v
A box model study of the Late Miocene Mediterranean
Strontium isotopes and evaporites
R.P.M. Topper 1 , P.Th. Meijer 1 , R. Flecker 2 and M.J.R. Wortel 1
Utrecht University, The Netherlands; e-mail: topper@geo.uu.nl;
2Bristol University, UK
Pre−MSC
LE HL
1
100
101
102
103
104
105
106
g
WA
[m
3
/s per g/l]
0.5 1.0 1.5 2.0
fR
50 100 150 200 250 300
Salinity [g/l]
Pre−MSC
LE HL
1
100
101
102
103
104
105
106
g
WA
[m
3
/s per g/l]
0.5 1.0 1.5 2.0
fR
0.7084 0.7086 0.7088 0.7090
87 Sr/ 86 Sr
1
100
101
102
103
104
105
106
g
WA
[m
3
/s per g/l]
0.5 1.0 1.5 2.0
fR
0.7084 0.7086 0.7088 0.7090
87 Sr/ 86 Sr
Rhone
Nile (0.706)
LE
Halite
UE
0.7085
0.7086 0.7087 0.7088 0.7089 0.7090 0.7091 0.7092
87
Sr/
86
Sr
4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0
Age [Ma]
Ocean curve
Sardella/M.dei Corvi, Italy
DSDP, Balearic
Pissouri/Cyprus
Gavdos, Greece
DSDP, Ionian
DSDP, Levantine
Northern Apennines, Italy
M. del Casino, Italy
San Severino, Italy
Sicily
Sorbas Basin, Spain
Southern Turkey
Tyrrhenian Sea
Vena del Gesso, Italy
Vera Basin, Spain
MSC
Time to gyp- sum saturation.
Shaded areas reach a gypsum concen- tration > 95% of the gypsum
saturation value.
(left) no stratification (right) with stratification
Modelled halite thickness as a layer with constant thickness
spread over the whole water covered area of the
Mediterranean.
(left) no stratification (right) with stratification
Riverine Oceanic
Salinity and
87Sr/
86Sr results without (above) and with (be- low) interpretation from a model with a late Miocene geo- metry and water budget as a function of gateway size (g
WA)
and river input (fR)
1
100
101
102
103
104
105
106
g
WA
[m
3
/s per g/l]
0.5 1.0 1.5 2.0
fR
50 100 150 200 250 300
Salinity [g/l]
Increasing river input in Med Increasing size of connection between Atlantic and Med
g WA g EW
E-P R Rhone E-P R Nile
Atl WMed EMed
Q
AW
Q
WA Q
WE
Q
EW
S
W SURF
S
W DEEP
S
E SURF
S
E DEEP
S
A
Increasing size of connection between WMed and EMed (Sicily Strait)
Increasing size of connection between Atlantic and WMed
Increasing river input in WMed
Increasing river input in EMed
2 5 10 20 50 100 200 500 1000
0.5
1.0
1.5
2.0
fR East
0.5
1.0
1.5
2.0
fREast
0.5
1.0
1.5
2.0
fR East
0.5
1.0
1.5
2.0
fREast
0.5
1.0
1.5
2.0
fR East
0.5 1.0 1.5 2.0
fR
West 0.5 1.0 1.5 2.0
fR
West 0.5 1.0 1.5 2.0
fR
West 0.5 1.0 1.5 2.0
fR
West 0.5 1.0 1.5 2.0
fR
West 0.5 1.0 1.5 2.0
fR
West
0.5 1.0 1.5 2.0
fR
West
0.5
1.0
1.5
2.0
fR East
0.5 1.0 1.5 2.0
fR
West
0.5
1.0
1.5
2.0
fR East
0.5 1.0 1.5 2.0
fR
West
2 5 10 20 50 100 200 500 1000 kyr
0.5
1.0
1.5
2.0
fREast
0.5
1.0
1.5
2.0
fREast
0.5
1.0
1.5
2.0
fREast
0.5
1.0
1.5
2.0
fREast
0.5
1.0
1.5
2.0
fREast
0.5 1.0 1.5 2.0
fR
West 0.5 1.0 1.5 2.0
fR
West 0.5 1.0 1.5 2.0
fR
West 0.5 1.0 1.5 2.0
fR
West 0.5 1.0 1.5 2.0
fR
West 0.5 1.0 1.5 2.0
fR
West 0.5 1.0 1.5 2.0
fR
West
0.5
1.0
1.5
2.0
fREast
0.5 1.0 1.5 2.0
fR
West
0.5
1.0
1.5
2.0
fREast
0.5 1.0 1.5 2.0
fR
West
0 50 100 150 200 250 300 350 400 450 500 m
0.5
1.0
1.5
2.0
fREast
0.5
1.0
1.5
2.0
fREast
0.5
1.0
1.5
2.0
fREast
0.5 1.0 1.5 2.0
fR
West 0.5 1.0 1.5 2.0
fR
West 0.5 1.0 1.5 2.0
fR
West 0.5 1.0 1.5 2.0
fR
West 0.5 1.0 1.5 2.0
fR
West 0.5 1.0 1.5 2.0
fR
West 0.5 1.0 1.5 2.0
fR
West
0.5
1.0
1.5
2.0
fREast
0.5 1.0 1.5 2.0
fR
West
0.5
1.0
1.5
2.0
fREast
0.5 1.0 1.5 2.0
fR
West
0 50 100 150 200 250 300 350 400 450 500 m
0.5
1.0
1.5
2.0
fREast
0.5
1.0
1.5
2.0
fREast
0.5
1.0
1.5
2.0
fREast
0.5 1.0 1.5 2.0
fR
West 0.5 1.0 1.5 2.0
fR
West 0.5 1.0 1.5 2.0
fR
West 0.5 1.0 1.5 2.0
fR
West 0.5 1.0 1.5 2.0
fR
West 0.5 1.0 1.5 2.0
fR
West 0.5 1.0 1.5 2.0
fR
West
0.5
1.0
1.5
2.0
fREast
0.5 1.0 1.5 2.0
fR
West
0.5
1.0
1.5
2.0
fREast
0.5 1.0 1.5 2.0
fR
West
−25 −20 −15 −10 −5 0 5 10 15 20 kyr
0.5
1.0
1.5
2.0
fR East
0.5
1.0
1.5
2.0
fR East
0.5 1.0 1.5 2.0
fR
West 0.5 1.0 1.5 2.0
fR
West 0.5 1.0 1.5 2.0
fR
West 0.5 1.0 1.5 2.0
fR
West
0.5 1.0 1.5 2.0
fR
West
−25 −20 −15 −10 −5 0 5 10 15 20 kyr
0.5
1.0
1.5
2.0
fR East
0.5
1.0
1.5
2.0
fR East
0.5 1.0 1.5 2.0
fR
West 0.5 1.0 1.5 2.0
fR
West 0.5 1.0 1.5 2.0
fR
West 0.5 1.0 1.5 2.0
fR
West
0.5 1.0 1.5 2.0
fR
West
Difference in time to saturation between the WMed and EMed
t
SAT WMed-t
SAT EMed
(left) no stratification (right) with stratification
1. Introduction
Many aspects of the Mediterranean massive evaporite layers, deposited during the Messinian Salinity Crisis (MSC), remain controversial.
Firstly, combining strontium isotope and salinity data with results from a box model, we can investigate if and how the fresh water budget and the size of the Atlantic-Mediterranean gateway varied during the Late Miocene.
Secondly, we examine the spatial and temporal evolution of salinity in the Mediterranean. We focus on the possibility of reaching gypsum saturation with different fresh water budgets and strait sizes, the difference in timing of onset of gypsum deposition between the WMed and EMed, and the
thickness of halite deposits during the MSC.
2. Model
Sr concentration, Sr-isotope ratio and a Late Miocene water budget are incor- porated in a 2-box model which quantifies the water and salt budget of the Me- diterranean basins under specified conditions of connectivity between the bas- ins, with the Atlantic ocean and atmospheric fluxes. Outflow from each basin is linearly dependent on the salinity contrast between the basins; Q = g (S 2 -S 1 ), where g is the linear exchange coefficient. g can be varied between 10 6 and 10 1 , corresponding to an open and very restricted gateway, respectively. Inflow compensates for outflow and fresh water deficit in order to keep the sea level constant.
The model allows for examination of the temporal evolution of salinity and Sr- isotope ratio as a function of the individual hydrologic fluxes (Atlantic in and out- flow, river input and evaporation).
Gypsum deposition commences in a basin once the salinity in one of the layers exceeds 145 g/l, halite deposition starts above 350 g/l.
3. Strontium isotopes
The Late Miocene Sr-curve
Pre-MSC 87 Sr/ 86 Sr values only de- viate from coeval oceanic values (shaded area) in marginal basins and the Tyrrhenian Sea. During the MSC 87 Sr/ 86 Sr values drop signifi- cantly from ocean values to 0.7085 in the upper evaporites. After the Zanclean reflooding 87 Sr/ 86 Sr va- lues shift back to oceanic values.
4. Evaporites: differences in the Mediterranean
While the observed thickness of gypsum in the WMed and EMed is simi- lar, the halite thickness in the EMed is more than twice that of the WMed.
These differences are as yet unexplained.
The time to reach saturation in each of the Mediterranean basins depends on the fresh water budget and the size of the gateways (Gibraltar arc and Sicily).
In general, a smaller connection with the Atlantic and less river input lead to faster increasing salinity in the Mediterranean basins. The size of the Strait of Sicily de- termines the distribution of salt between
WMed and EMed, a large connection leads
to a more even distribution and slower salinity increase in the EMed.
Synchronous onset of the MSC is evidenced by a similar age of the first LE gypsum bed in both WMed and EMed marginal sequences. Taking into account the uncertainties associated with the dating, gypsum saturation
should be reached with a maximum difference of ~20 kyr between the basins. The pa- rameter range in which gypsum saturation is reached in both basins and within 20 kyr of each other is very limited, and restricted to scenarios with a large Sicily strait.
Once halite saturation is reached in one of the Mediterranean basins, we can calculate the amount of halite deposition in the next 50 kyr to get an estimate of
possible halite accumulation during the second MSC stage. Only within a very limited parameter range is halite deposition in both basins possible.
5. Evaporites: the role of stratification
Stratification of the water column changes the vertical distribution of salt; the surface layer becomes fresher, the deep layer saltier.
With increasing stratification, surficial salt transport between the basins be- comes less, due to lower surface layer salinities, while deep layer transport
increases. This results in slower salinity increase in the EMed and a higher steady state salinity in the WMed.
With increasing stratification and a larger Sicily strait, the possibility of gypsum deposition in the WMed increa- ses.
Where the parameter range with gypsum saturation in both basins within 20 kyr of each other was very limited without stratification, it has significantly increased with stratification, especially when the Sicily strait is large.
Compared to the no stratification results the EMed halite thickness generally decreases whereas it increases in the WMed with stratification.
When comparing model results with field observations, one must keep in mind that modelled thicknesses are minimum estimates because deposition did not take pla- ce in the whole Mediterranean. In a smaller depositional area evaporite thicknesses would be larger.
6. Conclusions
- Simple gateway restriction can cause the onset of the MSC.
- The range of water budgets with which gypsum formation in both Mediterranean basins is possible, is small. It is largest with significant water column stratification and a large connection between WMed and EMed.
- The range of possible evaporite thicknesses, as modelled for the HL phase of the MSC, is large. There are many possible combinations of fresh water budget, gateway sizes and stratification with which observed thicknesses can be reproduced.
Model results
Salinity and 87 Sr/ 86 Sr values are known for three Late Miocene time intervals;
pre-MSC, Lower Evaporites (LE) and halite (HL). Locating the correspon-
ding range in our model results we ob- tain an estimate of fresh water budget and gateway restriction in these inter- vals.
Pre-MSC: open marine salinities and
87 Sr/ 86 Sr values correspond to a model with non-restricted gateway exchange with the Atlantic.
MSC (LE): gypsum saturation and
87 Sr/ 86 Sr values below the oceanic
range correspond to a model with res- tricted gateways and an average river input and evaporation
MSC (HL): halite saturation and low
87 Sr/ 86 Sr values correspond to a mo-
del with almost closed gateways and a slightly wetter climate.
The pathway between these stages
delineates simple gateway restriction
with a slightly wetter climate during the
HL stage.