive east-w est subtropical seaway during the Paleogene, offers a wide variety of neritic environments (Tunisia: Auberi and Berggren, 1976; Egypt: Luger, 1985 and Speijer, 1994). High-resolution paleoecologic studies indicate that in the deeper (bathyal) marginal basins of the Tethys (Egypt, Israel, Turkmenistan) the benthic foraminiferal extinction event was equally severe as in the deep-sea (Speijer et al„ 1997). Faunal changes at shallower middle neritic w ater depths (Egypt; Speijer e t al„ 1996) appear less abrupt and most disappearances of taxa were tem porary since they reappeared after the PETM or are described in younger Eocene sediments elsewhere. The observed benthic foraminiferal distribution patterns are incorporated into a general repopulation model describing the collapse and recovery of the oligotrophic bathyal-outer neritic ecosystem after the onset of anoxia at the Tethyan sea floor. This anoxic phase triggered downslope migration of opportunistic taxa (e.g. Anomalinoides aegyptiacus) from shallower waters (Speijer et al„ 1997; Speijer and Wagner, 2002). It was argued tha t these faunal changes resulted from upwelling of low oxygen interm ediate Tethyan w ater into the Egyptian epicontinental basin leading to enhanced biotic productivity and basin- wide sea floor anoxia.

In order to unravel the nature and spatial distribution of these tem porary eutrophic and anoxic conditions in the southern Tethys, additional studies are needed. In northw est Tunisia, the clays and marls of the El Haria Formation are exposed in wide incised valleys and plains, creating excellent opportunities for high-resolution studies on early Paleogene climatic events in the southern part of the Tethys Ocean. The purpose of this study is to reconstruct the Tunisian paleoenvironmental evolution in a shallow marine setting w ith special attention to changes in the foraminiferal associations across the Paleocene-Eocene boundary. This will enable us to determ ine the biotic response to rapid global warming during the PETM in a marginal marine setting dominated by deposition of fine grained terrigenous

sediments (clay). Additionally, as coastal ecosystems are very sensitive to sea level changes, the proposed link betw een the onset of the PETM and rapid eustatic sea level rise can be tested (Speijer and Wagner, 2002; Sluijs e t al„ 2008).

2. Regional setting

The Maastrichtian to lower Ypresian El Elaria Formation overlies Campanian-Maastrichtian limestones of the Abiod Formation and is overlain by Ypresian phosphatic marls of the Chouabine Fm. and limestones of the El Garia Fm„ both belonging to the Metlaoui Group. The El Haria Fm. consists mainly of dark gray shales and marls w ith thin limestone intercalations. Phosphate beds are scattered throughout the formation and represent condensed intervals corresponding to periods of maximum sediment starvation (Saint-Marc, 1992). Lateral facies and thickness variations of the El Haria Fm. are thought to be structurally controlled along basem ent lineaments resulting in a number of small tectonically controlled basins, surrounding the large emerged zone of the Kasserine Island (Zai'er e t al„1998). The result is a latest Cretaceous-Paleocene paleogeography characterized by subsiding troughs in the north and northeast (NW Tunisian Trough and NE Tunisian Basin) and the Gafsa Gulf in the southwest (Auberi and Berggren, 1976; Zai'er et al„ 1998). During the Paleocene, the studied region was situated in the southern proximal part of the subsiding Tunisian Trough and in the vicinity of the emerged Kasserine Island (Fig. 1). Prolonged marine sedimentation took place in a shelf setting w ith high subsidence rate, high sediment input and w ith reduced sedim ent thickness tow ards the Kasserine Island (Bensalem, 2002).

The paleoecologic aspects of the El Haria Formation have been the subject of several studies because of its relative richness in microfossils and it offers expanded and nearly continuous stratigraphie records of the Paleocene (e.g. Auberi and Berggren, 1976). Outcrops near El Kef are well known and distribution patterns of ostracodes (Peypouquet e t al„ 1986), benthic foraminifera (Kouwenhoven et al„ 1997) and dinoflagellata (Guasti et al„ 2005) have been studied intensively. These studies indicate that the lower and middle Paleocene part of the El Haria Formation is transgressive (zones P2 and P3), extending over large parts of Tunisia and, in the Tunisian Trough, is followed by a general shallowing trend during the late Paleocene (zones P4 and P5; Auberi and Berggren, 1976; Kouwenhoven et al„ 1997; Guasti et al„ 2005). The sections in this study (Sidi Nasseur and Wadi Mezaz) form the top part of an expanded and well exposed sequence of the El Haria Formation in the Kalaat Senan region (50 km to the southwest of El Kef). Studying the Paleocene-Eocene boundary in this area is complicated because of the few good exposures and the occurrence of small faults. The Sidi Nasseur and Wadi Mezaz sections expose the most expanded and complete Paleocene-Eocene boundary sequences.

3. Material and methods

The Sidi Nasseur (N35°48'18.48" and E08°26'48.36") and Wadi Mezaz (N35°47'52.12" and E08°26'31.95") sections are located around the Sidi Nasseur hill, close to Kalaat Senan. The Sidi Nasseur section is well exposed from the lowermost phosphate beds up to whitish limestones w ith a phosphatic basal layer. The section term inates where the dip of the slope decreases and the surface is covered by debris from overlying Eocene limestones. The Sidi Nasseur section has been sampled twice, a low-resolution sampling in 1999 (Nas'99) and a new high-resolution sampling in 2006 (Nas'06) w ith a special focus on the Paleocene-Eocene boundary. Both sample sets are converted to comparable Sidi Nasseur heights and described as the NAS samples (Fig. 2). The nearby Wadi Mezaz section (900 m to the south of Sidi Nasseur) is exposed in two trenches, the Mzg'06 and Mzh'06 sections, both sampled in 2006. The base of the Mzg'06

P. Stassen e t al. / Palaeogeography, Palaeoclimatology, Palaeoecology 317-318 (2012) 66-92

Tunis

Northern Basin

Sidi Nasseur & Wadi Mezaz

\ * Elies* I

KasserineIsland Southern

Basin/

■Jeffara i Island

[T j emerged zone

0 scale (kmj-jQQGafsa Gulf

I l emerqed zone

I I co asia I marine

I lepen marine

Fig. 1. A: Paleogeographic reconstruction of the Tethyan Ocean during the late Maastrichtian (modified after Dercourt e t a l, 2000). B: Detailed paleogeographic reconstruction of Tunisia during the deposition of the El Haria Formation (upper Maastrichtian to lower Eocene, modified after Zai'er e t al., 1998 and Saint-Marc and Berggren, 1988 after Burollet, 1967).

section is situated above a presumed fault as indicated by an accumulation of calcite veins and therefore sampling was started 2 m above this zone to ensure that the sediments were not distorted. The top of the Mzg'06 section corresponds to the top of a thick phosphate bed. The Mzh'06 section is only a few meters separated from Mzg'06 and the lowermost samples are taken just above the top of the thick phosphate bed. The Mzh'06 section term inates where the limestones are covered by debris. The composite section MEZ contains all Wadi Mezaz samples which are converted to comparable Wadi Mezaz heights and described as the MEZ samples (Fig. 2). Some samples (e.g. in the phosphate beds) are barren or contain only unidentifiable microfossils.

Calcareous nannofossils provide the main biostratigraphic framework. Planktic foraminifera are generally rare and provide limited additional stratigraphie refinem ent or represent reworked Cretaceous species. In order to identify the controlling environmental param eters, we determ ined foraminiferal numbers, percentages of planktic, calcareous and non-calcareous benthic foraminifera, the proportions of endo- and epibenthic morphotypes, faunal diversity changes, benthic foraminiferal abundance patterns and paleoecologic preferences of the most common species (m ore than 5% in at least one sample).

Foraminiferal residues were obtained following conventional procedures. About 80 g of sedim ent w ere dried at 50-60 °C and afterw ards soaked in a Na2C03 solution. After disintegration, the sam ples w ere washed over a 63 pm sieve and dried at 50-60 °C; this treatm ent was repeated w henever the washed residues rem ained som ew hat aggregated. After complete disaggregation, the dried residues were sieved into three size fractions (63-125 pm, 125-630 pm, >630 pm). A representative split for quantitative analysis (approximately 250-300 benthic specimens) was obtained from the 125-630 pm fraction using a microsplitter. From these splits, all benthic specimens were picked, identified, counted and perm anently stored in micropaleontologic slides (stored at the D epartm ent of Earth and Environmental Sciences, K.U. Leuven).

Benthic foraminifera were identified to species or genus level. The latter applies to Lenticulina spp„ other nodosariids and most agglutinated taxa. The lagenids are separated into five morphogroups based upon the ornam entation of the test. The preservation of the non-calcareous agglutinated taxa is rather poor, limiting detailed taxonomic assignments. These taxa are primarily distinguished from each other by large differences in overall morphology or wall composition.

Therefore rare Trochammina species are lumped together with Haplophragmoides species (mostly H. excavata/walteri). Consequently, this unavoidable lumping of non-calcareous taxa does not allow detailed insight into diversity patterns. To obtain representative assemblage patterns, non-calcareous agglutinated benthic foraminifera (agBF) and calcareous benthic foraminifera (calcBF) w ere generally picked from different splits, in order to increase their counted numbers and thus their representativeness. Calcareous and non-calcareous benthic assemblages are thus treated independently. Benthic foraminiferal num bers (num bers per gram sedim ent) w ere calculated for the 125-630 pm fraction. Diversity indices were calculated using initial data sets containing all encountered benthic species (PAST- software, Elammer et al„ 2001). Relative abundances are expressed as the proportion (percentage) of a species in the entire assemblage and the m ost common taxa are displayed, all rem aining species are lum ped in a rest group. To analyze the relative abundance data of Sidi Nasseur and Wadi Mezaz statistically (PAST-software, Hammer et al., 2001), detrended correspondence analysis was performed on the data set o f the common calcareous benthic foraminifera.

We compiled (semi-)quantitative data of other published Tethyan sections to generate paleobathymetric distribution patterns and paleodepth preferences of the m ost common taxa. A category reflecting the m ost preferred w ater depth (depth w ith highest abundances w ith an upper limit of middle neritic, fitting the shallow nature of the area) is assigned to each taxon. The paleodepth calculations are based on the relative abundances of each calcareous species: Paleodepth = 1 / ntota£ (n¡* (depthmin ¡ + dep thmaxi) /2 . In which n¡ represents the percentage of species i, minimal and maximal depth of a species i refers to the upper and lower limits of preferred w ater depth. Species w ith an unknow n depth preference or a wide range of occurrences are excluded from the calculations.

The ô13Corg record is based on powdered samples which were treated w ith 0.9 N hydrochloric acid to remove carbonates. Quantities required for analysis (betw een 5.5 and 16.5 mg) w ere calculated on the basis o f the TOC values (standard LECO carbon analyzer — CS- 200). Each sample was weighed into tin capsules and rolled into balls for continuous flow combustion ( 1025 °C) and isotope analysis using a Carlo Erba EA1110 elem ental analyzer coupled to a mass spectrom eter (ThermoFinnigan delta plus XP). Results are calibrated w ith the inter-laboratory standard sucrose 1AEA-CH and the overall precision of analyses is w ithin 0.2%» (1er).

P. Stassen e t al. / Palaeogeography, Palaeoclimatology, Palaeoecology 317-318 (2012) 66-92 69

1 Phosphatic marl

E Z 3 M arl A N a s ’9 9

Shale A N a s ’0 6

P .y / I Coprolites A M zh '0 6

[ ,11 Bioturbated shale ■ M zg '0 6

1 " Lim estone

Sidi NasseurNas'99 & Nas’06

7 7 7 T T 7/ B a r r e n / ■ ' •

u n c o n f o rm i ty '1

10.0 m HO F. aff. phosphatic a

unconform ity

Wadi MezazMzg'06 & Mzh’06

F ig . 2 . Litho- and biostratigraphy of the El Haria Formation at Sidi Nasseur (NAS) and Wadi Mezaz (MEZ) based on planktic foraminiferal and calcareous nannofossil biostratigraphy (Martini, 1971; Aubry, 1999; Berggren and Pearson, 2005). The PETM interval is highlighted by the gray segment and the onset is indicated by the highest occurrence (HO) of Frondicularia aff. phosphatica and lowest occurrence (LO) o f Discoaster araneus. Correlation between the sections suggests that a substantial package o f lower Eocene sediments is missing below the uppermost limestone interval at Sidi Nasseur.

4. Stratigraphy

The studied sections comprise the top part of El Haria Formation, grading into the more calcareous Chouabine Formation (Fig. 2). The lower part of the studied Sidi Nasseur section consists of upperm ost Paleocene shales of the El Haria Fm. These shales, from base to NAS 10.0 m, contain fairly well preserved, moderately diversified nannofossil associations and span the top of calcareous nannofossil Subzone

NP9a. Planktic foraminifera are nearly absent in the upper Paleocene sediments of the Sidi Nasseur section. The rare presence of Morozovella velascoensis and the associated calcareous nannofossil NP9a assemblage, suggests that the studied interval comprises the top part of Subzone P4c or Zone P5 (Stassen e t al., 2009).

The nannofossil associations from the upper shale unit (El Haria Fm.) of the Sidi Nasseur section, from NAS 10.1 m upward, are rather poor in species and number of specimens. Reworked Cretaceous taxa

70 P. Stassen et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 317-318 (2012) 66-92

and Coccolithus pelagicus/C. subpertusus are dominating. Typical PETM- taxa are present in the lowest part of these shales, although very rare: a few specimens of Discoaster araneus in sample NAS 10.1, 10.5 and 10.85 m, but there is no record of D. anartios. Assemblages of this unit refer to Subzone NP9b. Planktic foraminifera become more frequent from NAS 10.1 m onw ards and consist to a large extent of Acarinina species, including A. sibaiyaensis and A. multicamerata. This assem blage resembles the well-known planktic excursion assemblage of the PETM tha t is recognized w orldwide (El Zone; Kelly e t al., 1996; Berggren and Pearson, 2005; Guasti and Speijer, 2008). The uppermost marly interval, ranging from NAS 13.5 to 17.6 m, contains very impoverished nannofossil associations and recrystallized planktic foraminifera, probably due to weathering, leading to the preservation of only the most robust taxa. The composition of the nannofossil and planktic foraminiferal associations, although taphonomically biased, seems to indicate the continuation of Subzone NP9b and E l. The uppermost limestone beds with a barren phosphatic basal layer (El Garia Fm., from NAS 18.1 to 19.05 m) yields a poorly preserved, but fairly diversified nannofossil association and point to the middle to upper part of Martini's (1971) nannofossil Zone NP11. The NP10 interval is missing in the Sidi Nasseur section.

The W adi Mezaz section can be subdivided into similar units and the lower part is attributed to NP9b and El interval. The main Wadi Mezaz phosphate bed, ranging from MEZ 8.5 to 12.15 m, is almost completely devoid of calcareous nannofossils and foraminifera, except for MEZ 12.0 m. The assemblages, from MEZ 12.0 to 15.0 m,

are marked by a substantial decrease in reworked Cretaceous material and by a temporary increase in C. pelagicus. The presence o f Tribrachiatus bramlettei and the absence of T. orthostylus in the uppermost marly part (Chouabine Fm.) of the section allow an attribution to NP10. The NP10-NP11 boundary is situated near the phosphate level at MEZ19.5 m on the basis of lithostratigraphic correlation with another nearby section (MZD, unpublished). Planktic foraminifera become less frequent above the base of the main Wadi Mezaz phosphate bed and represent mostly reworked Cretaceous taxa, providing no extra stratigraphie information.

According to the overall composition of the calcareous nannofossil associations, the Paleocene-Eocene boundary is pinpointed between NAS 10.0 m (NP9a) and 10.1 m (NP9b). This level corresponds to the abrupt negative shift in ô13Corg values (Fig. 3), indicative of the start of the C1E and by definition the onset of the PETM (Dupuis et al., 2003). The PETM interval in the Sidi Nasseur section corresponds with the overlying shales and marls up to the base of the upperm ost phosphate bed (Fig. 2). Correlation with the nearby Wadi Mezaz section suggests a substantial stratigraphie gap across this phosphate bed. The lower part of the Wadi Mezaz section correlates with the PETM interval in the Sidi Nasseur section and is truncated at the top by the thick phosphate bed (Fig. 2). The overlying marls and phosphate beds are not encountered in the Sidi Nasseur section. Although the Wadi Mezaz section is also truncated by erosion at the top of the PETM interval, it contains the best record of lower Eocene sediments and thus provides insights into post-PETM environments. A composite sequence is

Sidi Nasseur (NAS)SCûrg (%o) TOC (%) CaC03 (%)

B a rren

15-

B arren (decalc ified)

-29 -27 -25 0,0 100

Fig. 3. The stable carbon isotope record of bulk organic carbon (013Corg) displays constant values in the uppermost Paleocene interval and the onset o f a negative isotope shift is pinpointed between NAS 10.0 and 10.1 m. There is no carbon isotope recovery trend in the Sidi Nasseur section.


formed by the lower to middle part of the Sidi Nasseur sequence and is extended by the middle to upper part of the Wadi Mezaz sequence.

5. Results

Distribution data of calcareous benthic foraminifera throughout the Sidi Nasseur and Wadi Mezaz sections (Figs. 4 and 5) allow the subdivision of the samples into distinct foraminiferal biofacies using detrended correspondence analysis (Fig. 6). The samples and species are clustered in three groups separating the four biofacies (biofacies 2 and 3 are closely related). Axis 1 (explaining 24.9% of the variance) essentially separates the PETM biofacies 2 and 3 from biofacies 1 and 4. Axis 2 (explaining 18.6% of the variance) separates biofacies 4 from the others. These four biofacies form the framework of the paleoenvironm ental interpretations and are supported by the overall distributional changes of non-calcareous agglutinated foraminifera and planktic foraminifera.

5.1. Subdivision in biofacies

Biofacies 1 is encountered in the lower part of the Sidi Nasseur section, from the base up to NAS 10.0 m and represents the latest Paleocene environm ent in the area. This biofacies is characterized by abundant Anomalinoides midwayensis, Lenticulina spp. and other associated characteristic Paleocene benthic foraminifera such as Frondicularia aff. phosphatica. Biofacies 2 (PETM) is encountered in the middle part o f the Sidi Nasseur section, from NAS 10.1 to13.5 m and is defined a t the base by the disappearance or dim inution of many latest Paleocene taxa and by comm on occurrences of taxa tha t are largely restricted to the PETM interval (e.g. Reophax species and lagenids). Biofacies 2 is also found in the Wadi Mezaz section, from MEZ 4.2 to 6.75 m. The transition from biofacies 1 to 2 was not sampled in the Wadi Mezaz section. Biofacies 3 (PETM) is encountered in the upper part of the Sidi Nasseur section (13.5 to17.5 m) beneath the barren phosphate layer. It is defined at the base by the decrease in abundance of the lagenids and Nonionella insecta and the increase of Neoeponides elevatus and Cibicidoides succedens. Reophax disappears below this level. Biofacies 3 is also found in the Wadi Mezaz section ( 6.75 to 8.5 m ), bu t is only represented by a few samples and the transition from biofacies 2 to 3 is less clear. As biofacies 2 and 3 are best developed in the Sidi Nasseur section, these are included in the composite sequence and the equivalent part of the W adi Mezaz section is not further considered. Biofacies 4 (lower Eocene) occurs in the upper part o f the Wadi Mezaz section and is characterized by the absence of lagenids, a strong increase in Stainforthia and reappearance of several biofacies 1 species (e.g. Lenticulina midwayensis). The main phosphate bed is mostly barren, bu t sample MEZ 12.0 m contains a foraminiferal assemblage attrib utable to biofacies 4.

5.2. Uppermost Paleocene biofacies 1

The encountered calcareous benthic foraminiferal assemblages are m oderately diverse, mostly dom inated by a few species (5 to 20 taxa per sample, average of 15, Figs. 7 and 8). The most common taxa tha t cause nearly all changes in the foraminiferal abundance patterns (Fig. 4, Plates 1 and 2) are Lenticulina spp. (diverse lenticulinid fauna), Anomalinoides midwayensis and buliminids (B. ovata and B. gr. trigonalis). Other species such as Anomalinoides umboniferus, Cibicidoides praecursorius, Valvulineria scrobiculata and Neoeponides elevatus occur with maximum abundances of 15%. Stainforthia and less common Anomalinoides, Cibiadoides and Gyroidinoides species make up the remaining part. Lenticulinid abundance gradually decreases from 64% (NAS 2.75 m) to 20% (NAS 7 m) and remains stable around 10-20% upwards (NAS 7.3 m-10.0 m) where Anomalinoides midwayensis becomes the dominant species (up to 78%). The buliminids (incl. Stainforthia) are

generally rare, but common to abundant ( 10 to 60%) in levels with lowered abundances of A. midwayensis. The last common occurrences of several characteristic Paleocene species (A. midwayensis, A. umboniferus, Pyramidulina spp.,F. aff. phosphatica) are at NAS 10.0 m. Non-calcareous agglutinated taxa are sometimes very frequent (agBF, Fig. 8). The most common group consists of Haplophragmoides species (mostly H. excavata/walteri), strongly dominating the Paleocene assemblage (average 86% of the agBF). Ammobaculites (A. expansus and A. midwayensis) is the only o ther common Paleocene com ponent w ith peak abundance (65.3%) near the top of the Paleocene.

Benthic foraminiferal numbers of calcareous taxa fluctuate widely and rapidly (0 to 800 specimens/g, average of -225 specimens/g) throughout the section (Fig. 7). Foraminiferal numbers of non- calcareous taxa are more stable (mean o f-100 specimens/g) with multiple peaks (>260 specimens/g). Lower numbers of calcareous benthic foraminifera correspond to intervals with very high abundance of non-calcareous agglutinated foraminifera (up to 100%). In contrast, absolute numbers of non-calcareous agglutinated foraminifera do not follow their relative abundance fluctuations. Planktic foraminifera are extremely rare (average <3/g) with most samples containing no planktic foraminifera and sporadic occurrences of large specimens in the phosphate layers.

5.3. PETM biofacies 2

The calcareous foraminiferal fauna is more diverse than in biofacies 1 (16 to 26 taxa per sample, average of 22 taxa) and shows a more equal distribution between the species (Fig. 8). The most common taxa are lagenids, buliminids (B. ovata, B. gr. trigonalis, B. aff. strobila and B. midwayensis), Nonionella insecta and Cibicidoides megaloperforatus (Plate 3). Species abundances are in general fairly stable (up to 20%) with peaks of some lagenids, Nonionella insecta and Cibiadoides megaloperforatus (up to 44%). Other species like Baggatella cf. coloradoensis, Gyroidinoides girardanus and Valvulineria? insueta occur with maximum abundance of 10%. Neoeponides elevatus, Stainforthia and less common Cibiadoides species are the remaining calcareous taxa. The assemblage is also characterized by the complete absence of relatively large benthic species (lenticulinids, nodosariids) and most specimens are smaller than 300 pm (visual observations). The non-calcareous agglutinated foraminiferal assemblage is also more diverse than in biofacies 1 and 5 taxa (morphotypes) are distinguished with an equal distribution between the species. The abundance of Haplophragmoides decreases upward. Reophax sp. 1 has a common and stable occurrence between NAS 10.1-12.7 m. Ammobaculites species are more common upwards and become dominant after the disappearance of Reophax sp. 1.

Calcareous benthic foraminiferal numbers drop to very low values, ranging from barely 1.5 to 20 specimens/g (average 6/g). These low numbers are associated with very high abundances of non-calcareous agglutinated foraminifera (up to almost 100% of the total benthic assemblage) but foraminiferal numbers of non-calcareous taxa also decrease to an average of 20 specimens/g, much lower than in the uppermost Paleocene biofacies 1 (Fig. 7). Planktic foraminifera are common in this biofacies. All samples contain planktic foraminifera with an average of 16 planktic foraminifera/g and an increase in abundance from 5 to 45%. The planktic assemblages consist to a large extent of acar- ininids and only a few reworked Cretaceous planktic foraminifera are encountered. Also, a striking change in microfossil content is the sudden appearance of numerous spines of irregular sea urchins (average of 11 spines/g).

5.4. PETM biofacies 3

The calcareous benthic foraminiferal assemblage of biofacies 3 is as diverse as in biofacies 2 (16 to 24 taxa per sample, average of 21 taxa), but with a less equal distribution between the taxa (Fig. 8, Plate 3). The most common taxa with stable occurrences are Neoeponides elevatus,

«3.C

©CL.2CLCOL L

a g B F.CO00c:

ISP

o>

C L(A©C

§-«D

H?©

X s 2 ©

I©OC

I03

©-QO

3o©-oo

■cII cë "ï £

1 1 1 1 1 1 1 1 ............... l l l l l l l l l l l l l l . i l

.s

.00©c

1 1a>O «- -X o> © © c .c

.g .g .goco

3 3 QQ GQL L i J L

F

3.O

1-3

I i I i

S-23,o-5

§ <A.©C©

.C

cI*©

. ;p ©

©

©© .C

S J U CÜ

1 c 'S IB2 Q*ç

. L ü L L L

©2 o c;

*=C i I i I i_Ll

F" - i w

c a lc B F

i l& i”• © ©

^ 330 o 3 . c Ç -'© ©1 s2 °

LjJ L

o© o© O CL<o E to© IA JZ.'S fcj LJ3 OïO® .W s«< -J -J -J

Ll±_ LuL L U i i

D ) O ) CD© © ©C C C

I §,

© © S s 3 ê'S — «w -gi- (=- 03

0,© © c 3 3 Ä& & §0 )0 )© © 3

- J ~ J QQJ L L LU

— r

.0000 3

1 ©

S

■c -2© OÔL» C-€ IK ■§t g© -2CD O©S <0 ©

~CS

.orQO

1111

I I

©

3c"o 422 §a > gW *■* © C-

T j .©ô.3 ©

jë :§=§S II© ,-X ©0 Q « 5 L d L L

)

g*7- S2 ©co ^ "O!=§ s ^ l 8o © © (A 05 10 ¿5 co.©.©.©.© <g d) -c-3-3-5"C3 ^i t ï t o o^ > 0 , 0 , 0 3 ^ . 3 . 3 . 3 . 3 . 0 .O ©45 © .© .-O ;-Q C0CQC0C0O Ö L L L L L L '

sJ j j j

F

decalcified

r F T n - r p T n r r r r 1 I M T f 11T M r ï ï l F P T H I“ P I H T T T P i r m P T T T H m r

n i n ' l 1111 0 50 100

■ h t r r m r r m m h t m r m r r r r r r r m r m

Fig. 4. Benthic foraminiferal abundance patterns of the most common taxa in the Sidi Nasseur section (>5% in at least one sample). The division into biofacies 1-3 is based upon distinct differences in benthic foraminiferal composition and is supported by correspondence analysis. Note that calcareous benthic foraminifera (calcBF) and non-calcareous benthic foraminifera (agBF) are treated separately and each group represents a 100% composition.

Barren

P. Stassen et al. / Palaeogeography, Palaeoclim

atology, Palaeoecology 317-318

(2012) 66-92

LU

1 9 - ■/,

10 -

«Æ £55 c S'r- — CU

i i ii i i <u ¡, o af e p gTJ V) V)-2 03 co ^ °

10.<0 <US?ü> <0py

w (5 0 0 S ) Q n 5 ÇJ (D <L>

,-g g g Q.o C E <13

m cu<0 CD g gQ) G)05 O) 0)0)

eT;2.£o.3 a QQ CQ

(o (DQQ GQ QQ

L L L L LU- l i l i :

[ > i M ► »barren o r un identifiab leI 1 II I I M MI I I II

barren

► ► M

m-n r n 11 m r m 111 n m n5 0 1 0 0

r r r r r niTi m m r r n r r rnrr m hth r r mr i m 111111 m-n r m i m i r 1111 m 11111 r m r rm pprn r r r r r r r r r r rm m m r r nr r m r m n m r r m rm m ht r r n r m

Fig.5. Benthic foraminiferal abundance patterns of the most common taxa in the Wadi Mezaz section (> 5% in at least one sample). The division into biofacies 2 -4 is based upon distinct differences in benthic foraminiferal composition and is supported by correspondence analysis. Note that calcareous benthic foraminifera (calcBF) and non-calcareous benthic foraminifera (agBF) are treated separately and each group represents a 100% composition.

u>

P. Stassen et al. / Palaeogeography, Palaeoclim

atology, Palaeoecology 317-318

(2012) 66-92


nj>

■

aeCD ËCO "njm

</) +

X<

-1

Eocene Chouabine Fm.

lagoonal to restricted • 0

✓

c o a s t a l

Lenticulina spp. #

.. •

V, scrobiculata'

■ ,

'o .

■p.

Paleocene E! Haria Fm.

coastal

o' o^ ■ ■ ^9, f polignaci

OAno. midwayensis A

A. umboniferus #

o ’• •

•Ár «

/ V4

b iofacies

1

’ Eocene (PETM) 1 El Haria Fm. 1

m idd le-inner neritic 1 1------------

3infaunal

+ te s t size Axis 1 (24.9 % var.)

Fig. 6. Two-dimensional ordination of the samples (detrended correspondence analysis, based upon the relative calcareous benthic foraminiferal abundances) confirms the subdivision into separated biofacies as indicated by non-overlapping convex hulls. Note that the scores of all samples are plotted together w ith the ordination of the most common benthic foraminifera (> 10% in a t least one sample).

Cibiadoides succedens, lagenids and buliminids (B. ovata, B. gr. trigonalis and B. aff. strobila). Other species such as Baggatella cf. coloradoensis and Cibicidoides megaloperforatus occur sporadically in higher abundances. The remaining taxa (“rest") in biofacies 3 consist mostly of poorly preserved trochospiral taxa. Non-calcareous agglutinated taxa become less frequent and Ammobaculites is dominant, only interrupted by a peak of Haplophragmoides at NAS 15.0 m.

Foraminiferal numbers of calcareous taxa remain low, ranging from 1 up to 53 specimens/g w ith an average around 20 (Fig. 7). These low numbers are associated w ith reduced abundance of non- calcareous agglutinated foraminifera. All samples contain planktic foraminifera (average of 35/g), mostly consisting of Acarinina, w ith an upwards increase in numbers and abundance from 45 to 86%. Shell fragments and small spines of irregular sea urchins occur in large numbers (average of 35 spines/g).

5.5. lower Eocene biofacies 4

The calcareous benthic foraminiferal assemblage of biofacies 4 is less diverse than PETM biofacies 2 and 3 and similarly diverse as biofacies 1 (11 to 17 taxa per sample) w ith variable abundances (Fig. 8) and the most common taxa are Lenticulina spp. (a low diverse lenticulinid fauna, mostly L. midwayensis, Plate 4), Valvulineria scrobiculata, Cibicidoides succedens and Stainforthia species (S. kamali, S. gafsensis and Stainforthia? sp. 1). Other species such as Bulimina occur sporadically in higher numbers. Non-calcareous agglutinated foraminifera are not frequent and Ammobaculites is the main component, interrupted by several peaks of Haplophragmoides.

Foraminiferal numbers of calcareous taxa (2 to 425 specimens/g, average of 85) are lower than in biofacies 1, but much higher than in biofacies 2 and 3. This increase in calcareous specimens corresponds w ith continuously low numbers of non-calcareous specimens (<20 specimens/g, average of 3) and reduced abundance of agglutinated taxa (Fig. 7). Planktic foraminifera are less frequent than in biofacies 3, generally decreasing upwards in numbers (average of 5/g) and abundance from 20% to less than 5% w ith several abundance peaks. Furthermore, the sediments contain common reworked Cretaceous planktic foraminifera and inoceramid crystals.

5.6. General faunal changes

Uppermost Paleocene biofacies 1 contains numerous microfossils and foraminiferal assemblages are poorly diverse due to the strong dominance of a few taxa (Haplophragmoides, Lenticulina and Anomalinoides midwayensis). The transition from biofacies 1 to 2 between NAS 10.0 and 10.1 m is very abrupt and coincides w ith the onset of the CIE, marking the onset of the PETM. Foraminiferal assemblages indicate a complete faunal turnover w ith the disappearance or dim inution of many Paleocene taxa and appearance of Eocene taxa. This abrupt faunal turnover is the most prom inent change in the studied interval. Benthic foraminiferal assemblages in biofacies 2 are more diverse and most species encountered in biofacies 2 do not occur in biofacies 1 but remain present in biofacies 3. Another remarkable feature is the very low number of benthic specimens and the increase in planktic ones leading to a higher P/B ratio in biofacies 2 and 3, reaching highest values in biofacies 3. The transition from biofacies 2 to 3 is gradual and correlates w ith a lithologie transition from shales to marls. Various faunal changes occur at different levels, such as the disappearance of Reophax sp. 1. Biofacies 3 is typified by a minor diversity decrease, slightly higher foraminiferal numbers and notable changes in foraminiferal abundance. Biofacies 2 taxa become gradually less frequent (e.g. lagenids) and species w ith a more common occurrence in biofacies 1 (e.g. Neoeponides elevatus) or biofacies 4 (e.g. Cibicidoides succedens) increase in abundance. The transition from biofacies 3 to 4 is abrupt, corresponding to the base of the thick phosphate bed. Biofacies 4 is characterized by the common occurrence of Stainforthia. Only part of the biofacies 1 taxa reappear (e.g. Lenticulina midwayensis, Valvulineria scrobiculata) and characteristic latest Paleocene components are absent (A. midwayensis, F. aff. Frondicularia, diverse lenticulinid fauna).

6. Discussion

6.Í. Sedimentation rates and completeness o f the sections

The studied sections w ere located on the edge of the emerged Kasserine Island w here deposition occurred in a shallow marine


NO DATA

Erosion

Foraminiferal numbers

NO DATA

PF/gram — ♦ calcB F/gram ---- agB F/gram

d isso lu tio n

no or minor dissolution

mainly agBF fauna

i i i i i i r i i i i 100 0 20 40 60 80 100

Fig. 7. General foraminiferal param eters (num ber of planktic (PF), calcareous benthic (calcBF) and non-calcareous benthic (agBF) specimens per gram sedim ent and planktic/ benthic ratio) indicate a collapse of benthic foraminiferal fauna a t NAS 10.1 m and gradual increase in percentage planktic foraminifera w ithin biofacies 2 and 3. Foraminiferal num bers increase again during biofacies 4.

environm ent (Stassen e t al., 2009). There are no lithologie or biotic indications for im portant unconformities or discontinuities in the upperm ost Paleocene part. Only the lower phosphate layer a t Sidi Nasseur could be regarded as a condensed interval corresponding to a period of maximum sedim ent starvation (Saint-Marc, 1992). The average sedim entation rate of the NP9a interval (total thickness is 18.75 m, based on observations in nearby sections) is roughly estimated at 1.5 cm/kyr (based on the biochronology of Agnini et al., 2007). The TOC ô13Corg record reveals a sharp negative excursion at NAS 10.1 m and is not associated with any lithologie change (Fig. 3). In addition, the drastic decline of biofacies 1 taxa at this level indicates little or no reworking of upperm ost Paleocene sediments into the basal Eocene, suggesting continuous sedimentation. The total duration of the preserved Sidi Nasseur PETM interval is difficult to estimate. If the PETM interval would cover the entire planktic foraminiferal zone El, then the minimum sedimentation rate is 5 cm/kyr (based on the biochronology of Berggren and Pearson, 2005). Furthermore, the slow drop of ô13Corg values over several meters and the lack of recovery carbon isotope values, leads to the assumption that the investigated PETM interval was deposited at higher sedimentation rates and reflects only the lower part of the C1E, representing less than 70 kyr (the C1E “core"; Röhl et al., 2007). Based upon this assumption, sedimentation

rates would be higher than 10 cm/kyr. In general, the absence of lithologie changes, high sedimentation rates and absence of reworking indicate that the early part of the PETM is complete and expanded in the Sidi Nasseur section.

The top of the PETM interval (biofacies 3) is truncated in both sections by barren phosphate beds. The overlying limestone bed at Sidi Nasseur is much younger in age (NP11), so a large stratigraphie gap is present around this phosphate level (a hiatus >1.7 myr). The sediments above the thick phosphate bed at Wadi Mezaz are older (upper NP9b and NP10) and consequently a smaller gap is present there. The biofacies 4 interval at Wadi Mezaz regularly contains thin phosphate layers and the overall sedimentation rate is probably low w ith periods of sediment starvation. The position of the NP9b-NP10 boundary is defined by the lowest occurrence of Tribrachiatus bramlettei and seems to be diachronous as this boundary is placed within the PETM interval in high latitude sections (e.g. North Atlantic, Cramer et al., 2003 and Walvis Ridge, Agnini et al., 2007) and above the PETM interval in low latitude settings (e.g. Tethys, Dupuis et al., 2003 and equatorial Pacific Ocean, Raffi e t al., 2005). This apparent diachrony can be partly attributed to taxonomic inconsistencies or taphonomic problems (see discussion in von Salis et al., 2000). Furthermore, the abundance patterns at Walvis Ridge (ODP site 1262)

76 P. Stassen e t al. / Palaeogeography, Palaeoclimatology, Palaeoecology 317-318 (2012) 66-92

Taxa

-o

Erosion

-o

Fisher-a H(s)

- calcBF agBF

t ó f

I f

Dominance % BF/ AgBF total BF

CalcBF

E n d o H E p i & E n d o d E p i

0 10 20 30 0 2 4 6 8 10 0 1 2 3 4 0 20 40 60 80 0 20 4 0 60 80 100 0 20 40 60 80 100

Fig. 8. Diversity trends (taxa, Fisher-a, Shannon index H(s) and dominance) and abundance o f morphogroups (epibenthic — Epi, endobenthic — Endo and endobenthic to epibenthic morphology — Endo and Epi) visualize the biotic turnover to a more diverse benthic fauna during the PETM, composed mainly of small endobenthic benthic foraminifera.

display scattered occurrences of T. bramlettei around the PETM interval (Agnini et al., 2007). For these reasons, we use the level of the common occurrence of T. bramlettei to calculate a minimum sedimentation rate. The average sedimentation rate of the Wadi Mezaz NP10 interval is estimated at >0.4 cm/kyr (based on the biochronology of Agnini e t al., 2007).

6.2. Taphonomy and test production

The calcareous benthic foraminiferal numbers display rapid fluctuations in biofacies 1, very low numbers in biofacies 2 and 3 and a return to higher numbers in biofacies 4. Foraminiferal numbers depend on three main parameters, namely test production, sedimentation rate (condensation or dilution by other sedimentary particles) and taphonomic alteration (e.g. transport, degradation and post-depositional dissolution of calcareous components). The production of tests depends on the rate of successful reproduction which is mostly regulated by competition, the availability of food and dissolved oxygen (see Section 6.4). The effect of condensation or dilution by sedimentary particles is not well constrained due to uncertainties in exact short-term sedimentation rates, but can be a significant factor in shallow environm ents. Partial dissolution causes an underestim ation of the foraminiferal numbers and the combination of high percentages of non-calcareous agglutinated taxa with low absolute numbers of calcareous individuals is often regarded as an indicator of strong taphonomic alteration (Murray, 2006). Using these dissolution criteria, several dissolution levels are suspected (Fig. 7). These levels occur in narrow zones in biofacies 1

(NAS 4.05-4.3, 6.3, 8.1 and 8.7-9.4 m), a broad zone in biofacies 2 (NAS 10.1-12.45 m) and only one thin zone in biofacies 4 (MEZ 17.75 m). The lack of elevated absolute numbers of non-calcareous agglutinated foraminifera in these beds indicates that the high relative abundance of this group is caused by the near absence of calcareous foraminifera, rather than a bloom of agglutinated foraminifera.

During the latest Paleocene (biofacies 1), sedimentation in the Sidi Nasseur area occurred in a shallow marine setting (Stassen et al., 2009). Studies on recent highly productive shallow water environments demonstrate that many calcareous benthic species live in coastal settings but a significant taphonomic loss in the transition from living to fossil assemblages may occur. The mechanism behind this is the oxidization of organic material in the taphonomically active zone, leading to higher levels of dissolved C02 in pore water, resulting in a drop in foraminiferal numbers and an overall relative increase in abundance of agglutinated foraminifera (Murray and Alve, 1999; Berkeley et al., 2007). Combined with the percentage carbonate still present in the Sidi Nasseur sediments (Fig. 3), the suspected dissolution levels can be divided into two groups separating the biofacies 2 interval from the rest. The carbonate content drops simultaneously with calcareous foraminiferal numbers in biofacies 1 and supports the idea that these levels are affected by post-mortem dissolution processes (Stassen et al., 2009). The Sidi Nasseur biofacies 1 levels with the lowest calcareous foraminiferal num bers may in fact reflect intervals with high test production, but with sufficient oxygen levels to degrade the organic material. Most fluctuations in foraminiferal numbers in biofacies 1 are probably the result from the combination of post-depositional taphonomic effects (dissolution)


and winnowing (condensation). Variations in calcareous test production cannot be positively determined. The sporadic enrichment in the phosphate layer by large robust planktic foraminifera in biofacies 1 is likely the result of removal of the smaller size fraction.

In contrast, the carbonate content in biofacies 2 and 3 remains more stable and relatively high in combination with very high abundances of non-calcareous agglutinated foraminifera in the lower part of biofacies2. Another approach is an evaluation of the preservation potential of foraminifera. Reduced planktic/benthic ratios can be indicative of selective dissolution because planktic foraminifera are in general more susceptible to dissolution (Boltovskoy and Totah, 1992; Nguyen et al., 2009).

Additionally dissolution susceptibility differs between size fractions, with the smaller specimens dissolving more rapidly, leading to a larger average size of the remaining association (Nguyen et al., 2009). In the studied sections, absolute numbers of planktic foraminifera are increasing in biofacies 2 whereas absolute numbers of all benthic species drop to very low values with a noticeable size reduction of the benthic foraminifera, pointing to another explanation than post-mortem dissolution. As average sedimentation rates increased from 1.5 cm/kyr to at least 10 cm/kyr, the effect of dilution may contribute to the low benthic foraminiferal numbers. However, we consider a lowered test production as a more important factor (see Section 6.4).

Plate I. Paleocene biofacies 1 fauna, scale bar represents 100 pm unless indicated otherwise, (see on page 78)

1) Spiroplectammina mexiaensis Lalicker (NAS 7.0 m).2) Haplophragmoides excavata/walteri Cushman & Waters/(Grzybowski) (NAS3) Haplophragmoides excavata/walteri Cushman & Waters/(Grzybowski) (NAS4) Haplophragmoides excavata/walteri Cushman & Waters/(Grzybowski) (NAS5) Ammobaculites expansus Plummer (NAS 9.5 m).6) Bulimina ovata d'Orbigny (NAS 7.85 m).7) Bulimina kugleri Cushman & Renz (MEZ 6.75 m).8) Bulimina gr. trigonalis (Ten Dam) (NAS 7.85 m).9) Bulimina gr. trigonalis (Ten Dam) (NAS 9.8 m).

10) a/b Lenticulina pseudomamilligera (Plummer) (NAS 7.3 m).H ) Stainforthia solignaci (Grignani and Cococcetta) (NAS 7.9 m).12) Stainforthia kamali Bou Dagher (NAS 7.3 m).13) Frondicularia aff. phosphatica (Russo) (NAS 2.75 m).14) Pyramidulina latejugata (Giimbel) (NAS 2.75 m).

Plate II. Paleocene biofacies 1 fauna, scale bar represents 100 pm. (see on page 79)

1) a/b/c Valvulineria scrobiculata (Schwager) (NAS 3.7 m).2) a/b I c Neoeponides elevatus (Plummer) (NAS 7.9 m).3) a/b/c Anomalinoides midwayensis (Plummer) (NAS 7.5 m).4) Anomalinoides midwayensis (Plummer) (NAS 7.0 m).5) a/b/c Cibicidoides praecursorius (Schwager) (NAS 7.3 m).6) a/b/c Alabamina midwayensis Brotzen (NAS 3.7 m).7) a/b/c Anomalinoides lordi Bou Dagher (NAS 9.9 m).8) a/b/c Gyroidinoides girardanus (Reuss) (NAS 3.7 m).9) a/b/c Anomalinoides umboniferus (Schwager) (NAS 3.7 m).

Plate III. PETM biofacies 2 and 3 fauna, scale bar represents 100 pm. (see on page 80)

1) Ammobaculites expansus Plummer (NAS 12.5 m).2) Ammobaculites midwayensis Plummer (NAS 10.1 m).3) Reophax sp.1 (NAS 10.6 m).4) Lagenammina sp.1 (NAS 10.1 m ).5) Stainforthia kamali Bou Dagher (NAS 10.85 m).6) Bulimina midwayensis (Cushman & Parker) (MEZ 6.0 m).7) Bulimina ovata d'Orbigny (NAS 14.0 m).8) Bulimina aff. strobila (Marie) (NAS 14.0 m).9) Bulimina gr. trigonalis (Ten Dam) (NAS 10.6 m).

10) Lagena hispida Reuss (NAS 12.5 m).11) Lagena polygonissima Willems (NAS 12.0 m).12) Lagena striata (Montagu) (NAS 12.5 m).13) Lagena laevis (Montagu) (NAS 12.5 m).14) Lagena sulcata (Walker & Jacob) (NAS 12.5 m).15) Neoeponides elevatus (Plummer) (NAS 14.0 m).16) a/b/c Nonionella insecta (Schwager) (MEZ 5.5 m).17) a/b/c Cibicidoides megaloperforatus (Said & Kenawy) (NAS 11.4 m).18) a/b/c Baggatella cf. coloradoensis Malumian (NAS 10.8 m).19) a/b/c Cibicidoides howelli (Toulmin) (NAS 11.4m).20) a/b/c Valvulineria? insueta (Cushman & Bermúdez) (NAS 12.5 m).

Plate IV. Eocene biofacies 4 fauna, scale bar represents 100 pm. (see on page 81)

1) Stainforthia gafsensis (Grignani & Cococcetta) (MEZ 12.0 m).2) Stainforthia kamali Bou Dagher (MEZ 12.0 m).3) Stainforthia? sp.1 (MEZ 15.25 m).4) Stainforthia? sp.1 (MEZ 15.25 m).5) a/b Lenticulina midwayensis (Plummer) (MEZ 19.5 m).6) a/b/c Cibicidoides succedens (Brotzen) (MEZ 18.5 m).7) a/b/c Cibicidoides howelli (Toulmin) (MEZ 16.75 m).8) a/b/c Valvulineria scrobiculata (Schwager) MEZ 16.75 m).9) Acarinina multicamerata Guasti and Speijer (NAS 12.0 m).

10) Acarinina multicamerata Guasti and Speijer (NAS 12.7 m).11) Acarinina multicamerata Guasti and Speijer (NAS 12.5 m).12) Acarinina multicamerata Guasti and Speijer (NAS 10.6 m).


Plate I (caption on page 77).


Plate II (caption on page 77).


Plate III (caption on page 77).

Plate IV (caption on page 77).


Sedimentation rates dropped again in biofacies 4, creating several condensed phosphate layers. The peak in foraminiferal numbers in the phosphate bed at MEZ 16.75-17.25 m is probably the effect of winnowing. Paleoenvironmental param eters of biofacies 4 point to a restricted environment (see Section 6.5) limiting to some degree overall test production, resulting in higher foraminiferal numbers compared to biofacies 2 and 3 but lower than biofacies 1. Low abundances of non-calcareous agglutinated foraminifera and continuous presence of planktic foraminifera (mostly reworked Cretaceous specimens) suggest that post-m ortem taphonomic effects had limited effects on the foraminiferal assemblages in biofacies 4.

Alternatively, em pty foraminiferal tests can be considered as sedimentary grains and therefore subject to transport. The foraminifera of biofacies 2 and 3 are well sorted (<300 pm), which may suggest transport effects. The most common mechanisms of transport in shallow marine environments are bed load and suspended load, which are both capable of moving tests from their original environm ent to another depositional area (Murray, 2006). We exclude the effect of bed load due to the lack of visual abrasion. The fauna could be transported as suspension load but the small benthic foraminifera are all endobenthic morphotypes; juvenile forms of larger trochospiral forms are absent and reworking of upperm ost Paleocene specimens did not occur, indicating a change in paleoenvironmental conditions rather than taphonomic processes. Furthermore, these small benthic foraminifera are mixed w ith larger spines of irregular sea urchins (>1 mm) indicating the absence of sorting. Based on these observations, we consider the reduction in test size as a genuine adaption to environmental conditions.

6.3. Paleobathymetric evolution

Studies on outcrops of the El Elaria Formation near El Kef indicate a Tunisian inner neritic open marine setting during the late Paleocene (Kouwenhoven et al., 1997). Since El Kef was located more offshore from the Kasserine high than the Sidi Nasseur area (Guasti et al.,2006), an even shallower inner neritic to coastal setting is predicted for this region (Stassen et al., 2009). W ater depth preferences (Fig. 9) and resulting paleodepth estimations are elaborated for both non-calcareous and calcareous benthic foraminifera but the better constrained paleoecologic preferences of calcareous benthic taxa allows the most straightforward paleodepth reconstruction (Fig. 10).

The non-calcareous benthic fauna of biofacies 1 is mainly composed of Haplophragmoides excavata/walteri, indicative for an open marine inner neritic to coastal setting (Auberi and Berggren, 1976; Saint-Marc, 1992). Planktic foraminifera are nearly absent in biofacies 1 w ith P/B ratios well below 1% in levels that do not appear affected by dissolution. The combination of characteristic inner neritic calcareous benthic foraminifera, combined w ith strong dominances of coastal species and the near absence of planktic foraminifera also suggest a shallow inner neritic and probably coastal environment with connections to the open marine Tethys as can be judged from the common presence of calcareous nannofossils (Stassen et al., 2009). In addition, an upward shallowing trend towards a more coastal setting is noted based on the increase in dominance of A. midwayensis and Ammobaculites species. The lower abundances of inner neritic species and the disappearance of species typical for middle neritic assemblages in the southern Tethys, initially occurring in minor proportions in the lower part of biofacies 1 also support the idea of a shallowing trend (Fig. 10).

Biofacies 2 is mainly composed of inner to middle and even outer neritic taxa (e.g. Bulimina aff. strobila); only a minor com ponent refers to the former coastal environment. Most of these species also occur in the upper Paleocene part of the El Kef section, tha t is located more offshore (Kouwenhoven e t al., 1997). The higher frequency of lagenids also points to deeper w aters as these taxa are less frequent at shallower depths, although little is known about their precise

environm ental preferences (Hermelin and Malmgren, 1980; Albani and Yassini, 1989; Schnack, 2000). Calcareous species assigned to a middle neritic environm ent become less abundant in biofacies 3 and species related to coastal environm ents (e.g. Neoeponides elevatus) increase again, indicating a renew ed shallowing trend. Biofacies 2 and 3 contain higher absolute num bers and percentages of planktic foraminifera. Under normal marine conditions, higher P/B ratios are commonly linked to an increase in w ater depth, but here benthic bottom life was affected by an increase in oxygen deficiency (see Section 6.5), rendering P/B ratios as unreliable proxies for w ater depth (e.g. van der Zwaan et al. 1990). However, taking into account tha t biofacies 2 and 3 w ere deposited under enhanced sedim entation rates, the small increase in planktic foraminiferal num bers certainly indicates more open marine conditions. Non-calcareous benthic foraminifera may suggest an overall shallowing trend across the PETM, but the exact paleoecologic preferences of these, partly lumped, taxa are less well-constrained and they are only a minor component of the total benthic assemblage within the PETM interval.

Biofacies 4 with high abundances of Stainforthia species strongly differs from previous biofacies. Similar Stainforthia assemblages are found in restricted nearshore or lagoonal facies in southern Tunisia (Auberi and Berggren, 1976; Bou Dagher, 1987). The benthic assemblages ofbio- facies 4 are linked to these restricted environments (Stainforthia) and to an open marine connection as suggested by the presence of calcareous nannofossils, Lenticulina species (mostly L. midwayensis), Valvulineria scrobiculata and Cibicidoides succedens. The absence of Anomalinoides midwayensis and other characteristic Paleocene taxa (Frondicularia and Pyramidulina spp.) suggest a more restricted coastal, perhaps lagoonal, environment (Fig. 10). Compared to biofacies 1, the higher percentages of planktic foraminifera are caused by reworked Cretaceous planktic foraminifera and lower benthic foraminiferal numbers. Non-calcareous agglutinated taxa are generally not frequent and the inner neritic- coastal Haplophragmoides fauna is largely replaced by a shallow water Ammobaculites fauna.

The benthic foraminiferal biofacies transitions suggest significant paleodepth changes across the studied intervals (Fig. 10). During biofacies 1, w ater depth decreased from a shallow inner neritic (> 30 m) to a coastal environment (<20 m). A similar latest Paleocene sea level fall just prior to the PETM was also observed at Gebel Duwi (Egypt, Speijer and Morsi, 2002). At Dababiya (Egypt), the presence of a distinct channel-like incision underlying the PETM interval indicates a discontinuity (Dupuis et al., 2003; Ernst et al., 2006; Aubry et al.,2007). These observations also suggest a sea level fall prior to the PETM. The sudden increase in w ater depth by approximately 30-40 m from coastal towards middle neritic deposition (50-60 m) is associated w ith the onset of the PETM and its timing suggests a strong link betw een sea level rise and the PETM (Fig. 10). Sea level rise of similar magnitude is also noted elsewhere (Speijer and Morsi, 2002; Ernst et al., 2006; Sluijs et al., 2008) and it has been proposed that eustatic rise during the PETM may be related to the m elting of small Antarctic ice sheets and thermal expansion of the w ater column (Miller et al., 2005; Sluijs et al., 2008). Yet, the early Paleogene is generally considered to have been ice-free at the poles, so this scenario remains controversial. The top part of the PETM is not preserved and is probably linked to a subsequent relative sea level fall w ith renewed sedimentation during the next Eocene transgression, filling erosion channels w ith phosphatic particles. The overlying sediments were deposited in a shallow restricted environment. A similar sedimentary sequence is also described at Ellès (Tunisia, Bolle e t al., 1999) where the PETM corresponds to a 2.7 m eter thick clay layer, cut off at the top by a thick phosphate layer.

6.4. Trophic and redox conditions

The total organic carbon content is low throughout the studied part of the Sidi Nasseur section, including the PETM (average of


S p e c ie s

Lag3

Co

Water d 0 5

IN

ep th 0 1C

MN0 2

ON0

B

P a le o d e p th p re fe re n c e

(u p p e r lim it isM N )

Lifeepi

n o d ee n d o

Salih ypo

nity to le r no rm al

an c eh y p e r

R e fe re n c e s

A JLLJLL-'-l- ' i -ftl_L A v 1 1 ^

CO IN 2 6 13R e o p h a x sp p .

S p iro p le c ta m m in a m e x ia e n s is5 .1 3

Al H 1 V* 1 _ _

A n o m a lin o id e s lord i CO 12

C ib ic id o id e s h o w elli C ib ic id o id e s m e g a lo p e r fo ra tu s

unknown X X 23 and epiphytic lifestyle

2, 3 ,7wide range

cM° ;L a g en a s p p . probably > IN

X

S ta in fo r th ia ? sp .1 . . . Lag-CO wide range wide range

X X s e e Stainforthia g a fsen sis

Valvulineria scro b icu la ta 2, 3, 7 ,1 3

Fig. 9. The paleodepth preferences of the m ost common taxa are arranged along a depth gradient. The terms neritic and bathyal are used to separate between paleodepths less than and in excess of 200 m. The terms coastal (Co), inner neritic (IN), middle neritic (MN) and outer neritic (ON) correspond with paleodepths respectively of 0-30 m, 30-50 m, 50-100 m, 100-200 m. The term lagoonal (Lag) corresponds with a shallow restricted environm ent Thick solid lines mean most frequent occurrences, thin solid lines common occurrences and discontinuous lines rare occurrences. Paleodepth preference refers to optimal living conditions, with an upper limit of middle neritic due to the shallow nature of this region, meaning that all benthic foraminifera with paleodepth preferences larger than 50 m are considered to be indicative of middle neritic. The terms hyper, normal and hypo refer to a salinity tolerance to respectively hypersaline, normal saline and hyposaline bottom water conditions. Based on data from: 1 — Kellough, 1965; 2 — Saint-Marc and Berggren, 1988 and Saint-Marc, 1992; 3 — Schnack, 2000; 4 — Kaiho, 1994; 5 — Murray, 2001; 6 — Jones and Chamock, 1985; 7 — Speijer and Schmitz, 1998; 8 — Berggren, 1974; 9 — Aubert and Berggren, 1976; 10 — Kouwenhoven et al, 1997; 11 — Bou Dagher, 1987; 12 — Bou Dagher, 1988,13 — Murray, 2006 and 14 — Stassen e t al, 2009.

0.24%) and probably reflects surface w eathering of the outcrops (Fig. 3). As the initial TOC content is unknown, it cannot be used as a proxy for organic flux. Insights into the trophic regime are therefore derived from the biotic record. Oxygen availability and organic flux (food supply) are considered to be the major parameters structuring benthic foraminiferal communities. These param eters are closely interrelated and it is difficult to separate their influences in shallow m arine environments (Jorissen et al„ 1995; van der Zwaan e t al„ 1999). In general, organic flux is im portant in regulating foraminiferal densities, but becomes subordinate as soon as oxygen starts to be the limiting factor (van der Zwaan et al„ 1999). Taxa w ith an endobenthic microhabitat become more abundant as the organic flux increases and these taxa also show a systematic population increase under dysoxic bottom w ater conditions (Jorissen, 1999).

The latest Paleocene coastal biofacies 1 assemblage is dominated by Lenticulina spp., Anomalinoides midwayensis, buliminids and non- calcareous agglutinated Haplophragmoides spp. Both Lenticulina and Haplophragmoides are ecologie generalists as these taxa are capable of alternating betw een an epibenthic and shallow endobenthic lifestyles in reaction to low-oxygen eutrophic conditions often occurring in organic-rich mud (Saint-Marc and Berggren, 1988; Ernst et al„2006). Early Paleogene buliminids are regarded as indicators of a high food flux in association w ith low oxygen conditions (e.g. Kouwenhoven et al„ 1997; Ernst et al„ 2006). The occasional interruptions by elevated percentages of the buliminid group in biofacies 1 suggest periods of possible elevated food supply and/or lowered oxygen conditions at the seafloor (Stassen e t al„ 2009). Furthermore, many agglutinated taxa have an inner organic layer to which the solid particles of the test are attached. These organic components are not preserved under oxic conditions due to microbial degradation (Berkeley et al„ 2007). The encountered agglutinated taxa at Sidi

Nasseur have siliceous cement, but presumable also organic components before fossilization. The high preservation potential of these taxa in biofacies 1, indicate more or less continuous lowered oxygen levels, but slightly more oxic conditions did occur regularly because post-m ortem dissolution is probably linked to temporarily enhanced oxidization of organic material.

The Tunisian benthic assemblages in biofacies 2 and 3 are more diverse w ith a uniform distribution betw een taxa, yet foraminiferal numbers are extrem ely low and partly linked to higher sedim entation rates. Most species have an endobenthic morphology (Fig. 11) adapted to high food and/or low oxygen living conditions (e.g. lagenids, buliminids, Reophax). A dominance by lagenids has not been described in the southern Tethys region before and it may indicate an opportunistic behavior, probably under dysoxic conditions (Kaiho, 1994). The reduced foraminiferal num bers are not the result of a severe deficiency in food supply because burrowing sea urchins, dependent on a large and stable food source, lived nearby (as indicated by large am ounts of spines). Furthermore, foraminiferal test size reduction occurred in biofacies 2 and 3. Small sized foraminifera might better survive episodes of oxygen depletion than larger species because of less oxygen requirem ents (Bernhard and Sen Gupta, 1999). In the modern shallow Adriatic Sea (Duijnstee et al„ 2005) and in Ría de Vigo (Spain; Diz et al„ 2006), inhibited grow th and elevated m ortality are indeed the effect of oxygen stress induced by extra organic flux. The persistence of endobenthic lifestyles, reduced test size and low foraminiferal num bers in biofacies 2 and 3 indicate the continuation of eutrophic conditions with increased oxygen deficiency in comparison w ith biofacies 1.

Biofacies 4 is less diverse w ith strong dominance of large sized Lenticulina spp. (mostly L midwayensis), Valvulineria scrobiculata, Stainforthia spp. and Cibicidoides succedens. The first three taxa are


Erosion

NO DATA

BCalcBF Paleodepth

Lag-CoCoCo-IN

Unknown

Erosion,

0Tethyan sea level

fluctuations

? * E□ 3

notstud ied

n o ts tud ied

Fig. 10. A: Total abundances of benthic foraminifera grouped into assemblages according to their paleodepth preferences under normal Paleocene to Eocene conditions (lagoonal to coastal — Lag-Co; coastal — Co; coastal to inner neritic — Co-IN, inner neritic — IN, middle neritic — MN, Wide range — species with a wide range of paleodepth preferences and unknown represent the remaining taxa of the benthic foraminiferal abundances). This figure visualizes the strong influx of middle neritic species during the PETM (biofacies 2 and 3) and the strong occurrence of lagoonal to coastal species in biofacies 4. B: Paleodepth trend estimations based upon paleodepth preferences (gray area represents minimal and maximal range of w ater depth and the black line is the mean value). C: Schematic representation of Tethyan sea level fluctuations (not to scale), based on several authors (Tunisia NAS & MEZ: this study; Tunisia El Kef: Kouwenhoven et al., 1997; Egypt Gebel Duwi, Speijer and Morsi, 2002; Egypt Dababiya, Ernst e t al., 2006).

ecologie generalists with a broad adaptive range with potentially endobenthic lifestyles (Fig. 8) and can tolerate large changes in environmental parameters. High abundances are often linked to high food and/or low oxygen conditions (Kaiho, 1994; Ernst et al„ 2006; Murray, 2006). The observed biofacies 4 fauna also reflects eutrophic conditions with oxygen deficiency in a restricted coastal or lagoonal environment.

6.5. Controlling environmental parameters

Benthic foraminifera occupy a specific ecologie niche in which optimal living conditions occur. Also, a species does not have to be perfectly adapted to its habitat; it simply m ust be sufficiently adapted to do better than other taxa. W ater depth (hydrostatic pressure) itself is not an ecologie factor regulating the distribution of benthic foraminifera; species living on the shelf are affected by various environm ental param eters tha t tend to vary proportionally w ith depth, like sea w ater chemistry, food availability, com petition and oxygenation. In many m odern shallow marine environm ents w ith a muddy substrate, organic flux is sufficiently high but prevailing low-oxygen levels place constraints on the foraminiferal comm unity and can lead to the developm ent of impoverished stressed faunas (Loubere and Fariduddin, 1999; van der Zwaan e t al„ 1999).

In the Sidi Nasseur and Wadi Mezaz sections, coastal species thrived during biofacies 1, being replaced by deeper dwelling taxa during the PETM (Fig. 11). Species adapted to lagoonal or restricted coastal conditions are present in biofacies 4. As the biofacies are a ttributed to specific paleodepths, the question is which environmental param eters destroyed or created suitable niches for the taxa in these

assemblages. The differences in biofacies and underlying environmental conditions are visualized by means of the detrended correspondence analysis (Figs. 6 and 11). Axis 1 is assumed to express bottom water oxygen conditions because most species with a potential endobenthic microhabitat have more positive scores and are represented in the PETM interval by small sized benthic foraminifera. Large and/or epibenthic species have more negative scores. Axis 2 is mainly controlled by the scores of the Stainforthia group versus Anomalinoides midwayensis. Axis 2 is believed to express bottom water salinity conditions as different salinity tolerances are attributed to these species (Figs. 9 and 11). Foraminifera can tolerate brackish to hypersaline conditions, but the majority of species and highest diversities are found in normal marine conditions (Murray, 2006). The results from the detrended correspondence analysis suggest that salinity plays an important role in the observed foraminiferal distribution patterns. Overall, the absence of miliolids or gypsum deposits indicates the lack of high salinity during the late Paleocene to early Eocene.

The m odern Ría de Vigo (Spain) environm ent is used as a potential modern equivalent of the Tunisian biofacies distribution. In this coastal environm ent, the muddy organic rich sedim ents impose a combination of factors close to the threshold of tolerance for many benthic species and there is a decreasing trend in diversity from deeper to shallower areas (Diz et al., 2006; Diz and Frances, 2008). In the shallowest parts, the combination of high food supply, low oxygen and, importantly, variations in salinity cause the establishm ent of a poorly diverse estuarine foraminiferal assemblage. Large sized epibenthic foraminifera live in normal saline near-shore areas and are considered to be k-strategists, spending energy on shell growth,


Erosion

infaunal/ Axis 1epifaunal oxygenation Paleodepth

E ndoben th ic Epi- & E ndoben th ic

I I E pibenth ic

Lag-Co

C o-IN

W id e ra n g e

Axis 2 salinity

= ■-w 15S £ <ura o.

Salinitytolerance

I hy p o salin e ■ hyposaline-norm al I norm al

I ~~1 hypersaline-norm al

0 20 40 60 80100 0 20 40 60 80100 0 1 2 3 4+ < ► -

0 20 40 60 8 0 1 0 0

Fig. 11. Samples scores of the first and second correspondences axis (detrended correspondence analysis) are plotted versus stratigraphie order. AU axes show relationship to paleodepth expressing underlying controlling environmental parameters such as bottom w ater oxygen levels (axis 1 ) and bottom w ater salinity gradients (axis 2).

rather than on offspring. The coastal benthic foraminiferal com m unities are poorly diverse because these k-strategists are capable of out- competing other foraminiferal taxa. The inner neritic benthic foraminiferal assemblages are more diverse in the deeper parts (40 m w ater depth) because more ecologie niches are available (e.g. fresh labile organic material derived from seasonal blooms of diatoms). In these specific m odern settings, the neritic assemblages are composed of dwarfed endobenthic species (r-strategist) related to seasonal upwelling, creating highly productive environm ents w ith reduced bottom w ater oxygen levels.

The low diverse inner neritic to coastal Paleocene biofacies 1 setting was mainly inhabited by large sized benthic foraminifera. Considering Ría de Vigo as a modern equivalent, the dominant biofacies 1 species were well adapted to the coastal setting and outcompeted other taxa. Overall, Axis 1 scores are lowest in biofacies 1, indicating the highest (variable) oxygen levels compared to the other biofacies, w ith lower oxygen levels coinciding with the levels of higher buliminid frequencies (Fig. 11). The observed biofacies 1 fauna with high dominance values are thus the result of continuous eutrophic conditions under moderate dysoxic regime in a shallow marine coastal environment. The increasing dominance of the large epibenthic A. midwayensis is also considered to be an adaptation to more saline coastal waters, probably due to the increasing effect of evaporation in a shallowing basin (Stassen et al., 2009). The mixture of normal marine taxa and taxa tolerant to more hypersaline conditions (A. midwayensis) indicate normal to slightly hypersaline conditions during the latest Paleocene (biofacies 1).

The Tunisian PETM benthic assemblages (biofacies 2 and 3) are more diverse and equable, suggesting a less stressed environment, yet axis 1 scores are highest in biofacies 2, suggesting the lowest oxygen levels (severe dysoxia). In muddy sediments, the redox boundary is normally situated a few centimeters w ithin the sediment even where the overlying bottom w ater is sufficiently oxygenated. Consequently, all muddy environments have endobenthic taxa living only a few centimeters away from potentially epibenthic taxa. In the Sidi Nasseur and Wadi Mezaz sections, the near absence of epibenthic morphotypes indicates the extension of dysoxia into the lower part of the w ater column. The few large epibenthic taxa have an epiphytic lifestyle (e.g. C. megaloperforatus) and may have lived well above the seafloor or were washed in. According to the interpretation of axis 1, severe continuous oxygen deficiency occurred during the early phase of the PETM. Prolonged periods of extreme dysoxia are unlikely due to the absence of laminated sediments and the continuous presence of benthic fauna. Moreover, in shallow seas, seasonal mixing or storm events can ventilate the basin leading to rapid restoration of benthic faunas. The observed smaller size of species w ith an endobenthic morphotypes and the increase in diversity (more niches available), may be the result of (seasonal?) changes in organic flux from the photic zone. Because most specimens probably did not reach adulthood due to limited oxygen supply, reproduction failed leading to low foraminiferal numbers. Salinity was not a limiting factor as foraminiferal associations point to normal saline bottom w ater conditions during the PETM.


Biofacies 4 w ith high abundance of Stainforthia strongly differs from the other biofacies. Bottom w ater oxygenation (m oderate dysoxia) was probably not the controlling environm ental param eter. Similar benthic associations are linked to restricted nearshore or lagoonal facies w ith hyposaline w ater conditions in Tunisia (Auberi and Berggren, 1976; Bou Dagher, 1987). The absence of Elphidiella prima, which is often the major constituent in hyposaline lagoonal assemblages, the continuous presence of coastal benthic species and nannofossils indicate conditions betw een normal and hyposaline, perhaps seasonally fluctuating (Fig. 11). Under these eu trophic, m oderately dysoxic hyposaline environm ental conditions, only taxa (e.g. Stainforthia) w ith a broader tolerance range can thrive and species adapted to normal marine conditions will disappear (e.g. nodosariids).

6.6. Epeiric sea dysoxia

Oxygen deficiency occurs w hen the oxygen dem and created by the decay of metabolizable organic m atter exceeds the rate of oxygen supply. Oxygen depletion of bottom w aters in shallow epeiric shelf environments indicates insufficient circulation to homogenize or resupply the dissolved oxygen content of bottom waters and is explicable by several general models, namely oxygen minimum zones, irregular bottom topography, upwelling regimes and stratified basins, which are not necessarily mutually exclusive (Tyson and Pearson, 1991). The establishment of an oxygen minimum zone is unlikely in this region as it often occurs at w ater depths deeper than 200 m. High subsidence rates can cause irregular bottom topography with small areas sinking below a depth where circulation is unable to ventilate the bottom waters. This is unlikely as sedimentation rate was very low during the latest Paleocene and a shallowing trend was observed, indicating infilling of the basin.

The idea of enhanced organic fluxes created by upwelling systems seems more plausible as upwelling is thought to have occurred during the Paleocene along the borders of Kasserine Island, creating enhanced fluxes of organic m atter (El Kef; Kouwenhoven et al., 1997). At Sidi Nasseur, upwelling did probably not occur during the latest Paleocene due to the shallow nature of the setting (Stassen et al.,2009). During the PETM, the increase in w ater depth may have shifted or expanded the upwelling regions towards Kasserine Island and consequently increased productivity in the photic zone, leading to more oxygen consumption. This upwelling would have been caused by wind stress causing offshore Ekman transport of surface waters. The increase or shift in w ind-driven circulation would also lead to a higher proportion of w ater column mixing and better bottom w ater ventilation in shallow marine environments and is consequently incompatible w ith the observed establishment of enhanced dysoxic conditions.

Natural anoxia-dysoxia due to stratification occurs mostly in w ater masses where the bottom w ater layer is thin (subpycnocline at w ater depths betw een 20 and 60 m) and where low levels of tu rbulence (winds, tides, currents) prevail, preventing complete mixing of the w ater column. As surface waters are warmed by the sun, a seasonal pycnocline (thermocline) develops, decoupling the bottom w ater from the surface water. The smaller the volume of bottom water, the lower its initial available oxygen reservoir and the quicker it can become dysoxic (Tyson and Pearson, 1991). Additionally, periods of intense runoff can produce a low salinity surface layer causing extra stratification of the w ater column and reduced vertical mixing. Moreover, density gradients along a halocline are generally stronger than along a thermocline.

We propose a model of enhanced seasonal stratification in the Tunisian basin. During the latest Paleocene, the shift to more saline bottom w ater conditions would have created an anti-estuarine circulation pattern (negative fresh w ater balance) w ith inflow a t the surface and outflow at depth (Fig. 12; Hay, 1995). Due to the sea level rise a t the onset of the PETM, w ater depths become sufficiently deep to develop a seasonal thermocline. Additional warming related to the PETM would have caused an enhancem ent of the duration of the seasonal therm ocline as milder w inters, earlier springs and longer sum mers will cause a smaller therm al gradient (not enough w inter cooling to cause overturning of the w ater column) leading to a thickening of the seasonal therm ocline (pycnocline) at the expense of the bottom layer (subpycnocline). The observed Tunisian elevated sedim entation rate during the PETM involves a higher input of detrital particles and other indications o f increased runoff are also observed in the Tethyan area (Tunisia: Bolle e t al., 1999; Egypt: Ernst e t al., 2006 and Spain: Schmitz and Pujalte, 2003). Higher sedim entation rates during the PETM are also reported in o ther neritic PETM sections along the North American continental margin (John e t al., 2008). These periods w ith increased riverine runoff would have influenced the freshw ater balance, enhancing stratification of the w ater column (Fig. 12; Hay, 1995).

The th inner bottom layer, less dissolved oxygen due to higher tem peratures and the extension of the stratified periods would have caused the observed increase in bottom w ater oxygen deficiency during the PETM, inhibiting the establishm ent of large benthic foraminiferal communities. These conditions could have resulted in the w idespread reduction of benthic foraminiferal communities over large coastal areas and recolonization by non-opportunistic slower growing taxa would have been impossible. Fluxes of labile organic m atter (e.g. seasonal surface blooms) would generate diverse food sources for the benthic communities. In combination w ith tem porary re-oxygenation of the bottom waters, this could enable short term reproduction cycles of several opportunistic species (Reophax species, buliminids, lagenids) leading to a diverse death assemblage.

River induced oxygen deficiency is often linked to enhanced supply of labile organic matter, which may trigger rapid opportunistic periodic reproduction of shallow endobenthic taxa (van der Zwaan and Jorissen, 1991; van der Zwaan e t al., 1999). The overall result would be a drop in diversity and higher densities of opportunistic species, contrasting our Tunisian observations w here more diverse benthic foraminiferal comm unities w ith low densities developed. Secondly, although perhaps biased by weathering, no increase in burial of organic material is observed. We postulate the absence of high primary production as land derived nutrients w ere not fully captured near the coast w here only the sedim ent load was dumped. The less saline surface w aters would also have had a negative effect on the population dynamics of planktic taxa (calcareous nannoplankton and planktic foraminifera) as they are adapted to normal open marine conditions and have a very low tolerance to salinity changes. Also, high suspension load would have reduced light penetration effecting photosynthesis in the surface layer. This further explains the low planktic abundances. The increasing planktic/benthic ratio during the PETM may reflect a shift from continuous low saline surface w ater (biofacies 2) towards seasonal controlled salinity changes (biofacies 3). Alternatively, dinoflagellate blooms, also causing acceleration of oxygen consumption, can become abundant during stratified conditions when high nutrient levels accumulate at the base of the pycnocline within the euphotic zone (Tyson and Pearson, 1991). Unfortunately, organic-walled dinocysts are absent in the studied samples, probably due to intense weathering (van Simaeys, pers. comm.).

Fig. 12. Summary of all faunal patterns, associated paleoenvironmental indicators and the interpreted paleoenvironmental reconstructions in this study. The largest biotic shifts in calcareous nannofossil, planktic, benthic and ostracode (Morsi e t a l, 2011) assemblages coincide with environmental perturbation associated with the onset of the PETM. See text for discussion on the environmental interpretations.


c2>e

LU

(D CUi m «

(Ü O ~

8■ o — o fl) (D o o .E o'■¿= (D £ i/) ij) CL <1) O O

Q - Í 3CD Z3

o ■*= ®

Ocß(ßTD

E(Dn

c/> oQ_OC

5 . 0

<D (D

E »5 S

O p d J« ■ D Ö ) O <D(D<ü O_ f(D <D O ^ O LU

P r- °O s êC O)

CD-Q X 0tu o w Í2 >> o u>

OJ o = 0 . - 2 _ . "OCU US "O> (D ‘c Q) (ß C

— (0m E o (ü i : nCß iß S

ißo

CD

S ho LUC û_M I I

d) ç o c °<ß 3 ^

* i lO B - O

o ) jS

* so oo .

. 2 <D ; ~ E

75 B

a ï ^c Q -

OJZD.

^ iß

Po o ~ o

E a>

c - c? d) O tfi! g s sB O i t

o■O B

E o S*- c .«2

3> T3C

> - £ D -S

>

’ECDC/)

a u ||P so d A i| o } l e i u i o u

|PLUJOU

auijPSjadAq A p i j ô ü s o \ i p l u j o l i

comc0)C I>1X

O

a .<v■oo_a>(D

a .

P jX O S A p 0 J 0 A 0 S

PIX O S Â p 0}PJ0pO L U

l e j s e o o p 9 ) D i j ) 5 0 j o\ |Buooße|

P jX O SA p 0JP J0pO L U

□ l|I J 0 U J0U U j o; i p | s p o o

3 ¡flJ 0LÍ 0 |p p iL U O ) J 0 U U !

A M u / .p s e j o e p i q G Z

sa jO E jO jgsaiD gjO jq I S d j o ^ o j q

e u n q O U 0 O O 3 c u n c i i A I ± 3 d c u n c j a u a o o s i e j

d d s enjiJO fU fB¡s (sisudAeMpiLu '7 A| u ¡eu)

d d s eu t/nonu& 7 s u d p e o o n s qiQ

u a j j e q ~B1EP 0|SJ

d d s d » ™ ,n s i .d d s e ü J lü ( ;n sW A S ? j ddsB(ya5

s sp iuoddO B fyji

d d s eu iL uijng ( e u n e j a s jaA jp ) d d s BU finopu& i

s î s u b Àb m pilu o u y

d d s sa p io L u d e jq d o /d e h d d s s3 )!inoB qotuL u\/

u a j j e q ~ e p p o n

d d s sd p io u jß e jq d o id e H d d s SBj!inoeqoLULU\/

Q. g>« 2

*3 9- â% t *

i f

d d s s a p i o u j ß B j q d o i d e H

( d d s n o a o e p o ps>jJ0MSJ Â usow ) p e ip n js p u

u s j je q - e j e p o n

d d S B U/UIJBOV p U J0 0 |q j u a s q e j s o iu |e

&«q

2 E

d d s ja f s e o o s iç j SEoejsuud

s n o fß e ia d sn q ii/o o o o Q0SJ0AIP (X |0 p j0 p O lil)

u s j je q - e p p o n

sniBipBjfuniu jaiseoosfQ s n s n ¡ ja d q n s /s n o i5 e ja d 0 0 3 s u j jo j . s n o e o e p j Q p ©>|j o m s j

p 0 S B 0 JO 0 P S p jB M d n 9 SJ0 AJP A|0 }ej0 p o o i 01 MO i

d d s sn q ifin o io s b j d d s jQ fse o o s iQ aeaoBjSuud

sn o iÔ B ied s n q m o o o o o e u n e j esiS A ip

cu <u ■o o o <02 - ° 2 ES*O

SfSU&Mnp eueiidAGaÿ s i s u e u o ó ip e a ja q jA o o d o o o /\ /

u a j j e q ~ e je p o n

s i s u a is run t ¿ e iu o iu n g s is u B u o ß ip B 9JB if¡A oodooo i\/

s i s u s jn a s s e u e is o o i fu a u jA a y f iu n o ts s e q b iso o i}ubluAb ^

( u i )ssau>(om_i_

Z B z e iA i i p e / w ^

t £ 3 2S0tOB^O|a jeq^j uajjeq

S0UOZ n o

p 0 ip n ;s jo u

QldN I 16dNu o q e u j jo j

s a p a s

e u j q e n o q oa u 0 0 0 3

!P!Sí 5 S 5 r t C N ^ O O J C Ö ^ i Ö l ß ’T O «

I1 3

q e d N

S d - ^ d

B6 d N

0 U0 O O 3 0U0OO0|ed

O O CD 'iZ

C ~o

.0 < •

O « 1

« •

P. Stassen et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 317-318 (2012) 66-92

6.7. Shallow water biotic turnover during the PETM

The paleoenvironmental setting of the studied sections offers much needed information about the shallow water biotic response to rapid global warming and the spatial distribution of the anoxic-suboxic eutrophic conditions as recognized in all Egyptian shelf environments (Speijer and Wagner, 2002). In the Egyptian basin, the environmental changes triggered by the PETM caused a large basin-wide taxonomic turnover in association with a transient downslope migration of opportunistic benthic taxa (e.g. Anomalinoides aegyptiacus), yet biotic turnover at shallower water (Gebel Duwi) was preceded by a gradual diversity decline prior to the PETM (Speijer et al„ 1996). The Tunisian coastal benthic foraminiferal turnover at the onset of the PETM was abrupt and drastic and was associated with a full turnover in ostracodes (Morsi et al„ 2011; Fig. 12). During the PETM, a middle neritic benthic fauna colonized the eutrophic environment, replacing a coastal community, a pattern that strongly contrasts with the downslope migration in Egypt. The increase in water depth during the PETM, in combination with stratification, seems to be an important controlling feature in these coastal settings, provoking a complete faunal turnover. Several common Paleocene taxa disappear at the P-E boundary (e.g. A. midwayensis and A. umboniferus). W hether these disappearances represent temporary, local or real extinctions cannot be resolved because the latest Paleocene Sidi Nasseur paleoenvironment is very different from the early Eocene Wadi Mezaz environment (e.g. water depth, salinity, oxygen, competition... Fig. 12) and no other post-PETM Tunisian coastal environments are currently available for comparison.

An increase of surface dwelling planktic foraminifera (flat-spired multi-chambered Acarinina species) during the PETM is observed along the borders of the Tethys (e.g. Guasti and Speijer, 2007). This photosym- biotic genus is generally regarded as being indicative of oligotrophic open ocean conditions (Norris, 1996; Kelly et al„ 1998) but also occur in higher abundance in the deep water Forada section (northern Italy, Luciani et al„2007) and Egyptian sections (Ernst et al„ 2006; Guasti and Speijer, 2007) where Acarinina species were able to tolerate and thrive within the eutrophic environments during the PETM. We observed comparable behavior in Tunisia (Fig. 12), although densities are fairly low in this area. We concur with Guasti and Speijer (2007) that the common presence of flat-spired Acarinina is not indicative of oligotrophic conditions, but rather reflects stressed conditions not directly related to nutrient availability.

7. Conclusions

The paleoenvironmental evolution at Sidi Nasseur and Wadi Mezaz during the latest Paleocene and early Eocene is interpreted using various indicators and reveal relationships between global warming during the PETM and perturbations of shallow marine ecosystems.

• A sharp faunal turnover marks the onset of the PETM biofacies with the settlement of a deeper and more diverse but low density benthic assemblage, mainly composed of opportunistic endobenthic taxa, accompanied by a planktic Acarinina assemblage.

• The repopulation pattern observed in this shallow setting strongly contrasts w ith the low diversity opportunistic fauna recolonizing the basin in Egypt. W hether the Tunisian disappearances of Paleocene benthic foraminifera represent temporary, local or real extinctions is currently irresolvable.

• The proposed link betw een the onset of the PETM and rapid eustatic sea level rise is supported by the present study. Immediately after the onset of the PETM, a deeper dwelling assemblage settled, indicating a sea level rise of -3 0 -4 0 m. This rise was preceded by a -2 0 m sea-level fall during the latest Paleocene.

• The onset of the PETM is linked to enhanced oxygen deficiency (severe dysoxia) in a eutrophic inner to middle neritic environment. The rising sea level and elevated temperatures caused more stressed sea floor conditions due to the initiation of (seasonal) stratification, intensified

by enhanced runoff. The oxygen deficiency resulted in inhibited growth and elevated mortality prior to reaching the adult reproductive phase.

To conclude, the PETM exerted a major influence on the studied shallow water ecosystem. Sea level rise and oxygen deficiency played an important role in restructuring the benthic ecosystem, immediately after the onset of global warming. Unfortunately, early Eocene erosion precludes an assessment of the aftermath of the environmental perturbation.

Acknowledgements

We thank Mohadinne Ben Yahia for assistance in the field and Ruth Martin for the review that helped to improve the manuscript. Financial support was provided by grants from the K.U. Leuven Research Fund and the Research Foundation Flanders (FWO) to Robert P. Speijer and Etienne Steurbaut.

Appendix A. Taxonomic notes

We largely adopted species concepts of well known publications dealing w ith Tethyan and Midway fauna (Plummer, 1926; Brotzen, 1948; Cushman, 1951; Leroy, 1953; Auberi and Berggren, 1976; Speijer, 1994). More detailed comparisons are made w ith the faunas described by Cushman (1951) (Gulf Coastal Region of the United States); Auberi and Berggren (1976) (Tunisia); Bou Dagher (1987, 1988) (Tunisia) and Speijer (1994) (Egypt). The most common species are illustrated in Plates 1-4.

Alabamina m idwayensis Brotzen1948 Alabamina midwayensis Brotzen, p. 99, pi. 16, Figs. 1-2. 1976 Alabamina midwayensis Brotzen in Auberi and Berggren, p. 428. pi. 8, Fig. 3.1994 Alabamina midwayensis Brotzen in Speijer, p. 114, pi. 3, Fig. 2.

Ammobaculites expansus Plummer1933 Ammobaculites expansus Plummer, p. 65, pi. 5, Figs. 4-6. 1951 Ammobaculites expansus Plummer in Cushman, p. 4, pi. 1, Figs. 5-7.

Note: All chambers are arranged in a tight and very strongly compressed coil.

Ammobaculites m idwayensis Plummer1933 Ammobaculites midwayensis Plummer, p. 63, pi. 5, Figs. 7-11.1951 Ammobaculites midwayensis Plummer in Cushman, p. 4, pi. 1, Figs. 8-12.

Anomalinoides lordi Bou Dagher1988 Anomalinoides lordi Bou Dagher, p. 139, pi. 3, Figs. 4-10.

Anomalinoides m idwayensis (Plummer)1926 Truncatulina midwayensis Plummer, p. 141, pi. 15, Fig. 3. 1951 Anomalina midwayensis (Plummer) in Cushman, p. 62, pi. 17, Figs. 17-19.1976 Anomalinoides midwayensis (Plummer) in Auberi and Berggren, p. 430, pi. 9, Fig. 3.

Anomalinoides umboniferus (Schwager)1833 Discorbis praecursoria var. umbonifera Schwager, p. 126, pi. 27, Fig. 14.1951 Anomalina umbonifera (Schwager) in Cushman, p. 62, pi. 17, Fig. 16.1976 Anomalinoides umbonifera (Schwager) in Aubert and Berggren, p. 430, pi. 9, Fig. 4.

Note: Only specimens in the adult stadium were distinguishable from A. midwayensis by the inflated last chambers. Juvenile specimens are probably lumped with A. midwayensis

Baggatella cf. coloradoensis Malumian1976 Baggatella cf. coloradoensis Malumian (1970) in Aubert and Berggren, p. 419, pi. 4, Fig. 8.


Bulimina gr. trigonalis (ten Dam)1944 Bulimina trigonalis ten Dam, p. 112, pi. 3, Figs. 16-17. 1976 Bulimina gr. trigonalis (ten Dam) in Aubert and Berggren, 1976, p. 423, pi. 5, Fig. 11.1987 Stainforthia troosteri (Drooger) in Bou Dagher, 1987, p. 145, pi. 3, Figs. 1-8.

Bulimina kugleri Cushman and Renz1942 Bulimina kugleri Cushman and Renz, p. 9, pi. 2, Fig. 9. 1951 Bulimina kugleri Cushman and Renz in Cushman, p. 41, pi. 11, Fig. 24.1976 Bulimina kugleri Cushman and Renz in Aubert and Berggren, p. 422, pi. 5, Fig. 6.

Bulimina m idw ayensis (Cushman and Parker)1936 Bulimina arkadelphiana var. midwayensis Cushman and Parker, p. 42, pi. 7, Figs. 9-10.1951 Bulimina arkadelphiana var. midwayensis Cushman and Parker in Cushman, p. 40, pi. 11, Figs. 25-26.1976 Bulimina midwayensis (Cushman and Parker) and in Aubert and Berggren, p. 422, pi. 5, Fig. 7.1994 Bulimina midwayensis (Cushman and Parker) in Speijer, p. 110, pi. 1, Fig. 9.

Bulimina ovata d'Orbigny1846 Bulimina ovata d'Orbigny, p. 185, pi. 11, Figs. 13-14. 1994 Bulimina quadrata-ovata plexus in Speijer, p. 110, pi. 1, Fig. 10.

Bulimina aff. strobila (Marie)aff. 1941 Bulimina strobila Marie, p. 265, pi. 32, Fig. 302.1976 Bulimina asperoaculeta Brotzen in Aubert and Berggren, p. 421, pl.5 Fig. 4.

Note: A distinct triangular species w ith irregular ridges at the margins.

Cibicidoides howelli (Toulmin)1941 Cibicides howelli Toulmin, p. 609, pi. 82, Figs. 16-18. 1951 Cibicides howelli Toulmin in Cushman, p. 67, pi. 19, Figs. 15-17.

Cibicidoides megaloperforatus (Said and Kenawy)1956 Cibicides megaloperforatus Said and Kenawy, p. 155, pi. 7, Fig. 13.2000 Cibicidoides megaloperforatus (Said and Kenawy) in Schnack, p. 51, pi. 8, Figs. 9-11.

Cibicidoides p ra ecu rso ria (Schwager)1883 Discorbina praecursoria Schwager, p. 125, pi. 27, Figs. 12-13.1951 Cibicidespraecursorius (Schwager) in Cushman, 1951, p. 65, pi. 19, Figs. 1-6.1976 Cibicidoides praecursoria (Schwager) in Aubert and Berggren, p. 432, pi. 10, Fig. 5.

Cibicidoides succedens (Brotzen)1948 Cibicides succedens Brotzen, p. 80, pi. 12, Figs. 1-2.1976 Cibicidoides succedens (Brotzen) in Aubert and Berggren, p. 432, pi. 11, Fig. 1.1994 Cibicidoides succedens (Brotzen) in Speijer, p. 114, pi. 2, Fig. i.

Frondicularia phosphatica Russo1934 Frondicularia phosphatica Russo, p. 358-359, pi. 16, Figs. 6- 8 , 12.

1976 Frondicularia phosphatica Russo in Aubert and Berggren, p. 414, pi. 2, Fig. 12.1994 Frondicularia phosphatica Russo in Speijer, p. 109, pi. 1, Fig. 4.

Note: Remarks: This is a very large and characteristic species w ith chevron-like chambers and strong vertical costae from the initial to apertural end. These elevated central ridge(s) on both sides of the test and occasional oblique costae leads to an easy identification of this large species. Frondicularia wanneri Nakkady, 1950 is considered to be a ju nior synonym.

Frondicularia aff. phosphatica (Russo)aff. 1934 Frondicularia phosphatica Russo, p. 358-359, pi. 16, Figs. 6-8, 12.non 1950 Palmula woodi Nakkady, p. 684, pi. 89, Fig. 24.1976 Palmula woodi Nakkady in Aubert and Berggren, p. 417, pi. 3, Fig. 12.

Note: The large (up to a few mm.) specimens encountered in the Sidi Nasseur section don 't display the typical elevated central vertical ridge(s) on both sides of the test in comparison to Frondicularia phosphatica bu t do have the additional oblique costae. Some specimens have a small elevation in the central part resembling to some exten t a central ridge. The species lack the typical triangular outline of Frondicularia nakkadyi Futyan, 1976 and is usually longer than broad. Furthermore, F. aff phosphatica distribution data display a similar (tem porary?) disappearance at the P-E boundary. It is possible tha t these specimens are shallower ecophenotypes of F. phosphatica lacking the central elevated ridge. This species does not resemble the illustrated holotype of Palmula woodi by Nakkady which belongs to the genus Palmula.

Gyroidinoidesgirardanus (Reuss)1851 Rotalina girardana Reuss, p. 73, pi. 5, Fig. 34.1976 Gyroidinoides subangulata (Plummer) in Aubert and Berggren, p. 429, pi. 8, Fig. 6.1994 Gyroidinoides girardanus (Reuss) in Speijer, p. 118, pi. 3, Fig. 3.

Haplophragmoides spp.1989 Trochammina walteri Grzybowski, p. 290, pi. 11, Fig. 31. 1927 Flaplophragmoides excavata Cushman and Waters, p. 82, pi. 10, Fig. 3.1976 Flaplophragmoides excavata/walteri Cushman/(W aters and Grzybowski) in Aubert and Berggren, p. 408, pi. 1, Figs. 4.

Note: This group is largely composed of Flaplophragmoides excavata/walteri. Evolute and involute forms have been grouped under this combined name due to poor preservation resulting from compression and deformation of the specimens.

Lagena spp.Note: The diverse lagenids are separated into 5 distinct groups depending on their type of test ornamentation. Lagena gr. hispid e.g. Lagena hispida Reuss

1863 Lagena hispida Reuss, p. 335, pi. 6, Figs. 11-I'd. Lagena gr. reticulate e.g. Lagena polygonissima Willems

1990 Lagena polygonissima Willems, p. 386, pi. 4, Fig. 22. Lagena gr. smooth e.g. Lagena laevis (Montagu)

1803 Vermiculum leave Montagu, p. 524 and pi. 1, Fig. 9 in Walker and Boys (1784).

Lagena gr. striate-costate e.g. Lagena sulcata (Walker and Jacob) & Lagena stria ta (Montagu)

1798 Serpula sulcata Walker and Jacob, p. 634, pi. 14, Fig. 5. 1803 Serpula striatum Montagu p. 523.

Lenticulina spp.Note: The diverse lenticulinids are identified at genus level, because differential preservation does not allow consistent identification of all specimens at species level. Two species are more common:Lenticulina m idw ayensis (Plummer)

1927 Cristellaria midwayensis Plummer, p. 95, pi. 13, Fig. 5. 1951 Robulus midwayensis (Plummer) in Cushman, p. 13, pi. 3, Figs. 14-17.1976 Lenticulina midwayensis (Plummer) in Aubert and Berggren, p. 414, pi. 2, Fig. 16.

Lenticulina pseudomamilligera (Plummer)1927 Cristellaria pseudomamilligera Plummer, p. 98, pi. 7, Fig. 11.

P. Stassen et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 317-318 (2012) 66-92

1951 Robuluspseudo-mamilligerus (Plummer) in Cushman, p. 13, pi. 4, Figs. 1-5.1976 Lenticulina pseudomamilligera (Plummer) in Aubert and Berggren, p. 415, pi. 3, Fig. 2.

Neoeponides elevatus (Plummer)1927 Truncatulina elevata Plummer, p. 142, pi. 11, Fig. 1.1951 Eponides elevatus (Plummer) in Cushman, p. 52, pi. 14, Figs. 18-19.1976 Eponides elevatus (Plummer) in Aubert and Berggren, p. 425, pi. 7, Fig. 2.

Nonionella insecta (Schwager)1833 Anomalina insecta Schwager, p. 128, pi. 28, Fig. 2.1951 Nonionella insecta (Schwager) in Cushman, p. 37, pi. 11, Fig. 1.1976 Nonionella insecta (Schwager) in Aubert and Berggren, p. 428, pi. 8, Fig. 1.

Pyramidulina latejugata (Gümbel)1868 Nodosaria latejugata Gümbel, p. 619, pi. 1, Fig. 32.1951 Nodosaria latejugata Gümbel in Cushman, p. 23, pi. 7, Figs. 1-2.1976 Nodosaria latejugata Gümbel in Aubert and Berggren, p. 412, pi. 2, Fig. 4.

Reophax sp.1 Lagenammina sp.1Spiroplectammina mexiaensis Lalicker

1935 Spiroplectammina mexiaensis Lalicker, p. 43, pi. 6, Figs. 5-6. 1951 Spiroplectammina mexiaensis Lalicker in Cushman, p. 6, pi. 1, Figs. 25-26.

Stainforthia gafsensis (Grignani and Cococcetta)1973 Pseudovirgulina gafsensis Grignani and Cococcetta, p. 311, pi. 1, Fig. 16-201987 Stainforthia gafsensis (Grignani and Cococcetta) in Bou Dagher, p. 134, pi. 1, Figs. 1-12.1976 Fursenkoina sp. 3 in Aubert and Berggren, p. 427, pi. 7, Fig. 8.

Stainforthia kamali Bou Dagher1987 Stainforthia kamali Bou Dagher, p. 139, pi. 1, Figs. 13-20, pi. 2, Figs. 15-19, pi. 3, Figs. 10, 12, 14.

Stainforthia solignaci (Grignani and Cococcetta)1973 Pseudovirgulina solignaci Grignani and Cococcetta, p. 309, pi. 1, Figs. 1-61987 Pseudovirgulina solignaci (Grignani and Cococcetta) in Bou Dagher, p. 142, pi. 2, Figs. 1-4, pi. 3, Figs. 9, 11, 13.1976 Fursenkoina sp. 1 in Aubert and Berggren, p. 426, pi. 7, Fig. 6.

Stainforthia? sp.1.Note: Specimens were always compressed and poorly preserved and co-occur w ith other Stainforthia spp. It is possible that they represent a different thin walled Stainforthia species or a compressed form of Stainforthia gafsensis.

Valvulineria? insueta (Cushman and Bermudez)1948 Valvulineria insueta Cushman and Bermudez, p. 85, pi. 14, Figs. 7-9.1951 Valvulineria insueta Cushman and Bermudez in Cushman, 1951, p. 51, pi. 23, Figs. 2-4.1994 Valvulineria? insueta (Cushman and Bermudez) in Speijer, p. 112, pi. 3, Fig. 4.

Valvulineria scrobiculata (Schwager)1833 Anomalina scrobiculata Schwager, p. 129, pi. 29, Fig. 18. 1953 Valvulineria scrobiculata (Schwager) in Leroy, 1953, p. 53, pi. 9, Figs. 18-20.1994 Valvulineria scrobiculata (Schwager) in Speijer, p. 112, pi. 4, Figs. 1-3.

Note: This species name seems to be used for a different taxon in Aubert and Berggren (1976); we followed the concept of Leroy (1953).

Appendix B. Supplementary data

Supplementary data to this article can be found online at doi:10. 1016/j.palaeo.2011.12.011.

References

Agnini, C., Fornaciari, E., Raffi, I., Rio, D., Röhl, U., Westerhold, T., 2007. High-resolution nannofossil biochronology of middle Paleocene to early Eocene at ODP Site 1262: implications for calcareous nannoplankton evolution. Marine Micropaleontology 64,215-248.

Albani, A.D., Yassini, I., 1989. Taxonomy and distribution of shallow-water lagenid- Foraminiferida from the Southeastern Coast of Australia. Australian Journal of Marine & Freshwater Research 40, 369-401.

Aubert, J., Berggren, W.A., 1976. Paleocene benthic foraminiferal biostratigraphy and paleoecology of Tunisia. Bulletin du Centre de Recherche de Pau 10, 379-469.

Aubry, M.-P., 1999. Late Paleocene-early Eocene sedimentary history in western Cuba: implications for the LPTM and for regional tectonic history. In: Fluegeman, R.H., Aubry, M.-P. (Eds.), Lower Paleogene Biostratigraphy of Cuba: Micropaleontology, 45, pp. 5-18. suppl. 2.

Aubry, M.-P., Ouda, K., Dupuis, C., Berggren, W.A., Van Couvering, J.A., Ali, J., Brinkhuis, H., Gingerich, P.D., Heilmann-Clausen, C., Hooker, J., Kent, D.V., King, C., Knox, R., Laga, P., Molina, E., Schmitz, B., Steurbaut, E., Ward, D., 2007. The Global Standard Stratotype-section and Point (GSSP) for the base of the Eocene Series in the Daba- biya section (Egypt). Episodes 30, 271-286.

Bensalem, H., 2002. The Cretaceous-Paleogene transition in Tunisia: general overview. Palaeogeography, Palaeoclimatology, Palaeoecology 178,139-143.

Berggren, W.A., 1974. Late Paleocene-Early Eocene benthonic foraminiferal biostratigraphy and paleoecology of Rockall Bank. Micropaleontology 20,426-448.

Berggren, W.A., Pearson, P.N., 2005. A revised tropical to subtropical Paleogene planktonic foraminiferal zonation. Journal of Foraminiferal Research 35, 279-298.

Berkeley, A., Perry, C.T., Smithers, S.G., Horton, B.P., Taylor, K.G., 2007. A review of the ecological and taphonomic controls on foraminiferal assemblage development in intertidal environments. Earth-Science Reviews 83, 205-230.

Bernhard, J.M., Sen Gupta, B.K., 1999. Foraminifera of oxygen depleted environments. In: Sen Gupta, B.K. (Ed.), Modern Foraminifera. Kluwer Academic Publishers, London, pp. 201-216.

Bolle, M.P., Adatte, T., Keller, G., von Salis, K., Burns, S., 1999. The Paleocene-Eocene transition in the southern Tethys (Tunisia): climatic and environmental fluctuations. Bulletin De La Société Géologique De France 170, 661-680.

Boltovskoy, E., Totah, V.l., 1992. Preservation index and preservation potential of some foraminiferal species. Journal of Foraminiferal Research 22, 267-273.

Bou Dagher, M.K., 1987. The Stainforthiidae (Foraminifera) in the Late Paleocene and Early Eocene o f Tunisia. Bulletin des Centres de Recherches Exploration-Production Elf Aquitaine 11 ,133-152.

Bou Dagher, M.K., 1988. Shallow-water smaller benthic foraminifera from the lower Eocene of the eastern and southern Tunisia. Bulletin des Centres de Recherches Exploration-Production Elf Aquitaine 12,129-142.

Bralower, T.J., 2002. Evidence of surface water oligotrophy during the Paleocene- Eocene thermal maximum: nannofossil assemblage data from Ocean Drilling Program Site 690, Maud Rise, Weddell Sea. Paleoceanography 17 PAI 023.

Brotzen, F., 1948. The Swedish Paleocene and its foraminiferal fauna. Sveriges Geolo- giska Undersökning, 42 .140 pp.

Burollet, P.F., 1967. General geology of Tunisia. Petroleum Exploration Society of Libya. 9th Annual Field Conference, Tripoli, pp. 51-58.

Cramer, B.S., Wright, J.D., Kent, D.V., Aubry, M.P., 2003. Orbital climate forcing of 013C excursions in the late Paleocene-early Eocene (chrons C24n-C25n). Paleoceanography 18 PAI 097.

Crouch, E.M., Heilmann-Clausen, C., Brinkhuis, H., Morgans, H.E.G., Rogers, K.M., Egger, H., Schmitz, B., 2001. Global dinoflagellate event associated with the late Paleocene thermal maximum. Geology 29, 315-318.

Cushman, J.A., 1951. Paleocene foraminifera of the Gulf Coastal Region of the United States and adjacent areas. Geological Survey Professional Paper, 232. 75 pp.

Dercourt, J., Gaetani, M., Vrielynck, B., Barrier, E., Biju-Duval, B., Brunet, M.F., Cadet, J.P., Crasquin, S., Sandulescu, M., 2000. Atlas Peri-Tethys, palaeogeographical maps. Commission de la carte Géologique du Monde/Commission for the Geological Maps of the World, Paris, p. 269.

Dickens, G.R., Castillo, M.M., Walker, J.C.G., 1997. A blast of gas in the latest Paleocene: simulating first-order effects of massive dissociation of oceanic methane hydrate. Geology 25, 259-262.

Diz, P., Frances, G., 2008. Distribution of live benthic foraminifera in the Ria de Vigo (NW Spain). Marine Micropaleontology 66,165-191.

Diz, P., Frances, G., Roson, G., 2006. Effects of contrasting upwelling-downwelling on benthic foraminiferal distribution in the Ria de Vigo (NW Spain). Journal of Marine Systems 60 ,1 -18 .

Duijnstee, I.A.P., de Nooijer, L.J., Ernst, S.R., van der Zwaan, G.J., 2005. Population dynamics of benthic shallow-water foraminifera: effects of a simulated marine snow event. Marine Ecology Progress Series 285, 29-42.

Dupuis, C., Aubry, M.-P., Steurbaut, E., Berggren, W.A., Ouda, K., Magioncalda, R., Cramer, B.S., Kent, D.V., Speijer, R.P., Heilmann-Clausen, C., 2003. The Dababiya Quarry section: lithostratigraphy, clay mineralogy, geochemistry and paleontology. Micropaleontology 49,41 -59.

Ernst, S.R., Guasti, E., Dupuis, C., Speijer, R.P., 2006. Environmental perturbation in the southern Tethys across the Paleocene/Eocene boundary (Dababiya, Egypt): foraminiferal and clay mineral records. Marine Micropaleontology 60, 89-111.


Gingerich, P.D., 2006. Environment and evolution through the Paleocene-Eocene thermal maximum. Trends in Ecology & Evolution 21, 246-253.

Guasti, E., Speijer, R.P., 2007. The Paleocene-Eocene Thermal Maximum in Egypt and Jordan: an overview of the planktic foraminiferal record. In: Monechi, S., Coccioni, R., Rampino, M.R. (Eds.), Large Ecosystem Perturbations: Causes and Consequences: Geological Society of America Special Paper, 424, pp. 53-67.

Guasti, E., Speijer, R.P., 2008. Acarinina multicamerata n. sp. (Foraminifera): a new marker for the Paleocene-Eocene thermal maximum. Journal of Micropalaeontology 27,5-12.

Guasti, E., Kouwenhoven, T.J., Brinkhuis, H., Speijer, R.P., 2005. Paleocene sea-level and productivity changes at the southern Tethyan margin (El Kef, Tunisia). Marine Micropaleontology 5 5 ,1 -17 .

Guasti, E., Speijer, R.P., Brinkhuis, H., Smit, J., Steurbaut, E., 2006. Paleoenvironmental change at the Danian-Selandian transition in Tunisia: Foraminifera, organic- walled dinoflagellate cyst and calcareous nannofossil records. Marine Micropaleontology 59, 210-229.

Hammer, O., Harper, D.A.T., Ryan, P.D., 2001. PAST: paleontological statistics software package for education and data analysis. Palaeontologica Electrónica 9.

Hay, W.W., 1995. Paleoceanography of marine organic-carbon-rich sediments. In: Hue, A-Y. (Ed.), Paleogeography, Paleoclimate and Source Rocks: American Association of Petroleum Geologists Studies in Geology, 40, pp. 21-59.

Hermelin, J.O.R., Malmgren, B.A., 1980. Multivariate analysis of environmentally controlled variation in Lagena: late Maastrichtian, Sweden. Cretaceous Research 1,193-206.

John, C.M., Bohaty, S.M., Zachos, J.C., Sluijs, A., Gibbs, S., Brinkhuis, H., Bralower, T.J., 2008. North American continental margin records of the Paleocene-Eocene thermal maximum: implications for global carbon and hydrological cycling. Paleoceanography 23 PA2217.

Jones, R.W., Charnock, M.A., 1985. Morphogroups of agglutinating foraminifera. Their life positions and feeding habits and potential applicability in (paleo)ecological studies. Revue de Paléobiologie 4, 311-320.

Jorissen, F.J., 1999. Benthic foraminiferal microhabitats below the sediment-water interface. In: Sen Gupta, B.K. (Ed.), Modern Foraminifera. Kluwer Academic Publishers, London, pp. 161-179.

Jorissen, F.J., de Stigter, H.C., Widmark, J.G.V., 1995. A conceptual model explaining benthic foraminiferal microhabitats. Marine Micropaleontology 26, 3-15.

Kaiho, K., 1994. Benthic foraminiferal dissolved-oxygen index and dissolved-oxygen levels in the modern ocean. Geology 22, 719-722.

Kellough, G.E., 1965. Paleoecology of the Foraminiferida of the Wills Point Formation (Midway Group) in northeast Texas. Transactions—Gulf Coast Association of Geological Societies 15, 73-153.

Kelly, D.C., Bralower, T.J., Zachos, J.C., Silva, I.P., Thomas, E., 1996. Rapid diversification of planktonic foraminifera in the tropical Pacific (ODP Site 865) during the late Paleocene thermal maximum. Geology 24,423-426.

Kelly, D.C., Bralower, T.J., Zachos, J.C., 1998. Evolutionary consequences of the latest Paleocene thermal maximum for tropical planktonic foraminifera. Palaeogeography, Palaeoclimatology, Palaeoecology 141,139-161.

Kennett, J.P., Stott, L.D., 1991. Abrupt deep-sea warming, palaeoceanographic changes and benthic extinctions at the end of the Palaeocene. Nature 353, 225-229.

Kouwenhoven, T.J., Speijer, R.P., van Oosterhout, C.W.M., van der Zwaan, G.J., 1997. Benthic foraminiferal assemblages between two major extinction events: the Paleocene El Kef section, Tunisia. Marine Micropaleontology 29 ,105-112.

Leroy, L.W., 1953. Biostratigraphy of the Maqfi section, Egypt. Geological Society of America, Memoir, 54. 73 pp.

Loubere, P., Fariduddin, M., 1999. Benthic foraminifera and the flux of organic carbon to the seabed. In: Sen Gupta, B.K. (Ed.), Modern Foraminifera. Kluwer Academic Publishers, London, pp. 181-199.

Luciani, V., Giusberti, L, Agnini, C., Backman, J., Fomaciari, E., Rio, D., 2007. The Paleocene- Eocene thermal maximum as recorded by Tethyan planktonic foraminifera in the For- ada section (northern Italy). Marine Micropaleontology 64,189-214.

Luger, P., 1985. Stratigraphie der marinen Oberkreide und des Alttertiärs im südwestlichen Obernil-Becken (SW Ägypten) unter besonderer Berücksichtigung der Mikropaläontologie, Palökologie und Paläogeographie., Berliner Geowissenschaf- tliche Abhandlungen. Reihe A. Geologie und Paläontologie, 63, p. 151.

Martini, E., 1971. Standard Tertiary and Quaternary calcareous nannoplankton zonation. Proceedings of the 2nd Planktonic Conference, Roma, pp. 739-785.

Miller, K.G., Wright, J.D., Browning, J.V., 2005. Visions of ice sheets in a greenhouse world. Marine Geology 217, 215-231.

Morsi, A.-M.M., Speijer, R.P., Stassen, P., Steurbaut, E., 2011. Shallow marine ostracode turnover in response to environmental change during the Paleocene-Eocene thermal maximum in northwest Tunisia. Journal of African Earth Sciences 59,243-268.

Murray, J.W., 2001. The niche of benthic foraminifera, critical thresholds and proxies. Marine Micropaleontology 41 ,1 -7 .

Murray, J.W., 2006. Ecology and Applications of Benthic Foraminifera. Cambridge University Press, Cambridge.

Murray, J.W., Alve, E., 1999. Natural dissolution of modern shallow water benthic foraminifera: taphonomic effects on the palaeoecological record. Palaeogeography, Palaeoclimatology, Palaeoecology 146,195-209.

Nguyen, T.M.P., Petrizzo, M.R., Speijer, R.P., 2009. Experimental dissolution of a fossil foraminiferal assemblage (Paleocene-Eocene Thermal Maximum, Dababiya, Egypt): implications for paleoenvironmental reconstructions. Marine Micropaleontology 73, 241-258.

Norris, R.D., 1996. Symbiosis as an evolutionary innovation in the radiation of Paleocene planktic foraminifera. Paleobiology 22, 461-480.

Orue-Etxebarria, X., Pujalte, V., Bemaola, G., Apellaniz, E., Baceta, J.I., Payros, A, Nunez-Betelu, K., Serra-Kiel, J., Tosquella, J., 2001. Did the Late Paleocene thermal maximum affect the evolution of larger foraminifers? Evidence from calcareous plankton of the Campo Section (Pyrenees, Spain). Marine Micropaleontology 41,45-71.

Peypouquet, J.P., Grousset, F., Mourguiart, P., 1986. Paleo-oceanography of the Mesogean Sea based on ostracods of the northern Tunisian continental shelf between the late Cretaceous and early Paleogene. Geologische Rundschau 75, 159-174.

Plummer, H.J., 1926. Foraminifera of the Midway Formation in Texas. Texas University, Bulletin, 2644, p. 206.

Raffi, I., Backman, J., Pälike, H., 2005. Changes in calcareous nannofossil assemblages across the Paleocene/Eocene transition from the paleo-equatorial Pacific Ocean. Palaeogeography, Palaeoclimatology, Palaeoecology 226, 93-126.

Röhl, U., Westerhold, T., Bralower, T.J., Zachos, J.C., 2007. On the duration o f the Paleocene-Eocene thermal maximum (PETM). Geochemistry, Geophysics, Geosystems 8, Q12002.

Saint-Marc, P., 1992. Biogeographic and bathymetric distribution o f benthic foraminifera in Paleocene El-Haria Formation o f Tunisia. Journal o f African Earth Sciences 15, 473-487.

Saint-Marc, P., Berggren, W.A., 1988. A quantitative analysis of Paleocene benthic foraminiferal assemblages in Central Tunisia. Journal of Foraminiferal Research 18, 97-113.

Scheibner, C., Speijer, R.P., 2008. Late Paleocene-early Eocene Tethyan carbonate platform evolution — a response to long- and short-term paleoclimatic change. Earth- Science Reviews 90, 71-102.

Scheibner, C., Speijer, R.P., Marzouk, AM., 2005. Turnover of larger foraminifera during the Paleocene-Eocene Thermal Maximum and paleoclimatic control on the evolution of platform ecosystems. Geology 33,493-496.

Schmitz, B., Pujalte, V., 2003. Sea-level, humidity, and land-erosion records across the initial Eocene thermal maximum from a continental-marine transect in northern Spain. Geology 31, 689-692.

Schnack, K., 2000. Biostratigraphie und fazielle Entwicklung in der Oberkreide und im Alttertiär im Bereich der Kharga Schwelle. Westliche Wüste, SW Ägypten, Berichte, Fachbereich Geowissenschaften, Universität Bremen Nr. 151, Bremen, p. 142.

Sluijs, A., Bowen, G.J., Brinkhuis, H., Lourens, L.J., Thomas, E., 2007. The Paleocene- Eocene Thermal Maximum super greenhouse: biotic and geochemical signatures, age models and mechanisms of global change. In: Williams, M., Haywood, AM., Gregory, F.J., Schmidt, D.N. (Eds.), Deep-Time Perspectives on Climate Change: Marrying the Signal from Computer Models and Biological Proxies. The Micropalaeontological Society, Special Publications, London, pp. 323-350.

Sluijs, A., Brinkhuis, H., Crouch, E.M., John, C.M., Handley, L., Munsterman, D., Bohaty, S.M., Zachos, J.C., Reichart, G.J., Schouten, S., Pancost, R.D., Sinninghe Damsté, J.S., Welters, N.L.D., Lotter, A.F., Dickens, G.R., 2008. Eustatic variations during the Paleocene-Eocene greenhouse world. Paleoceanography 23 PA4216.

Speijer, R.P., 1994. Extinction and recovery patterns in benthic foraminiferal paleocom- munities across the Cretaceous/Paleogene and Paleocene/Eocene boundaries. Geológica Ultraiectina 124,191 Utrecht.

Speijer, R.P., Morsi, A.-M.M., 2002. Ostracode turnover and sea-level changes associated with the Paleocene-Eocene thermal maximum. Geology 30, 23-26.

Speijer, R.P., Schmitz, B., 1998. A benthic foraminiferal record of Paleocene sea level and trophic/redox conditions at Gebel Aweina, Egypt. Palaeogeography, Palaeoclimatology, Palaeoecology 137, 79-101.

Speijer, R.P., Wagner, T., 2002. Sea-level changes and black shales associated with the late Paleocene thermal maximum: organic-geochemical and micropaleontologic evidence from the southern Tethyan margin (Egypt-Israel). Geological Society of America Special Paper 356, 533-549.

Speijer, R.P., van der Zwaan, G.J., Schmitz, B., 1996. The impact of Paleocene/Eocene boundary events on middle neritic benthic foraminiferal assemblages from Egypt. Marine Micropaleontology 28, 99-132.

Speijer, R.P., Schmitz, B., van der Zwaan, G.J., 1997. Benthic foraminiferal extinction and repopulation in response to latest Paleocene Tethyan anoxia. Geology 25, 683-686.

Stassen, P., Dupuis, C., Morsi, A.-M.M., Steurbaut, E., Speijer, R.P., 2009. Reconstruction of a latest Paleocene shallow-marine eutrophic paleoenvironment at Sidi Nasseur (Central Tunisia) based on foraminifera, ostracoda, calcareous nannofossils and stable isotopes (Ô13C, ÔlsO). Geológica Acta 7, 93-112.

Steineck, P.L., Thomas, E., 1996. The latest Paleocene crisis in the deep sea: ostracode succession at Maud Rise, Southern Ocean. Geology 24, 583-586.

Thomas, E., 1998. Biogeography of the Late Paleocene benthic foraminiferal extinction. In: Aubry, M.-P., Lucas, S.H., Berggren, W.A. (Eds.), Late Paleocene-Early Eocene Climatic and Biotic Events in the Marine and Terrestrial Records. Columbia University Press, New York, pp. 214-243.

Tjalsma, R.C., Lohmann, K.C., 1983. Paleocene-Eocene bathyal and abyssal benthic foraminifera from the Atlantic region. Micropaleontology Special Publication4 .1 -9 1 .

Tyson, R.V., Pearson, T.H., 1991. Modern and ancient continental shelf anoxia: an overview. In: Tyson, R.V., Pearson, T.H. (Eds.), Modern and Ancient Continental Shelf Anoxia: Geological Society Special Publication, 58, pp. 1-24. London.

van der Zwaan, G.J., Jorissen, F.J., 1991. Biofacial patterns in river-induced shelf anoxia. In: Tyson, R.V., Pearson, T.H. (Eds.), Modern and Ancient Continental Shelf Anoxia: Geological Society Special Publication, 58, pp. 65-82. London.

van der Zwaan, G.J., Jorissen, F.J., Destigter, H.C., 1990. The depth dependency of planktonic benthic foraminiferal ratios - constraints and applications. Marine Geology95.1-16.

van der Zwaan, G.J., Duijnstee, LAP., den Dulk, M., Ernst, S.R., Jannink, N.T., Kouwenhoven, T.J., 1999. Benthic foraminifers: proxies or problems? A review of paleocological concepts. Earth-Sdence Reviews 46,213-236.

von Salis, K., Monechi, S., Bybell, L.M., Self-Trail, J., Young, J., 2000. Remarks on the calcareous nannofossil markers Rhomboaster and Tribrachiatus around the Paleocene/ Eocene boundary. GFF 122,138-140.


Zachos, J., Pagani, M., Sloan, L., Thomas, E., Billups, K., 2001. Trends, rhythms, and aber- Zaïer, A., Beji-Sassi, A., Sassi, S., Moody, R.T.J., 1998. Basin evolution and deposition rations in global climate 65 Ma to present Science 292, 686-693. during the Early Paleogene in Tunisia. In: Moody, R.T.J., Clark-Lowes, D.D. (Eds.),

Zachos, J.C., Dickens, G.R., Zeebe, R.E., 2008. An early Cenozoic perspective on green- Petroleum Geology of North Africa: Geological Society Special Publication, 132,house warming and carbon-cycle dynamics. Nature 451, 279-283. pp. 375-393. London.

Documents

Perturbation of a Tethyan coastal environment during the ... · calcareous taxa (Flaplophragmoides) and rare planktic foraminifera, indicating a slightly hypersaline eutro- phic inner