A shallow water record of the onset of the Messinian salinity crisis in the Adriatic foredeep (Legnagnone section, Northern Apennines)

(1)

AUTHOR QUERY FORM

Journal: PALAEO

Please e-mail or fax your responses and any corrections to:

Krishnaveni Kunchala

E-mail:

Corrections.ESCH@elsevier.spitech.com

Fax: +1 619 699 6721

Article Number: 6504

Dear Author,

Please check your proof carefully and mark all corrections at the appropriate place in the proof (e.g., by using on-screen

anno-tation in the PDF

ﬁle) or compile them in a separate list. Note: if you opt to annotate the ﬁle with software other than Adobe

Reader then please also highlight the appropriate place in the PDF

ﬁle. To ensure fast publication of your paper please return

your corrections within 48 hours.

For correction or revision of any artwork, please consult

http://www.elsevier.com/artworkinstructions.

Any queries or remarks that have arisen during the processing of your manuscript are listed below and highlighted by

ﬂags in

the proof. Click on the

‘Q’ link to go to the location in the proof.

Location in article

Query / Remark:

click on the Q link to go

Please insert your reply or correction at the corresponding line in the proof

Q1

Please conﬁrm that given names and surnames have been identiﬁed correctly.

Q2

Citation

“Reuss, 1850” has not been found in the reference list. Please supply full details for this

reference.

Q3

Citation

“Baird, 1850” has not been found in the reference list. Please supply full details for this

reference.

Q4

Citation

“Roemer, 1838” has not been found in the reference list. Please supply full details for this

reference.

Q5

Citation

“Kollmann, 1960” has not been found in the reference list. Please supply full details for this

reference.

Q6

Citation

“Ruggieri, 1953” has not been found in the reference list. Please supply full details for this

reference.

Q7

Citation

“Capeder, 1902” has not been found in the reference list. Please supply full details for this

reference.

Q8, Q9

Citation

“Müller, 1894” has not been found in the reference list. Please supply full details for this

reference.

Q10

Please check the country name provided.

Q11

The citation

“Krijgsman et al., 1999a” has been changed to match the author name/date in the

reference list. Please check here and in subsequent occurrences, and correct if necessary.

Q12, Q17

Citation

“Sierro et al., 2003” has not been found in the reference list. Please supply full details for

this reference.

Q13

Citation of Fig. 13 is found in the text but the corresponding artwork is missing. Please check and

amend if necessary.

Q14

Citation

“de Stitger et al., 1998” has not been found in the reference list. Please supply full details

for this reference.

Our reference: PALAEO 6504

P-authorquery-v11

(2)

Q15

Citation

“Van der Laan et al. (2006)” has not been found in the reference list. Please supply full

details for this reference.

Q16

Citation

“Manzi et al., 2010” has not been found in the reference list. Please supply full details for

this reference.

Q18

Citation

“Krijgsman and Hilgen, 1999” has not been found in the reference list. Please supply full

details for this reference.

Q19

Uncited references: This section comprises references that occur in the reference list but not in the

body of the text. Please position each reference in the text or, alternatively, delete it. Thank you.

Q20

Please check the page range in Ref. (Ruggieri, 1964).

Q21

Please check the author names in Ref. (Suc et al., 1995b).

Q22

Supplementary caption was not provided. Please check suggested data if appropriate and correct if

necessary.

Please check this box if you have no

corrections to make to the PDF

ﬁle.

□

Thank you for your assistance.

Our reference: PALAEO 6504

P-authorquery-v11

(3)

UNCORRECTED PR

OOF

1

Highlights

2 Palaeogeography, Palaeoclimatology, Palaeoecology xxx (2013) xxx_{– xxx}

4

5 A shallow water record of the onset of the Messinian salinity crisis in

6 the Adriatic foredeep (Legnagnone section, Northern Apennines)

7

8 Rocco Gennaria,b,_⁎_{, Vinicio Manzi}a,b_{, Lorenzo Angeletti}c_{, Adele Bertini}d,e_{, Ulderico Bif}_ﬁf_{, Alessandro Ceregato}c_{, Costanza Faranda}g_, 9 Elsa Gliozzig_{, Stefano Lugli}h_{, Elena Menichetti}d_{, Antonietta Rosso}i_{, Marco Roveri}a,b_{, Marco Taviani}c

10

11 a_{Università degli Studi di Parma}_{, V.le G.P. Usberti 157/A, 43100 Parma,}_Italy

12 b_{ALP Laboratory}_{, Via Madonna dei Boschi 76, 12016 Peveragno (CN),}_Italy

13 c_ISMAR-CNR_{, Via Gobetti 101, 40129 Bologna,}_Italy

14 d_{Università degli Studi di Firenze}_{, Via La Pira 4, 50121 Firenze,}_Italy

15 e_{IGG-CNR, Sezione di Firenze}_{, Via G. La Pira, 4, 50121 Firenze,}_Italy

16 f_ENI-AGIP_{, Italy}

17 g_{Università Roma Tre}_{, Largo S. Leonardo Murialdo, 1, I-00146, Roma,}_Italy

18 h_{Università degli Studi di Modena e Reggio Emilia}_{, Piazza S. Eufemia 19, 41100 Modena,}_Italy

19 i_{Università di Catania}_{, Corso Italia, 55, 95129, Catania,}_Italy

20 21 22 • Shallow water record of the pre-evaporitic/evaporitic transition of the MSC.

23 • High-resolution bio-magnetostratigraphic framework.

24 • Multi-proxy palaeoenvironmental and palaeoclimatic reconstructions.

25 26

Palaeogeography, Palaeoclimatology, Palaeoecology xxx (2013) xxx

Contents lists available atSciVerse ScienceDirect

Palaeogeography, Palaeoclimatology, Palaeoecology

j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / p a l a e o

(4)

Q22

Supplementary material.

(5)

UNCORRECTED PR

OOF

1

A shallow water record of the onset of the Messinian salinity crisis in the Adriatic

2

foredeep (Legnagnone section, Northern Apennines)

3

Rocco

Q1

Gennari

a,b,

⁎

,

Vinicio

Manzi

a,b

,

Lorenzo

Angeletti

c

,

Adele

Bertini

d,e

,

Ulderico

Bif

ﬁ

f

,

4

Alessandro

Ceregato

c

,

Costanza

Faranda

g

,

Elsa

Gliozzi

g

,

Stefano

Lugli

h

,

Elena

Menichetti

d

,

5

Antonietta

Rosso

i

,

Marco

Roveri

a,b

,

Marco

Taviani

c

6 a_{Università degli Studi di Parma}_{, V.le G.P. Usberti 157/A, 43100 Parma,}_Italy 7 b_{ALP Laboratory}_{, Via Madonna dei Boschi 76, 12016 Peveragno (CN),}_Italy

8 c

ISMAR-CNR, Via Gobetti 101, 40129 Bologna,Italy

9 d

Università degli Studi di Firenze, Via La Pira 4, 50121 Firenze,Italy

10 e

IGG-CNR, Sezione di Firenze, Via G. La Pira, 4, 50121 Firenze,Italy

11 f

ENI-AGIP, Italy

Q10

12 g

Università Roma Tre, Largo S. Leonardo Murialdo, 1, I-00146, Roma,Italy

13 h_{Università degli Studi di Modena e Reggio Emilia}_{, Piazza S. Eufemia 19, 41100 Modena,}_Italy

14 i

Università di Catania, Corso Italia, 55, 95129, Catania,Italy

15 16

a b s t r a c t

a r t i c l e i n f o

17 Article history: 18 Received 16 July 2012

19 Received in revised form 2 May 2013

20 Accepted 7 May 2013 21 Available online xxxx 22 23 24 25 Keywords: 26 Messinian 27 Northern Apennine 28 Multi-proxy palaeoenvironmental 29 reconstruction 30 Palaeoclimate 31 Palaeoceanography 32 Integrated stratigraphy 33 The Legnagnone section (North-eastern Apennines) represents one of the few shallow water records of the

34 onset of the Messinian salinity crisis. Here we present a detailed description of a ~200kyr time interval

35 encompassing the pre-/syn-evaporitic transition based on a multidisciplinary approach, integrating

sedimen-36 tological, bio-magnetostratigraphical, palaeontological and stable isotope data. Such a shallow water setting

37 is potentially more sensitive to the palaeoenvironmental change leading to the MSC than the more often

38 studied deeper Mediterranean basin. The aquatic palaeoenvironmental reconstruction proposed here is

39 based on the study of foraminifer, ostracod and mollusc assemblages. It depicts a change from infralittoral

40 (20–50 m) to inner circalittoral environment (60–100 m) that, since 6.12 Ma, was progressively affected

41 by a reduction of oxygen at the seaﬂoor punctuated by short-lived anoxic events. At least three cooling

42 events have been recognized on the basis of relative abundance data in mid to high altitude pollen, which,

43 before 6.03 Ma, are in phase with abundance peaks of Turborotalia spp., a taxon indicating eutrophic and

44 cool surface waters. The absence of stress-tolerant benthic foraminifers during these peaks points to strong

45 ventilation episodes triggered by a generally cooler climate. The proximity of a deltaic system and the

conse-46 quent riverine input probably caused a salinity decrease of the surface waters, hindering the proliferation of

47 planktonic foraminifers in the water column, which prevalently occur in short inﬂuxes and disappear at ca.

48 6 Ma. Our results suggest that the onset of the crisis occurred during a phase of relative sea level high

49 stand, whereas no evidences of sea level drop can be envisaged. The palaeoclimatic reconstruction based

50 on palynological data indicates the dominance of a“subtropical humid forest” vegetation type, where fresh

51 water swamps are well represented. From 6.03 Ma onward, the transition to the salinity crisis is marked

52 by more pronounced cyclical expansions of the temperate broad-leaved deciduous forest, along with

herba-53 ceous taxa. The establishment of the strongly evaporative condition at the crisis onset is not associated with

54 major vegetational changes towards drier conditions, but linked to a sudden increase ofδ18_{O and the}

disap-55 pearance of benthic foraminifers just prior to the deposition of the 1st laminated carbonate, which represents

56 the base of the Primary Lower Gypsum unit.

58 59 60

61

62 1. Introduction

63 A better appreciation of causal processes leading to the onset of

64 the Messinian salinity crisis (MSC) necessarily requires a thorough

65

understanding of the palaeoenvironmental evolution responsible of this

66

event. The broadly acceptedCIESM (2008)scenario describes a

synchro-67

nous onset of the MSC in all the geological settings of the Mediterranean

68

at 5.96 Ma (Krijgsman et al., 1999) and indicates that the Primary Lower Q11

69

Gypsum unit (PLG;Roveri et al., 2008) was deposited in shallower

oxy-70

genated marginal settings. At the same time, deeper settings experienced

71

the deposition of organic-rich anoxic deposits, starting to receive clastic

72

evaporite deposits only later, derived from the dismantlement and the

Palaeogeography, Palaeoclimatology, Palaeoecology xxx (2013) xxx–xxx

⁎ Corresponding author at: Università degli Studi di Parma, V.le G.P. Usberti 157/A, 43100 Parma, Italy. Tel.: + 39 0521905324.

E-mail address:rocco.gennari@gmail.com(R. Gennari).

Contents lists available atSciVerse ScienceDirect

Palaeogeography, Palaeoclimatology, Palaeoecology

j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / p a l a e o

(6)

UNCORRECTED PR

OOF

73 en-masse resedimentation of the PLG to form the Resedimented Lower

74 Gypsum (RLG;Roveri et al., 2008). The Messinian pre-evaporitic timing,

75 palaeoceanographic and palaeoenvironmental history are relatively

76 well known regarding its distal, slope to basin settings (Greece,

77 Krijgsman et al., 2004; Cyprus, Kouwenhoven et al., 2006; Sicily,

78 Hilgen and Krijgsman, 1999; Bellanca et al., 2001 and

Blanc-79 Valleron et al., 2002; Northern Apennines, Roveri et al., 2006;

80 Manzi et al., 2007). Usually four main palaeoenvironmental steps

81 are envisaged from the base of the Messinian to the onset of the

82 MSC, describing the progressive establishment of stressing

condi-83 tion both at the seaﬂoor and in the water column (Kouwenhoven

84 et al., 2006). Unfortunately, these studies generally refer to

succes-85 sions older than 6.5–6.3 Ma and/or that were deposited in basins

86 lacking the PLG unit and characterized by the presence of limestone

87 or diatomite, usually alternated with organic-rich clay devoid of

cal-88 careous microfossils, thus making the de_{ﬁnition of the salinity crisis}

89 onset more problematic (Gennari et al., 2009; Lugli et al., 2010;

90 Manzi et al., 2011).

91 On the contrary, pre-evaporitic shallow water marine settings are

92 less documented (Los Yesos;Goubert et al., 2001) in spite of their

rel-93 evance as monitors of the changes in surface water masses, coastal

94 habitats and sea level changes.

95 Here we discuss the Legnagnone section, a rare record of the pre-/

96 syn-evaporitic transition in a coastal setting, through a comprehensive

97 approach integrating sedimentological, bio-magnetostratigraphical,

98 palaeontological and geochemical data.

99 2. Geological setting

100 The Legnagnone section sits on the allochtonous“Val Marecchia”

101 Ligurian nappe, a portion of the Apennines accretionary prism, which

102 migrated since the early Oligocene on top of the deformed

autochtho-103 nous Umbro-Marchean-Romagna unit (Fig. 1a, b).

104 Up to the Messinian times, the Val Marecchia terrains were still

105 located in a more inner position, closer to the Apennines divide.

106 Consequently, the Messinian succession of the Val Marecchia was

107 deposited in a coastal shallow environment.

108 The situation therefore differs from the well-known Vena del

109 Gesso basin, where the Messinian Evaporites were deposited on

110 top of an open-marine succession laying on the autochthonous

111 Umbro-Marchean-Romagna unit (Vai, 1997; Roveri et al., 2003).

112 In the Val Marecchia, the Tortonian–Messinian succession rests

un-113 conformably and relatively undeformed on the Miocene calcarenites

114 of the“San Marino Fm”. This succession (Fig. 2) consists from the base

115 ofﬂuvio-deltaic conglomerates and sandstones of the “Aquaviva Fm.”

116 abruptlyﬁning-upward and grading in the marly unit of the “Casa i

117 Gessi Fm.” (see Ruggieri, 1958, 1970 for further information). This

118 unit is capped in turn by the Gessoso-Solﬁfera Fm.

119 During the Early Pliocene the“Val Marecchia nappe” translated to

120 its present location in the Adriatic foredeep. Thank to this

mecha-121 nism, an almost unique shallow water record of the marginal setting

122 of Northern Apennines has been preserved to date.

123 3. The Legnagnone section

124 The Legnagnone section (recorded asCa'Seriola section inCarloni

125 et al., 1974) is located at 43°55′10″N–12°21′08″E. The section is a

126 55 m-thick monotonous unit, extending from the uppermost sandy

127 layer of the Acquaviva Formation, up to theﬁrst gypsum bed of the

128 Gessoso-solﬁfera Group (Roveri and Manzi, 2006), mainly consisting

129 of marly and clay deposits with minor intercalation of sandstone

130 and indurated limestone bed related to differentiated cementation

131 (Fig. 3). The transition with the overlying Gessoso-solﬁfera Gr. is

132 recorded by two couplets of laminated limestone and organic-rich

133 shale (overall thickness 1.2 m). These limestone layers consist of a

134 partly clotted micrite matrix crossed by contractional cracks and

135

including intraclasts and silt-size quartz and muscoviteﬂakes. Both

136

layer contain minor amounts of dolomite and appear very similar to

137

the carbonate beds associated to the PLG deposits in the Piedmont

138

Basin (Dela Pierre et al., 2011).

139

Based on facies and stacking pattern characteristics of

Gessoso-140

sol_{ﬁfera Gr. in the Val Marecchia area,}Lugli et al. (2010) showed

141

that these limestone/shale couplets are actually a lateral equivalent

142

of the two lowermost gypsum cycles of the Primary Lower Gypsum

143

unit (Roveri et al., 2008); accordingly, theﬁrst gypsum bed of the

144

Legnagnone section correlates with 3rd cycle of the PLG unit and

145

the transition between pre- and syn-evaporitic stages of the MSC

146

lies below the two carbonate-shale couplets.

147

4. Material and methods

148

4.1. Stable isotope geochemistry

149

Bulk samples were collected from 9 levels in the uppermost

150

1.50 m of the section (LW samples,Fig. 3B), just below the lowermost

151

gypsum bed. The analyses were performed at the Laboratory of

Iso-152

tope Geochemistry of the Earth Sciences Department of Parma. The

153

isotopic composition of bulk carbonates was measured on CO2

devel-154

oped after reaction of the powdered solid with 100% H3PO4in vacuo

155

at 25 °C. A selective acid extraction method has been used to measure

156

the stable isotopic composition of samples containing both, calcite

157

and dolomite. The samples were (~ 40 mg) reacted in three steps:

158

1) with >100% H3PO4 at 25 °C for 2 h in vacuum to extract CO2

159

from the calcite fraction, 2) continuously with > 100% H3PO4 at

160

25 °C for 4 h in vacuum to extract CO2from the calcite–dolomite

mix-161

ture (CO2obtained in second step was pumping out from the system)

162

3) the remaining material was reacted at 25 °C for more than 72 h to

163

obtain CO2from the dolomite.

164

The isotopic composition of CO2was measured on a Finnigan Delta S

165

mass spectrometer vs. an internal laboratory CO2standard gas obtained

166

by the reaction at 25 °C of extra pure Carrara marble powder with 100%

167

phosphoric acid. The standard deviation of these measurements was

168

systematically equal to or lower than ±0.15‰(1σ). The CO2standard

169

is periodically calibrated against NBS-19 revealing an isotopic

com-170 position of−2.43‰(δ18_{O vs. VPDB) and +2.45}_‰₍_δ13_C _{vs. VPDB)} 171 respectively. 172 4.2. Palynology 173

For palynological studies (pollen, dinocysts and palynofacies) 55

174

subsamples (about 12 g) were processed at the“ENI-E&P Division”

175

laboratory of Milan, using a standard methodology that involves the

re-176

moval of carbonate with hydrochloric acid (HCl) and the silicate fraction

177

with hydroﬂuoric acid (HF). Pollen concentration, ranging from 259 to

178

42,278grains/g, was calculated using marker grains (Matthews, 1969).

179

Pollen counts, ranging from 100 to 1073 grains, were expressed as

per-180

centages in a summary palynological diagram; the calculation sum

in-181

cluded pollen of all the vascular plants. The main components of

182

sedimentary organic matter were organized inﬁve main groups and

183

expressed as percentages: 1) Amorphous Organic Matter (AOM) and

184

among the Structured Organic Matter (SOM): 2) Black debris (black

185

elongate woody fragments essentially of fusinite); 3) Brown woody

186

fragments; 4) Cuticles (leaf-epidermal tissue; cutinite tissues, etc.); 5)

187

Terrestrial (pollen and spores) and marine (essentially dinoﬂagellate

188

cysts and other marine phytoplankton) palynomorphs. Detailed data

189

on pollen, dinocysts and palynofacies are archived and available on

190

request.

191

4.3. Foraminifers

192

A total of 93 samples were washed on a 60μm mesh sieve and

193

(7)

UNCORRECTED PR

OOF

194 planktonic and benthic specimens was extremely variable from one

195 sample to the other, as well as the weight of the > 125μm residue.

196 The preservation was generally good in the lower part, up to about

197 18 m and moderate in the upper part, where in most of the samples

198 the tests are smoothed and re-crystallized. The total number of

ben-199 thic or planktonic foraminifers in the samples was often too low

200 (minor than 300 specimens) to allow quantitative counts throughout

201 the whole section. In order to obviate this problem and to check the

202 abundance pattern of each species or group of species along the

sec-203 tion, we performed a semi-quantitative analysis (Turco et al., 2011).

204 Semi-quantitative analysis was performed on 80 samples by picking

205 planktonic and benthic specimens in 9 out of 45 squares of a standard

206 picking tray; the counting was stopped at about 30 specimens per

207

taxa and then normalized to one square. In case of abundant taxa, all

208

specimens were counted in the ﬁrst square, even if the number

209

exceeded 30 individuals. Such quantities were plotted against the

strat-210

igraphic quote in order to compare relative abundanceﬂuctuations

211

throughout the stratigraphic column. All specimens were identi_ﬁed

212

and listed inTable 1, while the counting is reported as a supplementary

213

content. Depending on the weight of the residue (>125μm) and on the

214

abundance of foraminifers, the samples were split in order to obtain a

215

sufﬁcient quantity of material to cover the whole picking tray (the

216

weight of the picked residues varies from 0.01 g and 0.58 g, with a

217

mean value of 0.2 g). Due toneogloboquadrinid'spaucity, additional

218

counting up to at least 20 specimens was performed in order to achieve

219

a reliable sinistral (sx) to dextral (dx) coiling ratio. Ratio of planktonic to

Rome Study area Bologna Florence Ancona Bologna trace of section Rimini

ADRIA

TIC SEA

ADRIA TIC SEA Monte Tondo Monticino

Vena del Gesso basin

Forlì Line

12° 00' 13° 00' 44° 00' Fanantello Forlì Sillaro Line Pesaro N STRATIGRAPHY Pliocene and Quaternary Messinian

Early Miocene - Tortonian deposits Macigno-Cervarola unit (late Oligocene - early Miocene) Ligurian allocthonous units Legnagnone

a

b

p-ev

₁

p-ev

₁

p-ev

₂

T

₂

T

₂

PLG

(in situ)

PLG

(in situ)

PLG

(slided blocks)

RLG

(chaotics)

RLG

(turbidites)

turbiditic lobes turbiditic lobes slides slides euxinic shales euxinic shales thin-bedded turbidites

_MES

Emilian Ligurids

Val Marecc

hia Ligurids

thin-bedded turbidites thin-bedded turbidites Lago-Mare fluvio-deltaic hypohaline deposits fluvio-deltaic Pliocene deposits fluvio-deltaic Pliocene deposits open marine Pliocene deposits

Monticino Fanantello Legnagnone

Monte Tondo

Forlì line

Sillaro line

M-P boundary

NW

100 m 20 km

SE

Vena del Gesso basin adriatic foredeep Eastern Romagna basins Val Marecchia wedge-top basin Emilian wedge-top basins

Val Marecchia allocthonous Emilian allocthonous

Fig. 1. (a)—Schematic geological map of the study area and cross section (blue line in (a), modiﬁed afterRoveri et al., 2005), indicating the main structural and stratigraphic features of the Tortonian–Messinian succession (b).(For interpretation of the references to color in thisﬁgure legend, the reader is referred to the web version of this article.)

(8)

UNCORRECTED PR

OOF

220 benthic foraminifer was calculated as P/(P + B) * 100 and is referred

221 to as P/B ratio in the following chapters. Diversity of benthic and

plank-222 tonic foraminifers is here considered as a qualitative observation.

223 Bolivina dentellata is rare along the section and was therefore counted

224 together with B. dilatata in the B. dilatata/dentellata group.

225 4.4. Ostracods

226 Of the 96 processed samples only 28 yielded ostracod valves.

When-227 ever possible, up to 300 ostracod valves per sample were extracted from

228 the 125μm sieved sample, identiﬁed, and counted (Table 2and

supple-229 mentary content). The abundance of each species was normalized to

230 10 g of dry residue. The obtained abundance matrix was processed

sta-231 tistically [diversity indexes, R- and Q-mode hierarchical Cluster Analysis

232 applying the Morisita–Horn distance measure and the un-weighted

233 pair group method using arithmetic average (UPGMA), and Q-mode

234

Detrended Correspondence Analysis (DCA)] using the software package

235

PAST—PAleontological STatistics (ver. 2.04;Hammer et al., 2001).

236

4.5. Macrofossils

237

Macrofossils were searched for all along the succession and were

238

handpicked whenever visible, although in most cases they had to be

239

excavated from the embedding sediment. With very few exceptions

240

(for example oysters), the macrofossils appear severely decalciﬁed

241

and deformed requiring extreme care for their collection in the

242

ﬁeld, bagging and transportation. Further cleaning was performed in

243

the lab by using dental tools and many specimens were hardened

244

with paraloid liquid. Selected specimens were photographed by

245

using a digital Coolpix camera or a binocular microscope MEIJI Techno

246

RZ equipped with digital camera. The obtained collection was then

la-247

beled and is stored in the Geological Museum Capellini (Via Zamboni,

248

63, 40126 Bologna, Italy).

249

4.6. Magnetostratigraphy

250

The magnetostratigraphy of the Legnagnone section is based on

251

two different sample suites (LM and LZM,Fig. 3A). A total of 38

252

magnetostratigraphic samples (marked with LM) coming from the 0

253

to 35 m interval were measured at Fort Hoofddijk Laboratories of

254

Utrecht. Thisﬁrst study permitted to recognize a magnetic inversion

255

at around 18 m below the base of the ﬁrst gypsum bed (Manzi,

256

2001). The sampling is here extended downward until the base of

257

the“Casa i Gessi” argillaceous unit and this new sample set (marked

258

with LZM) was measured at the ALP laboratory (Cuneo). Both sets

259

were thermally demagnetized by means of 20–30 °C steps for

inter-260

nal analytic consistency. Samples measured at Fort Hooddjik Lab.

261

were heated up to 420 °C, but yet at 370 °C NRM intensity usually

in-262

creased and directions get randomly oriented. On this basis the 13

263

LZM samples were heated up to 370 °C. The demagnetization data

264

were digitally acquired by means of Paleodir software (LM) and

265

Remasoft (LZM samples,Chadima and Hrouda, 2006).

266

5. Results

267

5.1. Stable isotope

268

The lowermost three bulk samples, corresponding to the laminated

269

dark marls, show oxygen values ranging from−0.62 to 0.17‰ VPDB

270

and carbon from−3.89 and −3.18‰VPDB (Fig. 4). From the top of

271

this marly interval oxygen values increase (3.37‰ VPDB in LW4)

272

whereas carbon values decrease (−5.14‰VPDB). The calcite and

dolo-273

mite fractions of sample LW5 from theﬁrst laminated carbonate show

274

different isotopic signature. Calcite oxygen is much lighter (−0.15‰

275

VPDB) than in dolomite (6.27‰VPDB), while a smaller difference is

276

noted in the carbon signal as calcite is−6.96‰and dolomite−3.94‰

277

VPDB. The laminated marly interval between the two carbonate is

278

made up of thin alternation of white and dark gray laminae (sample

279

LW6A and LW6B, respectively) yielding slightly heavier oxygen values

280

in the white portion. Again dolomite displays enrichedδ18_{O and}13_C

281

values with respect to calcite. Calcite carbon isotope displays the

mini-282

mum value in the uppermost carbonate layer (−13.15‰VPDB).

Oxy-283

gen in the total bulk ranges from 6.07 to 6.97‰VPDB and displays a

284

remarkable difference between calcite (3.17‰VPDB) and dolomite

285

(7.56‰VPDB) in the uppermost laminated carbonate.

286

5.2. Palynoloy

287

5.2.1. Pollen

288

A pollen record consisting of 70 taxa has been reconstructed from

289

44 samples. Reworked taxa are present throughout the section,

290

among them especially Classopollis and Araucariacites (from early

Argille azzurre Fm. Argille azzurre Fm. Monte Perticara sandstones M. Sabatino Fm. Gessoso-solfifera Fm. (Primary Lower Gypsum)

Legnagnone section Casa i Gessi Fm. Acquaviva Fm. Montebello Fm. Monte Fumaiolo Fm. San Marino Fm. Val Marecchia Ligurian sheet Campaolo marls Monte Senario sandstones

Pliocene

Messinian

Tortonian

Serrav. Langhian Olig. Burd. Aquit.

upper Cretaceous

Eocene

T

₂

LT

₁

MP

MES

LP

depositional sequences after Ricci Lucchi, 1986

5.33 age [Ma] 7.25 11.0 14.7 16.2 21.0 23.5 36.0 83.0

Lithostratigraphic units

Fig. 2. Schematic succession of the Val Marecchia area. MES stands for Messinian erosional surface (Loﬁ et al., 2005).

(9)

UNCORRECTED PR

OOF

291 Cretaceous), bisaccate (probably Mesozoic) and spore (Cretaceous).

292 The non-reworked palynoﬂora is dominated by arboreal taxa as

293 expressed by the percentage values of subtropical humid (especially

294 Taxodium–Glyptostrobus type followed by Engelhardia, Myrica, Nyssa,

295 Sciadopitys, Sequoia, Distylium, etc.) and temperate broad-leaved

decid-296 uous (especially Quercus followed by Carpinus, Juglans, Carya, Ulmus,

297 Zelkova, Tilia, Pterocarya, Liquidambar, etc.) forest taxa as well as of

298 Pinaceae saccate. The latter include principally Pinus but also mid to

299 high elevation forest taxa such as Cedrus, Tsuga, Abies and Picea. Among

300 the much less abundant non-arboreal taxa, Poaceae, Chenopodiaceae,

301 Ericaceae are dominant. Other non-arboreal taxa, such as Asteraceae

302

(including Artemisia), Brassicaceae, Cyperacee, Plantago, Rosaceae,

303

Apiaceae, Sparganium, Tricolporopollenites sibiricum, Borraginaceae,

304

Primulaceae, Cannabaceae, are present in low percentages.

305

The interval is characterized by repeated changes in the pollen

306

assemblages, summarized by eight main pollen assemblage zones

307

(Pol-1 to Pol-8;Fig. 5).

308

POL1 (section base–35 m)—Pollen grains are not well preserved

309

and reworked taxa, including a large component of Classopollis are

310

dominant; concentration is generally very low, always below

311

600g/g. Such features did not permit to obtain relevant pollen data.

D A C B L 15 L 16 L 17 L 18 L 19 L 20 L 21 L 22 L 23 L 24 L 25 L 26 L 04 L 03 L 01 L 05 L 06 L 07 L 08 L 09 L 10 L 11 L 12 L 13 L 14 L 02 E F H G L 27 L 28 L L 34 L 35 L 36 L 37L 38 L 39 L 40 L 29 L 30 32 I J K L N Nbis M P Q * L 42 L 43 L 44 L 45 L 46 L 47 L 48 L 49 L 50 L 51 L 52 L 53 L 54 L 55 L 56 L 57 L 58 L 59 L 60 L 61 L 62 L 63 L 64 L 65 L 66 L 67 L 68 L41 Lz 16 Lz 15 Lz 14 Lz 17 Lz 13 Lz 1 Lz 2 Lz 3 Lz 4 Lz 5 Lz 6 Lz 7 Lz 8 Lz 9 Lz 10 Lz 11 Lz 12 LW1 LW2 LW3 LW4 LW5 LW6 LW8LW9 LW7 1,5 1 0,5 0 m San Marino Fm. Aquaviva Fm. Casa i Gessi Fm. Gessoso-Solfifera Fm. not to scale

0 m

5

20

25

30

35

40

45

50

10

15 selenitic gypsum

claystone

sandstone

conglomerate

biocalcarenite

marls

lamination

limestone

marly limestone

a

c

b

LZM LM 18 23 01 36 P samples M

Fig. 3. a—Lithological section and position of the palaeontological (P) (letters for macrofossil samples) and magnetostratigraphic (M) samples. b—Magniﬁcation of the uppermost 1.5 m of the section. c—Panoramic view of the Legnagnone outcrop with the main litostratigraphic units described in the text.

(10)

UNCORRECTED PR

OOF

312 POL2 (35–27 m)—A quite good expansion of humid, subtropical

313 to warm-temperate broad-leaved deciduous forest taxa with a

314 completely subordinate occurrence of herbs (principally

315 Chenopodiaceae at 1.4%) characterizes this interval. At the same

316 time pollen concentration increases signiﬁcantly, up to 4100g/g

317 (prevalently AP), whereas the reworked palynomorphs decrease.

318 Subtropical, humid forest taxa, especially Taxodium/Glyptostrobus

319 type, and warm temperate taxa, especially Quercus show repeated

320 ﬂuctuations, that are often in phase between them. However, the

321 subtropical, humid forest taxa show an overall decreasing trend

322

(29.5% at 33.5 m to 13.6% at 27 m) towards the top whereas an

in-323

creasing trend is observed among the temperate broad-leaved

de-324

ciduous forest taxa, spanning between 3.5% and 11.2%. Pinus plus

325

other indeterminable Pinaceae saccate show as well repeated

per-326

centageﬂuctuations throughout the succession, usually opposite

327

to all the previous groups. Mid to high altitude coniferous show

328

a percentage increase between 33.5 m and 31.5 m with maximum

329

values for Cedrus plus Tsuga (24%) as well as Abies (5%) at 32.5 m.

330

Herbaceous taxa show a major peak, up to 15.5% just above, at

331

28.25 m, which includes also the major occurrence in the section

332

of Artemisia (1.8%).

333

POL3 (27–22 m)—Subtropical taxa show a clear increasing trend

334

but also repeatedﬂuctuations, which are in phase with those of

335

the warm temperate taxa. Pinus plus other indeterminable Pinaceae

336

saccate show as well repeated percentageﬂuctuations, usually

op-337

posite to all the previous ones. Mid to high altitude coniferous are

338

constantly present but always respectively lower than 7.6% and

339

2.6%. Herbaceous taxa show some increases matching with

subtrop-340

ical and temperate broad-leaved deciduous taxa expansions at

341

25.6 m (L55).

342

POL4 (22–15.8 m)—The increasing trend of subtropical taxa is still

343

evident; however the latter as well as the warm temperate taxa

suf-344

fer, after successive in phaseﬂuctuations, a sharp fall at 16 m (67%)

345

linked to a major rise of Pinus plus other indeterminable Pinaceae

346

saccate. Mid altitude coniferous taxa (especially Cedrus) are

con-347

stantly present in good percentages, never below 8% and up to

348

21.5% at 16.4 m. Herbaceous taxa show some increases matching

349

with temperate broad-leaved deciduous taxa expansion at 21.5 m

350

(L47) and 18.3 m (L39). A strong increase in concentration is

ob-351

served between 20.2 and 18.3 m.

352

POL5 (15.8–12.5 m)—This interval shows an increase of humid,

353

subtropical (27.5%) and especially warm temperate (34.2%) pollen.

354

Again, they both show repeated in phaseﬂuctuations, usually in

355

opposition to those of Pinus plus other indeterminable Pinaceae

356

saccate. Mid to high altitude coniferous are constantly present;

357

Abies followed by Picea show a good increase at 13.55 m (L27)

358

reaching the 4.3%. Herbaceous taxa show a large increase at

359

15.3 m (L30) reaching 10.47%, which is half related to an Ericaceae

360

expansion in phase with subtropical to warm- temperate

broad-361

leaved deciduous taxa.

362

POL6 (12.5–4.7 m)—Warm temperate taxa largely increase at the

363

base of the interval, whereas the subtropical ones reach the higher

364

values at top. Previous groups show oppositeﬂuctuations regarding

365

those of Pinus plus other indeterminable Pinaceae saccate. Mid to

366

high altitude coniferous are constantly present. Herbaceous taxa

367

show three successive peaks (up to 15.2%) matching with

subtropi-368

cal (9 m and 5.2 m)/or temperate broad-leaved deciduous (11.3 m)

369

taxa expansions.

Table 1 t1:1

t1:2 List of the collected foraminiferal species.

t1:3 Benthic foraminifera Melonis padanum

t1:4 Agglutinants Melonis soldanii

t1:5 Ammonia tepida Miliolids

t1:6 Anomalinoides helicinus Nonion sp. t1:7 Bolivina spatulata Oridorsalis stellatus t1:8 Bolivina dilatata/dentellata Oolina squamosa t1:9 Bolivina cf. hebes Ortomorphina sp. t1:10 Bolivina sp. Protoelphidium granosum t1:11 Bulimina aculeata Rectuvigerina gaudryinoides t1:12 Bulimina costata Rosalina globularis t1:13 Bulimina echinata Textularia sp. t1:14 Bulimina elegans Trifarina sp. t1:15 Bulimina enlongata Uvigerina peregrina t1:16 Cancris oblungus Valvulineria t1:17 Cassidulina neocarinata

t1:18 Cibicides lobatulus Planktonic foraminifera: t1:19 Cibicides sp. Globigerinita glutinata t1:20 Cribroelphidium Globigerinella siphonifera t1:21 decipiens/translucens/poeyanum Globigerina bulloides t1:22 Ellipsoidina ellipsoides Globigerinoides spp. t1:23 Elphidium advenum/macellum Globorotalia scitula t1:24 Florilus boueanum Globoturborotalita sp. t1:25 Globobulimina subglobosa Neogloboquadrina acostaensis t1:26 Gyroidinoides sp. Orbulina universa

t1:27 Hanzawaia boueana Turborotalita multiloba t1:28 Hopkinsina bononiensis Turborotalita quinqueloba t1:29 Lenticulina spp.

Table 2 t2:1

t2:2 List of the collected ostracod species. t2:3 Acanthocythereis hystrix (

Q2 Reuss, 1850)

t2:4 Aurila (Alboaurila) albicans (Ruggieri, 1958) t2:5 Aurila (Aurila) convexa (

Q3 Baird, 1850)

t2:6 Bosquetina sp.

t2:7 Callistocythere antoniettae t2:8 Callistocythere pallida praecedens t2:9 Carinocythereis galileaRuggieri, 1972 t2:10 Celtia clatrataMiculan, 1992 t2:11 Cyamocytheridea sp. t2:12 Costa edwardsii ( Q4 Roemer, 1838) t2:13 Cytherella pulchella t2:14 Cytheridea neapolitana Q5 Kollmann, 1960 t2:15 Hemicytherura deﬁorei Q6 Ruggieri, 1953 t2:16 Keijella lucida t2:17 Keijella punctigibba ( Q7 Capeder, 1902) t2:18 Leptocythere elliptica t2:19 Leptocythere sanmarinensis t2:20 Olimfalunia stellata t2:21 Palmoconcha agilis

t2:22 Palmoconcha dertobrevis (Ruggieri, 1967) t2:23 Phlyctenophora sp. t2:24 Rectobuntonia subulata t2:25 Ruggieria tetraptera t2:26 Sagmatocythere variesculpta t2:27 Sagmatocythere versicolor ( Q8 Müller, 1894)

t2:28 Semicytherura sanmarinensisRuggieri, 1967 t2:29 Tenedocythere sp.

t2:30 Xestoleberis cf. X. dispar (

Q9 Müller, 1894)

t2:31 Xestoleberis reymentiRuggieri, 1967 t2:32 Xestoleberis sp.

Table 3 t3:1

t3:2 Lists of foraminiferal and magnetostratigraphic events referred toFig. 11. Code for

ref-t3:3 erences: S01 =Sierro et al., 2001; L04 =Lourens et al., 2004; R09 =Roveri et al.,

t3:4 2009; ps = present study.

t3:5

events age (Ma)

t3:6 1. N. acostaensis dominance sinistral forms 6.108–6.140S01

t3:7 2. T.multiloba inﬂux 6.122S01 t3:8 3. G. scitula gr. 2nd inﬂux 6.098–6.107S01 t3:9 4. a-N. acostaensis sx>40%

b-Last inﬂux T. multiloba

6.078–6.082S01 6.08S01 t3:10 5. Base C3r Chron 6.033L04 t3:11 6. HO planktonic foraminifer ≈6.0ps t3:12 7. HO benthic foraminifer ≈5.974R09

(11)

UNCORRECTED PR

OOF

370 POL7 (4.7–2 m)—A progressive increase of Pinus plus other

inde-371 terminable Pinaceae saccate occurs, the most important at 2.7 m

372 (82.6%) just after a notable concentration increase at 4.25 m; all

373 the other vegetal groups are decisively subordinate.

374 POL8 (2 m–base 1st gypsum bed) —Humid, subtropical forest

375 taxa especially Taxodium/Glyptostrobus type, and warm temperate

376

taxa, especially Quercus show successive in phase ﬂuctuations

377

(1.5 m, 0.65 m, 0.15 m). Mid to high altitude coniferous are

378

constantly present with some percentage increases, the most

rel-379

evant at 1.5 m, where Abies followed by Picea reach the 3.4% and

380

at 0.15 m, where Picea, Abies and Fagus reach together the 4.2%

381

and Cedrus plus Tsuga the 14.3%. Herbaceous taxa show some

-2 -1 ₀ ₁ ₂ ₃ ₄ ₅ ₆ ₇ ₈ -14 -12 -10 -8 -6 -4 -2 ₀ calcite dolomite calcite dolomite LW1 LW2 LW3 LW4 LW5 LW6 LW8 LW9 LW7 1,5 1 0,5 0 m

Fig. 4. Isotope signature (O and C) of the pre-evaporitic/evaporitic transition at Legnagnone. Black arrows highlight the different values of calcite and dolomite in the laminated limestone respectively equivalent to theﬁrst and second PLG cycles (samples LW5 and 8). Full and empty symbols in LW6 refer respectively to dark and withish laminae (see de-scription in the text). The MSC onset is placed at the base of the lowermost laminated limestone.

Pollen (%) Palynomorphs concentration Palynofacies (%) 0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100 AOM FUSINITE WOOD CUTICLES PALYNOMORPH PA1 PA2 PA3 POL1 POL2 POL3 POL4 POL5 POL6 POL7 POL8 “Subtropical” humid forest taxa

Tsuga, Cedrus Abies, Picea, Betula, Fagus Temperate broadleaved deciduous forest taxa other arboreal plants non arboreal plant Pinus + other Pinaceae indet. Pinus haploxylon+ cf. cathaya Schlerophyll forest taxa gr./g 0 5 20 25 30 35 40 45 50 10 15 15k 30k 2.4k 550 NAP AP Dinocysts

Fig. 5. Plots of cumulative percentage of palynofacies and pollen groups and their relative palaeoenvironmental intervals. Shaded rectangles at the base of the pollen diagram in-dicate intervals with no data.

(12)

UNCORRECTED PR

OOF

382 ﬂuctuations within 1.1% and 6.1% again matching with temperate

383 broad-leaved deciduous taxa expansions (1.5 m and 0.65 m).

384 5.2.2. Palynofacies

385 Three main palynozones were recognized based upon the analysis

386 of the dispersed organic matter (Fig. 5):

387 PA1 (section base–35 m) — Amorph Organic Matter (AOM),

388 fusinite, brown phytoclast and cuticles are equally present, with

389 a slight predominance of the two last components; palynomorphs

390 are subordinated.

391 PA2 (35–22 m)—An expansion of fusinite occurs, which ranges

392 from 40% to 60%; AOM initially drops to lessthan5% and gradually

393 increase up to 20%. Both brown phytoclast and cuticles gradually

394 decrease down to 10%.

395 PA3 (22 m–base 1st gypsum bed)—Palynofacies percentage shows

396 repeatedﬂuctuations. Antiphase and ample ﬂuctuations affected

397 AOM (4% to 63%) and fusinite (23% to 79%), the latter reaching its

398 maximum value in the second laminated carbonate. An antiphase

399 relationship was found between brown phytoclasts/cuticles and

400 fusinite once removed the AOM and palynomorph signal. Despite

401 repeatedﬂuctuations, brown phytoclasts and cuticles continue the

402 descendent overall trend already enlightened the previous interval.

403 Palynomorphs show two low frequencyﬂuctuations with

percent-404 age peaks of about 10 and 20% respectively at 17 and 2 m.

405 5.2.3.Phytoplankton

406 The in situphytoplanktonis decisively very rare and only expressed

407 by the occurrence of long-ranging taxa such as Spiniferites spp. and

408 Operculodinium spp. Reworked taxa are present throughout the

sec-409 tion and include principally dinoﬂagellate cysts from Cretaceous

410 (Aptian/Albian) as well as Paleogene and early Miocene.

411 5.3. Foraminifers

412 5.3.1. Planktonic foraminifers

413 Four distinct intervals can be recognized based upon planktonic

414 foraminiferal distribution, abundance and diversity (Fig. 6):

415 PF1 (section base_–35 m)_—Planktonic foraminifers are very rare

416 and discontinuously present. Globigerina spp. and Neogloboquadrina

417 (scattered) are the only represented taxa.

418 PF2 (35–19 m)—The presence of planktonic foraminifers becomes

419 more continuous; a few barren levels are alternated with

occur-420 rences ranging from 5 to 32 specimens/_{ﬁeld. The number of species}

421 is still low, but rises up to a maximum of 7 species between 22.5 and

422 23.7 m. The two major planktonic inﬂuxes at 33.5–31.5 m (P/B =

423 56%—total abundance = 62.5 specimens/ﬁeld) and 19.7–20.2 m

424 (63% — 37–40 specimens/ﬁeld) are mainly characterized by

425 Turborotalita quinqueloba and T. multiloba, with a minor

abun-426 dance of both dx- and sx-coiled Neogloboquadrina acostaensis.

427 The interval between these two inﬂuxes is characterized by an

in-428 crease in diversity; the assemblage is prevalently made up of

429 Globigerina bulloides, Orbulina universa and N. acostaensis. Minor

430 percentage of Globigerinita glutinata and Globigerinella siphonifera

431 are present. Globorotalia scitula is present only between 23.1 and

432 24.3 m, together with very rare Globigerinoides.

433 PF3 (19–9 m)—Planktonic foraminifers are present only in three

434 prominent inﬂuxes at 14.9–15.7 m (99%), 11.3–12 m (97%) and

435 9 m (99%). The lower one is characterized by the maximum absolute

436 abundance of planktonic foraminifers (>200 specimens/ﬁeld), which

437 decrease down to 40 specimens/ﬁeld in the other two peaks. In this

438 interval the assemblage is rather constant (O. universa, G. bulloides,

439

N. acostaensis and rare T. quinqueloba, G. siphonifera, G. glutinata and

440

Globoturborotalita sp.). O. universa is the most common taxa in the

441

14.9-15.7 m and 9 m inﬂuxes, while at 11.3-12 m N. acostaensis

442

and G. bulloides are prevalent.

443

PF4 (9 m–base 1st gypsum bed)—Planktonic foraminifers are

444

absent with the only exception of very rare small globigerinids

445

at 5.7 m.

446

5.3.2. Benthic foraminifers

447

Seven stratigraphic intervals are identiﬁable on the basis of the

448

vertical distribution of benthic foraminifers (Fig. 6):

449

BF1 (section base–35 m)—Benthic foraminifers are scarce to

abun-450

dant (42–39 m) and diversity follows the abundance pattern.

451

Ammonia tepida is dominant, while Cribroelphidium decipiens and

452

Protoelphidium granosum are common throughout the interval.

453

Other taxa, such as Hanzawaia boueana, Valvulineria bradyana,

454

Florilus boueanum, Elphidium sp., Cibicides sp., Lenticulina spp. and

455

Bulimina echinata are discontinuously present with minor

percent-456

age. B. echinata is present only in the topmost sample of this interval.

457

BF2 (35_{–30.5 m)} _—Diversity and abundance; the assemblage

458

is characterized by a drastic change, as Bulimina (B. echinata,

459

B. aculeata and B.elongata) and Bolivina genera (B. spathulata

460

and B. dilatata/dentellata) start to be present and dominant.

461

H. boueana ranks the most abundant among the secondary taxa

462

(miliolids, rare C. decipiens and P. granosum).

463

BF3 (30.5–22 m)—The number of species remain relatively high

464

reaching the maximum at 23.1 m. The overall abundance increases;

465

the assemblage is dominated by B. spathulata up to 27 m and by

466

B. dilatata/dentellata up to the top of the interval. Buliminids are

con-467

tinuously present in minor percentage. Accessories species are rare

468

to common and range between 5 and 30 specimens/ﬁeld. Among

469

them, V. bradyana is the most abundant. Also Uvigerina peregrina,

470

Rectuvugerina gaudryinoides and Hopkinsina bononiensis are present

471

within the whole interval, while other taxa, such as Cibicides sp.,

472

Bulimina costata, Lenticulina spp. are scarce and present only within

473

short-lived inﬂuxes. At around 23 m several other species are

474

present: Oridorsalis stellatus, Gyroidinoides sp. Cancris oblungus,

475

Ortomorpina sp. and Globobulimina subglobosa

476

BF4 (22–13 m)—Diversity drops and barren levels intercalate

477

with peak of abundance, which decrease in amplitude from 18 m

478

up to the top of the interval. B. dentellata/dilatata is dominant

be-479

tween 22 and 18 m, where it occurs in short-lived inﬂuxes; B. cf.

480

hebes has been observed at 15 m. B. echinata and B. aculeata are

481

more regularly present, but with minor abundance as in previous

482

interval. H. boueana is common during short inﬂuxes at about 20,

483

18 and 15 m. As well as in BF2, the occurrence of H. boueana is

as-484

sociated to a reduction in benthic abundance, diversity and to an

485

abundance decrease of the bolivinids.

486

BF5 (13–7 m)— Diversity remains low; meanwhile abundance

487

further decreases leading to occurrence of barren intervals. Rare

488

buliminids and bolivinids and scattered secondary taxa are present.

489

BF6 (7–1 m)—Diversity is still low, but abundance progressively

490

increases following the abundant occurrence of B. echinata and

491

B. aculeata and of B. spathulata. The former taxa reach here their

max-492

imum absolute abundance values. Oolina hexagona, Haplophragmoides

493

sp., Protoelphidium granosum and Cribroelphidium decipiens become

494

important among the accessory taxa, but they are far less abundant

495

than in BF1 and only Haplophragmoides sp. and Cribroelphidium

496

decipiens exceed 5 specimens/ﬁeld.

497

(13)

UNCORRECTED PR

OOF

C3r C3An.1n P/(P+B)% 0 25 50 75 100 Bolivina 0 100 200 300 B.dilatata/ dentellata B.spathulata B.hebes 0 5k 10k 15k 20k 25k Ostracod abundance (n/10g of dry residue) BF3 BF1 BF2 BF4 BF5 BF6 BF7 O2 O1 O3 O4 G. b ulloides 20 0 40 60 80 N. acostaensis 20 0 O . univ ersa 20 0 40 60 80 20 0 Tutborotalita40 PF1 PF2 PF3 PF4

Benthic foraminifers (n/field)

0 10U20 . pereg rina 0 100 Bulimina200 300 0 10 20 Haplophr moides spp . 0 10A. 20 tepida 0 10 20 C. decipiens 0 10 20F. 30 40 bouean um 0 10 20 P. g ranosum 0 10 V.20 30 brady ana 0 10 H.20 30 boueana 0 5 20 25 30 35 40 45 50 10 15

Planktonic foraminifer (n/field)

B.echinata B.aculeata T.quinqueloba T.multiloba B.elongata 0 100 200 300 400 159 Foraminif er abundance (n/field) planktonic benthic

Fig. 6. Plot of P/B ratio, total abundance of (planktonic and benthic) foraminifers, distribution of selected benthic and planktonic foraminiferal taxa and plot of absolute abundance of ostracods against lithological log of Legnagnone section. The black triangle indicates the position of the G. scitula inﬂux. Vertical boxes indicate the palaeoenvironmental foraminiferal and ostracod intervals discussed in the text.

9 R. Gennar i et al. / Palaeogeography, Palaeoclimatology, Palaeoecology xxx (2013) xxx – xxx e a se ci te th is ar ti cl e a s: G e n n ar i, R. , e t a l., A sh al lo w w at er re co rd o f th e o ns e t of th e M es si n ian sa lini ty cr is is in th e A dr ia tic for ed ee p. .., la eo g e og ra p h y ,P a la eo cl im at ol o g y ,Pa la eo ec o lo g y (2 0 13 ), ht tp :/ /d x. d o i.or g /1 0 .1 0 1 6 /j. p a la eo .2 01 3. 05 .0 1 5

(14)

UNCORRECTED PR

OOF

498 5.4. Ostracods

499 Ostracods are very well preserved and abundant from the base to

500 35 m; diversity and abundance decrease from this level upwards;

501 they are absent in certain intervals, particularly in the upper portion

502 of the section (samples L38_–L29; L24_–L19; L13_–L11; L5_–L1). On

503 the whole, 31 species have been identiﬁed, referable to 22 genera

504 (Table 1). Only few species are common and dominant within their

505 assemblages (Aurila (Alboaurila) albicans, Cytheridea neapolitana,

506 Ruggieria tetraptera and Palmoconcha dertobrevis) while the others

507 are scattered along the section.

508 Four intervals have been deﬁned on the basis of the ostracod

509 assemblages:

510 O1 (section base–35 m)—This portion is characterized by rich but

511 poorly diversiﬁed (3–10 species) ostracod assemblages, strongly

512 dominated by A. (A.) albicans and C. neapolitana, accompanied by

513 several species exclusive of this interval like Carinocythereis galilea,

514 Celtia clatrata, Leptocythere sanmarinensis, Sagmatocythere versicolor,

515 Palmoconcha agilis and Xestoleberis reymenti.

516 O2 (35–17 m)—Ostracod assemblages are very poorly

diversi-517 ﬁed (1–5 species) dominated by R. tetraptera, accompanied by

518 Acanthocythereis hystrix and, sporadically by Costa edwardsi,

519 Loxoconcha dertobrevis and Keijella punctigibba.

520 O3 (17–2.7 m)—Diversity slightly increases respect to O2 (2–8

521 species) and the assemblages are generally characterized by the

522 co-dominance of R. tetraptera, P. dertobrevis and Keijella lucida

523 (this latter species exclusive of this interval), which are

accompa-524 nied by Sagmatocythere variesculpta, Rectobuntonia subulata and

525 Bosquetina sp.

526 O4 (2.7 m–base 1st gypsum bed)— This interval is barren of

527 ostracods.

528 5.4.1. Statistical analyses

529 Diversity indexes, cluster analysis and Detrended Correspondence

530 Analysis (DCA) applied on the normalized abundance matrix are

531 illustrated inFig. 7. By selecting a cut-off value very close to 0 for

532 the across-cluster similarity in the R-mode dendrogram (Fig. 7a),

533 species are statistically discriminated into two groups: Cluster 1

in-534 cludes, among other, R. tetraptera, R. subulata, A. hystrix, P. dertobrevis

535 and C. edwardsi, which are all species characteristic of marine inner

536 circalittoral environments (50–100 m) (Ruggieri, 1962, 1967, 1972;

537 Bonaduce et al., 1976; Lachenal, 1989; Barra, 1991); Cluster 2

in-538 cludes species more representative of marine infralittoral

environ-539 ments such as C. antoniettae, C. pallida praecedens, L. sanmarinensis,

540 C. neapolitana, C. galilea, A. (A.) albicans, S. versicolor, S. sanmarinensis

541 and X. reymenti (Aruta, 1966, 1983; Ruggieri, 1967, 1996; Bonaduce et

542 al., 1976; Lachenal, 1989; Barra, 1991; Miculan, 1992). In the Q-mode

543 dendrogram (Fig. 7b), very close to the 0.05 level of similarity,

sam-544 ples are grouped into two clusters: Cluster A groups all LZ samples

545 (i.e. the lower portion of the Legnagnone section) and samples L63

546 and L21. This cluster includes samples bearing species grouped in

547 Cluster 2 of the R-mode dendrogram, thus it groups infralittoral

sam-548 ples; Cluster B groups all the remaining L samples. This latter cluster

549 includes species grouped in Cluster 1, referable to inner circalittoral

550 environment. At a cut-off value near 0.15 Cluster B splits into two

551 groups: Cluster B1 that includes samples characterized by the

abun-552 dance of R. tetraptera and K. lucida and Cluster B2 that groups samples

553 mainly dominated by R. tetraptera.

554 Results from the Q-mode DCA are reported inFig. 7c. The axes of

555 the biplot account for 18% (axis 1) and 14.2% (axis 2) of variance.

556 Concerning axis 1, all samples LZ plus L63 and L21 (corresponding

557 to Cluster A of the UPGMA analysis) are located in the left portion

558 of the biplot and are closely clustered in Group I. All the infralittoral

559

species of Cluster 2 are located around this group, such as A. (A.)

560

albicans (alb), C. neapolitana (nea), L. sanmarinensis (san), S. versicolor

561

(ver), C. galilea (gal), P. agilis (agi), C. clatrata (cla) and X. reymenti

562

(rey). Within Group I two very close subgroups can be separated:

563

theﬁrst one, located more to the left, includes samples without the

564

circalittoral R. tetraptera; the second, located slightly more on the

565

right of the biplot, includes samples with few R. tetraptera.

566

Moving towards the right along Axis 1, samples of the upper part

567

of the section are displaced along with circalittoral species dots.

568

Groups are not well deﬁned, probably owing to the oligotipy and

569

very low equitability of the assemblages. Another cloud of samples

570

can be possibly detected, Group II, including mainly the lower

sam-571

ples of the upper portion of the Legnagnone section. The dot of

572

R. tetraptera (tet), the dominant species of the assemblages, plots in

573

this group. The remaining samples from this part are more or less

574

scattered in the right part of the biplot. Finally, it is possible to

recog-575

nize Group III as including samples still dominated by R. tetraptera

576

together with A. hystrix (hyx), C. edwardsi (edw), and sporadically

577

by R. subulata (sub), and Group IV clustering samples showing the

578

co-dominance of R. tetraptera and K. lucida (luc) and the presence of

579

Bosquetina sp. (BOS).

580

Based on the autoecological parameters reported inTable 2, it is

581

possible to infer that Axis 1 represents the ecological parameter

582

depth; the inner infralittoral environment is represented by the left

583

extreme of the axis while the inner circalittoral by the right one.

584

5.5. Macrofossils

585

Macrofossils are relatively abundant along the section, most

no-586

ticeably in the bottom to middle part of it (Fig. 8). They are absent

587

(or at least un-noticed) in the very upper part of the succession. The

588

specimens are by large very decalci_{ﬁed and deformed, making their}

589

identiﬁcation commonly problematic. Overall, about 50 macrofossils

590

taxa have been identiﬁed at Legnagnone, belonging to Mollusca,

591

Echinodermata, Crustacea, Bryozoa and Polychaeta. By far mollusks,

592

especially bivalves, dominate the fossil assemblages. On the base of

593

the vertical distribution of macrofossils assemblages, we recognized

594

7 zones (Fig. 8):

595

M1 (section base–35 m)—The macrofaunal assemblage is

domi-596

nated by infaunal suspension and deposit feeding bivalves

597

(Nucula, Saccella, Modiolus, Acanthocardia, Atrina, Parvicardium,

598

cf. Tellina, Abra, Veneridae spp., Corbula), suspension feeding

599

(Turritella) and predatory/scavenging gastropods (Naticidae spp.,

600

Nassarius), spatangid echinoids, serpulid polychaetes (including

601

Ditrupa) and bryozoans. Their abundance varies conspicuously,

602

from fossil poor layers where macrofossils are very sparse to levels

603

of high concentration of shells (e.g. ca. 39.5 and 40–41 m) locally

604

including oyster (Neopycnodonte) aggregates.

605

M2 (35–22 m)—Macrofauna is rather sparse, dominantly serpulids

606

in the lower part and Corbula aff. C. gibba and Nassarius ex gr

607

semistriatus in the upper part.

608

M3 (22–18 m)—Macrofossils are more common and the

assem-609

blages contain infaunal bivalves such as the lucinid Myrtea sp.

610

The bivalve fauna is relatively diverse and includes other infaunal

611

and semi-infaunal deposit- and suspension-feeding, and carnivore

612

bivalves (Yoldia, Modiolus, Cuspidaria, Cardiomya).

613

M4 (18–14 m)—This interval contains sparse macrofossils (more

614

common at ca. 17 m), such as bivalves (Cardiomya, Corbula),

615

Nassarius ex gr. semistriatus, serpulids (Ditrupa?) and spatangid

616

echinoids.

617

M5 (14_{–3 m)}_—Here we found no obvious macrofossils, a possible

618

indication of nearly prohibitive conditions on the seabottom for

619

(15)

UNCORRECTED PR

OOF

Fig. 7. Results of the multivariate analyses performed on the ostracod assemblages of the Legnagnone succession. a. and b.: dendrograms resulting from the Cluster Analysis (UPGMA– Morisita–Horne distance) in R-mode (a) and Q-mode (b); c. Q-mode biplot (Detrended Correspondence Analysis). Abbreviations: agi: P. agilis; alb: A. albicans; ant: C. antoniettae; BOS: Bosquetina sp.; cla: C. clatrata; con: A. convexa; cri: S. cristatissima; CYA: Cyamocytheridea sp.; def: H. deﬁorei; der: P. dertobrevis; dis: X. dispar; edw: C. edwardsi; ell: Leptocythere elliptica; gal: C. galilea; hyx: A. hystrix; luc: K. lucida; mar: S. sanmarinensis; nea: C. neapolitana; pal: C. pallida praecedens; PHL: Phlyctenophora sp.; pul: C. pulchella; pun: K. punctigibba; rey: X. reymenti; san: L. sanmarinensis; ste: O. stellata; sub: R. subulata; TEN: Tenedocythere sp.; tet: R. tetraptera; var: S. variesculpta; ver: S. versicolor; XES: Xestoleberis sp.

(16)

UNCORRECTED PR

OOF

620 M6 (3_{–1.5 m)}_—Macrofossils are again present, with the occurrence

621 of small articulated bivalves (Parvicardium, Abra, cf. Cardiomya),

622 M7 (1.5 m–base 1st gypsum bed)—No macrofossils have been

623 identiﬁed in this interval.

624 5.6. Magnetostratigraphy

625 A ChRM component was successfully isolated in the 180°/210° to

626 330°/370 °C interval by means of a principal component analysis on

627 Zijderveld diagrams (Zijderveld, 1967). Results from palaeomagnetic

628 analysis showed that both sets of samples (LZM from 0 to 35 m and

629 LZ from 35 down to 52 m) generally display a randomly oriented

vis-630 cous component at room temperature. LZM diagrams displayed a

nor-631 mal polarity/low temperature component up to 160–180 °C, while

632 from 180°to210 °C both reverse and normal polarity are present

633 and interpreted as the primary signal (ChRM;Fig. 9a to d). ChRM

di-634 rections for LZM samples are quite dispersed (Fig. 9e), this is evident

635 from the mean directions for reversal and normal polarity, which

636 were calculated by Fisher's statistics, respectively 189.4°N/−66.6°

637 (N = 11, k = 8.6,α95 = 16.6) and 55.0°N/55.0° (N = 15, k =11.7,

638 α95 = 11.6). LZ samples, collected in the lowermost part of the

sec-639 tion, are all of normal polarity and is often difﬁcult to distinguish the

640 low (80°–260 °C) from the high (280°–390 °C) temperature

compo-641 nents, as they show overlapping directions. Mean directions of both

642 components are normally oriented and, by removing the bedding

cor-643 rection, they are generally distributed very close to the geocentric

644 axial dipole (GAD) direction in geographic coordinates (Fig. 9f) for

645 the Legnagnone area (D = 2.15°N; I = 60.23°). Thus, we interpreted

646 the LZ samples as possibly remagnetized by the present dayﬁeld and

647 not suitable for magnetostratigraphic purposes.

648 The new palaeomagnetic analyses allow to adjust the reversal

649 boundary previously placed at 18 m below the base of theﬁrst

gyp-650 sum bed (Manzi, 2001). In fact, by plotting the VGP (Virtual

Geomag-651 netic Pole) obtained from stable direction of remanent magnetization

652

(Fig. 10), a normal polarity interval is identiﬁed from 32 up to 16 m;

653

furthermore, a 3 m-thick undeﬁned polarity interval occurs between

654

16 and 13 m, from where a reverse polarity zone occurs up to the

655

base of theﬁrst gypsum bed. Accordingly, we now suggest that the

656

magnetic inversion is better placed between 16 and 13 m.

657

6. Discussion

658

6.1. Stable isotope interpretation

659

The general trend shown by the carbonates, both calcite and

dolo-660

mite, in the uppermost meter of the section toward heavierδ18

O

661

values indicates an inﬂuence of strong evaporative condition

increas-662

ing progressively up to theﬁrst gypsum layer (Fig. 4). The lowerδ18_O

663

values of calcite in comparison to the coexisting dolomite suggest

664

that the two minerals are not coeval and may re_{ﬂect different}

forma-665

tion conditions and/or diagenetic overprint.

666

A specular trend to the oxygen curve is drawn by theδ13_C_that

667

yield depleted values toward the top suggesting a concomitant

pro-668

gressive inﬂuence of isotopically light carbon from organic matter

669

degradation, in particular sulfate bacterial reduction (Wacey et al.,

670

2008), which may be a result of increasing stagnation condition at

671

the beginning of sulfate deposition.

672

Analogous trends were described in the Sutera section in Sicily by

673

Oliveri et al. (2010).

674

6.2. Biostratigraphy and age model

675

Biostratigraphy is based on the semi-quantitative counting of

plank-676

tonic foraminifers, as calcareous nannofossils are largely reworked from

677

older sediments. The presence of planktonic foraminifers is

discontinu-678

ous; a common feature shared with the reference Mediterranean

679

sections, where lower Messinian markers are often represented by

in-680

ﬂuxes (Sierro et al., 2001). N. acostaensis is common to rare along the

681

section and its coiling ratio is prevalently sinistral (70%) at 34 m, from

682

above this level dextral specimens are prevalent except at 20 m,

683

where sx and dx specimens match. Two peaks of T. multiloba are present

684

at 19.95 and 32.5 m. Between theﬁrst and second peak of T. multiloba,

685

an inﬂux of Globorotalia scitula is present at 23.5 m. According to the

po-686

sition of planktonic foraminiferal bioevents in the astronomically tuned

687

Molinos/Perales section of the Sorbas basin (Sierro et al., 2001) the

fol-688

lowing correlation and age attribution can be drawn (Fig. 11): 1) the

689

predominance of N. acostaensis sx at 34 m could fall within the sinistral

690

form dominance, ranging from 6.140 to 6.108 Ma (cycles UA27–UA28);

691

2) the lowermost inﬂux of T. multiloba can be correlated with the

pen-692

ultimate inﬂux at Molinos/Perales, which falls in the upper part of

693

cycle UA28, dated at 6.121 Ma; 3) the G. scitula inﬂux correlates with

694

its 2nd inﬂux dated between 6.099 and 6.105 Ma (cycle UA29); 4a)

695

the N. acostaensis sx inﬂux at 40% can be correlated with the 2nd inﬂux

696

of N. acostaensis dated between 6.078 and 6.082 Ma (cycle UA30); 4b)

697

the last inﬂux of T. multiloba is dated at 6.08 Ma (cycle UA30).

698

According to this biostratigraphic framework, the normal

699

magnetozone between 32and16 m correlates with the upper part of

700

sub-chron C3An.1n and the reverse magnetic interval (13–0 m) with

701

the lower part of sub-chron C3r. The C3r/C3An.1n reversal is placed at

702

the midpoint of the undeﬁned polarity interval between 16 and 13 m,

703

at 14.5 m (event 5 inFig. 11). It represents the astronomically calibrated

704

event that best approximates the onset of the MSC. Its age at Ain el

705

Beida ranges between 5.998and6.040 Ma (Krijgsman et al., 2004). A

706

ﬁner astronomical tuning was achieved in the Sorbas basin, where

707

this reversal is dated at 6.033 ± 0.003 Ma (Sierro et al., 2001).

708

According toLourens et al. (2004)the C3r base is 6.033 Ma old.

709

Finally, the recognized bioevents and the reversal boundary

sug-710

gest astronomically calibrated ages for the 34–14.5 m interval,

corre-711

sponding to the 6.122–6.033 Ma time interval (Fig. 11). From 14.5 m

712

(6.033 Ma) to the base of the lowermost shale/limestone laminated D A CB E F HG I J K L NNbis M P Q * 0 5 20 25 30 35 40 45 50 10 15 Kelliella abyssicola Brachyura gen. sp. ind. Ostrea cf. lamellosa Acanthocardia sp. Thracia sp. Turritella sp. Malletiidae gen. sp. ind. Trochidae sp. ind. Cuspidaria sp. M2 M5 M6 M7 M1 M3 M4

Fig. 8. Macrofossils from selected levels of the Legnagnone section. Intervals M1 to M7 are based on the vertical distribution of macrofossil assemblages.