Reconstructing the topography and water level of the Mediterranean Sea during the Messinian Salinity Crisis

Resumen   Abstract   Índice   Conclusiones


Heida, Hanneke

2023-A
Descargar PDF  



Resumen

Durante la Crisis de Salinidad del Messiniense (MSC, 5,97 – 5,33 Ma), una crisis ambiental sin precedentes en la historia geológica reciente, se depositaron gruesas evaporitas en el Mar Mediterráneo asociadas con una gran erosión de los márgenes continentales. Se cree que el MSC provocó una caída del nivel del agua alrededor de un kilómetro debido a la evaporación provocada debido a la restricción en la conexión marina entre el Atlántico y el Mediterráneo. Debido a la incertidumbre en los movimientos verticales posteriores a la MSC y la falta de correlaciones claras entre los registros sedimentarios marginales y abisales, la amplitud de esta caída y su ubicación en el tiempo siguen estando en discusión.
En esta tesis, mi objetivo es determinar el nivel del agua del Mediterráneo durante la MSC proporcionando una reconstrucción paleobatimétrica de las subcuencas mediterráneas y estimaciones de profundidad para el emplazamiento de depósitos de evaporita y marcadores de erosión. Estimo la magnitud de los movimientos verticales inducidos por la acumulación de evaporita y otras unidades de sedimentos, el hundimiento isostático y térmico y la deformación tectónica en tres regiones clave, que son: la cuenca de Alborán, el resto del Mediterráneo occidental y el delta del Nilo.
En la Cuenca de Alborán (Capítulo 3), las terrazas erosionales se formaron originalmente en un amplio rango de profundidad. La terraza menos profunda se reconstruye a 250-550 m de profundidad, mientras que la terraza más profunda tiene un rango de profundidad reconstruido de 750-1500 m. Esta variación se interpreta como resultado de la fluctuación del nivel de agua durante la fase de evaporación y de un evento de inundación de alta energía de la cuenca. En el Mediterráneo occidental (Capítulo 4), utilizamos la halita de la “Unidad móvil” y el yeso de la “Unidad superior” del MSC como marcadores de las paleo-líneas costeras, y estimamos que se formaron a profundidades de 1500 m y 1100 m respectivamente. Además, la halita se encuentra en pequeñas cuencas restringidas originalmente hasta a una profundidad tan poco profundo como 500 m a lo largo del Promontorio Balear, lo que sugiere que la deposición de halita abarcó una profundidad más amplia que el sugerido por su conservación actual y posteriormente fue eliminada por la erosión durante la exposición subaérea durante la fase de descenso y nivel bajo de la crisis de salinidad de Messiniense. El modelo de caja basado en la física (Capítulo 6) de los flujos de agua y sal en la Depresión Central de Mallorca en el Promontorio Balear permite evaluar más a fondo esta hipótesis, mostrando que el yeso identificado en esta cuenca restringida solo podría haberse formado por la salinización general del Mediterráneo occidental en el nivel alto del agua, mientras que el volumen de halita sugiere que su precipitación comenzó solo después de que el nivel del agua había descendido al menos 850 m. Además, la nueva estimación del volumen de halita en el Mediterráneo occidental profundo concuerda con la precipitación que comienza con el inicio de la bajada del nivel del mar, en contraste con la del Mediterráneo oriental. En el Delta del Nilo (Capítulo 5), la restauración topográfica muestra la profundidad original del nivel base
geomorfológico del río Nilo a ~600 m por debajo del nivel actual del mar, con una cascada de 400 m que separa el cañón Messiniense río abajo del valle superior más antiguo. Esta caída del nivel base es de 2 a 4 veces menor que la determinada por otros criterios para la cuenca oriental, lo que nuevamente sugiere fluctuaciones y diacronismo de los episodios de erosión de MSC entre el Mediterráneo occidental y oriental.
Mi trabajo demuestra que la batimetría de las cuencas mediterráneas no fue radicalmente diferente de la actual en áreas no afectadas por la deformación tectónica de fallas o la subducción de placas. Se encontró que tanto las evaporitas como las evidencias de erosión se formaron en una amplia gama de profundidades, no vinculadas claramente a un único nivel base estable en toda la cuenca, sino más bien afectadas por los balances de agua fluctuantes en cada subcuenca. Propongo que estas variaciones temporales y entre subcuencas fueron impulsadas por variaciones en la escorrentía del continente y posiblemente por la captura de aguas del Paratethys, durante una etapa de desconexión total del Atlántico.



Abstract

During the Messinian Salinity Crisis (MSC, 5.97 – 5.33 Ma), an environmental crisis unparalleled in recent geological history, thick evaporites were deposited in the Mediterranean Basin associated with major erosion of the continental margins. The MSC is thought to have led to a kilometre-scale water level drop by evaporation due to restriction of the Atlantic-Mediterranean marine connection, but the timing and amplitude of this drop have remained controversial. This is due to uncertainty in the post-MSC vertical motions and lack of clear correlations between the marginal and abyssal sedimentary records.
In this thesis I aim at constraining the Messinian water level by way of providing a paleobathymetric reconstruction of the Mediterranean sub-basins and depth estimates for the emplacement of evaporite deposits and erosional markers. I constrain the magnitude of vertical motions induced by the accumulation of evaporite and other sediment units, isostatic and thermal subsidence, and tectonic deformation in three key regions, being: the Alboran Basin, the rest of the Western Mediterranean, and the Nile Delta.
In the Alboran Basin (Chapter 3), erosional terraces were formed originally at a wide depth range. The shallowest terrace is reconstructed to 250-550 m depth, while the deepest terrace has a reconstructed depth range of 750-1500 m. This variation is interpreted as the result of fluctuating water levels during the drawdown phase and to a high-energy basin reflooding event. In the Western Mediterranean (Chapter 4), we use the MSC “Mobile Unit” halite and “Upper Unit” gypsum as markers for paleoshorelines, and we estimate them as having formed at depths of 1500 m and 1100 m respectively. In addition, halite is found in small silled basins originally as shallow as 500 m along the Balearic Promontory, suggesting that halite deposition during the evaporative drawdown spanned a wider depth range than suggested by its current preservation, and was subsequently removed by erosion during subaerial exposure during the drawdown and lowstand phase of the Messinian Salinity Crisis. Physics- based box modelling (Chapter 6) of water and salt fluxes to the Central Mallorca Depression on the Balearic Promontory allow to further evaluate this hypothesis, showing that the gypsum identified in this silled basin could only have formed by overall salinification of the Western Mediterranean at high water level, while the volume of halite suggests that its precipitation started only after the water level had dropped by at least 850 m. Also, the new, smaller, estimate of halite volume in the deep Western Mediterranean is in agreement with precipitation starting at or soon after the onset of drawdown, in contrast to that in the Eastern Mediterranean. In the Nile Delta (Chapter 5) topographical restoration shows the original depth of the geomorphological base level of the Nile River at ~600-m below present sea level, with a 400 m waterfall separating the downstream Messinian canyon from the older upper valley. This baselevel drop is 2-4 times smaller than derived from other criteria for the Eastern basin, again suggesting fluctuations and diachronism of the MSC erosion episodes between the Western and Eastern Mediterranean.
I show that the bathymetry of the Mediterranean basins was not radically different from the modern day in areas unaffected by fault tectonic deformation or plate subduction. Both the evaporites and the erosional features are found to have formed at a wide range of depths, not clearly linked to a single stable basinwide base level but rather affected by fluctuating water budgets in each subbasin. I propose that these variations in time and between subbasins were driven by variations in runoff from the continent and possibly by the capture of Paratethyan waters, during a stage of complete disconnection from the Atlantic.



Índice

SUMMARY ………………………………………………………………………………………………………………………………….. 1
RESUMEN ……………………………………………………………………………………………………………………………………3
PUBLICATIONS …………………………………………………………………………………………………………………………….. 6
TABLE OF CONTENTS……………………………………………………………………………………………………………………..7
THE SALTGIANT ETN………………………………………………………………………………………………………………………9
CHAPTER 1: INTRODUCTION ………………………………………………………………………………………………………… 11
1.1. THE MESSINIAN SALINITY CRISIS …………………………………………………………………………………………………….. 12
1.2. OBJECTIVES ……………………………………………………………………………………………………………………………… 22
1.3. OUTLINE…………………………………………………………………………………………………………………………………. 23
CHAPTER 2: METHODS FOR THE RECONSTRUCTION OF MESSINIAN TOPOGRAPHY AND WATER LEVEL …….. 25
2.1. ISOSTASY AND LITHOSPHERIC FLEXURE………………………………………………………………………………………………. 26
2.2. BACKSTRIPPING…………………………………………………………………………………………………………………………. 31
2.3. TISC (NUMERICAL MODEL FOR FLEXURE CALCULATION) ………………………………………………………………………….. 34
2.4. TIME-DEPTH CONVERSION……………………………………………………………………………………………………………..35
2.5. EVIDENCE FOR SUBAERIAL EXPOSURE IN THE MEDITERRANEAN SEAFLOOR: DATA AND ASSUMPTIONS ………………………. 36
2.6. INPUT DATA GRIDS FOR THE ISOSTATIC CALCULATION ……………………………………………………………………………… 38
2.7. HYDROLOGICAL BOX MODELLING …………………………………………………………………………………………………….. 39
CHAPTER 3: THE ALBORAN BASIN …………………………………………………………………………………………………. 41
3.1. CHAPTER SUMMARY …………………………………………………………………………………………………………………… 42
3.2. INTRODUCTION …………………………………………………………………………………………………………………………. 43
3.3. DATA AND MODEL SETUP …………………………………………………………………………………………………………….. 49
3.4. RESULTS …………………………………………………………………………………………………………………………………. 56
3.5. DISCUSSION …………………………………………………………………………………………………………………………….. 62
3.6. CHAPTER CONCLUSIONS………………………………………………………………………………………………………………..68
CHAPTER 4: THE WESTERN MEDITERRANEAN BASINS EAST OF ALBORAN ……………………………………………. 69
4.1. CHAPTER SUMMARY …………………………………………………………………………………………………………………… 70
4.2. INTRODUCTION …………………………………………………………………………………………………………………………. 71
4.3. DATA AND MODEL SETUP …………………………………………………………………………………………………………….. 79
4.4. RESULTS …………………………………………………………………………………………………………………………………. 86
4.5. DISCUSSION …………………………………………………………………………………………………………………………….. 94
8 Table of Contents
4.6. CHAPTER CONCLUSIONS ……………………………………………………………………………………………………………..102
CHAPTER 5: THE NILE DELTA……………………………………………………………………………………………………….. 103
5.1. CHAPTER SUMMARY …………………………………………………………………………………………………………………. 104
5.2. INTRODUCTION ……………………………………………………………………………………………………………………….. 105
5.3. GEOLOGICAL OBSERVATIONS ……………………………………………………………………………………………………….. 106
5.4. DATA AND MODEL SETUP …………………………………………………………………………………………………………… 109
5.5. RESULTS ……………………………………………………………………………………………………………………………….. 109
5.6 DISCUSSION AND CONCLUSIONS ……………………………………………………………………………………………………. 115
CHAPTER 6: BOX MODELLING OF EVAPORITE DEPOSITION IN THE CENTRAL MALLORCA DEPRESSION…….. 117
6.1. CHAPTER SUMMARY …………………………………………………………………………………………………………………. 119
6.2. INTRODUCTION ……………………………………………………………………………………………………………………….. 120
6.3. DATA AND MODEL SETUP …………………………………………………………………………………………………………… 128
6.4. RESULTS ……………………………………………………………………………………………………………………………….. 134
6.5. DISCUSSION …………………………………………………………………………………………………………………………… 146
6.6. CHAPTER CONCLUSIONS …………………………………………………………………………………………………………….. 157
CHAPTER 7: GENERAL DISCUSSION………………………………………………………………………………………………. 159
7.1. WATER LEVELS DURING MSC: CONNECTING THE DOTS …………………………………………………………………………. 159
7.2. IMPLICATIONS FOR CONDITIONS DURING EVAPORITE DEPOSITION ………………………………………………………………165
7.3. WATER BUDGETS AND BASELEVEL VARIATION …………………………………………………………………………………….. 170
7.4. FUTURE PERSPECTIVE ……………………………………………………………………………………………………………….. 174
CHAPTER 8: GENERAL CONCLUSIONS …………………………………………………………………………………………… 176
ACKNOWLEDGEMENTS ……………………………………………………………………………………………………………… 180 BIBLIOGRAPHY…………………………………………………………………………………………………………………………. 181



Conclusiones

The aim of this thesis was to reach a better understanding of the timing and magnitude of water level variations during the Messinian Salinity Crisis, and their implications for conditions during evaporite deposition. This has been done through a quantitative reconstruction of the Mediterranean subbasins, constraining the depth of paleoshoreline markers from these reconstructions, and examining the plausible basin connectivity and environmental conditions that led to evaporite deposition.
The results show that the Western Mediterranean was of considerable depth before the onset of the MSC, similar to the modern bathymetry. Shoreline terraces were formed at a wide range of depths from 250 to 1500 m in the West Alboran Basin, although these cannot be linked univocally to the drawdown stage of the MSC and their formation by a flooding event cannot be excluded. The mostly submarine paleotopography obtained for the Alboran Volcanic Arc indicates that marine corridors should have allowed for water exchange between the West Alboran and the rest of the Mediterranean, and that therefore the shallower Gibraltar tectonic arc was the most likely position for the topographic barrier separating the Atlantic Ocean and the Mediterranean Sea.
In the Valencia Basin, the relationship of MSC stratigraphy and erosional surfaces allow identification of two paleoshoreline levels, one formed originally at -1500 m related to the halite “Mobile Unit” (MU) and the Bottom Erosion Surface, and one at -1100 m related to the “Upper Unit” (gypsum/anhydrite) deposits. This, like the variable terrace depth in the Alboran Basin, indicates variations in the West Mediterranean water levels from the onset of the lowstand stage to the eventual reflooding at the base of the Pliocene. The presence of halite in small silled basins at a wide depth range on the Balearic Promontory shows that halite was originally deposited in shallower regions than suggested by its modern preservation. However, it is unclear whether the shallow halite deposits were formed only in these silled basins due to local brine accumulation, or if halite was formed on all of the later exposed margins and subsequently eroded everywhere but in the silled basins.
In the Eastern Mediterranean, re-examination of the Nile Canyon allowed identification of a morphological expression of a shoreline in the form of a 400 m high knickzone formed as a response the drop in Messinian baselevel. Downstream from the knickzone, a gently sloping subaerial river canyon transitions into a steeper, subaqueous canyon, with this change in the character of the canyon geometry being interpreted as an expression of the paleoshoreline formed during the Messinian lowstand. This shoreline was reconstructed to a level 600 m below sea level, which implies a significantly larger water level drop than can be explained by eustatic sea-level variations, but this new estimate is 2-4 times smaller than previous water level reported based on the Nile
Canyon incision. The difference between the Nile Canyon paleoshoreline depth and previous estimates based on coastal erosion features on the Malta Escarpment and the potentially subaerially exposed Nahr Menashe deposit in the deep Levant Basin suggests, similarly to the Western Mediterranean, strong changes in water level during the lowstand stage of the Messinian during which the connection to the global ocean was closed, probably controlled by climatic-induced changes in the water budget of each subbasin.
The box-model of the hydrological and salinity budget of the Central Mallorca Depression based on the restored Messinian bathymetry illustrates how the Stage 1 gypsum deposits in this unique basin formed over a large depth range under high water level, in a Mediterranean that had surface waters with high salinity. In contrast, based on its preserved volume the halite observed in the CMD was formed in a basin undergoing significant drawdown, suggesting that halite saturation was reached when the CMD disconnected from the surrounding deep basin when the water level drop reached approximately -800 m. However, the presence of the aforementioned halite basins on the Balearic Promontory suggests that at least in some areas halite saturation was reached already at higher water level, before the drawdown reached -500 m depth relative to the global sea level. The box model also shows that the entire halite volume in the Western Mediterranean could potentially be formed during the drawdown from the brine present in the basin at the time of disconnection.
Based on the results and ideas presented above I tentatively propose the following model for the evolution of Mediterranean water level and connectivity, illustrated in Figure 8.1:
– After deposition of the Primary Lower Gypsum in a full Mediterranean, progressive restriction at the strait of Gibraltar initiated Stage 2 with the complete disconnection between the Atlantic Ocean and Mediterranean Sea and a major water level drop down to -1500 m below global sea level during which the majority of the Western Mediterranean halite volume was deposited. In the Eastern Mediterranean the initial water level drop might have been even larger, potentially reaching -1900 m. Rebound in response to the water level drop at the Gibraltar Arc region would prevent the reconnection of the Atlantic inflow by eustatic sea-level variation and initiate a ~220 kyr phase of endorheism in the Mediterranean.
– After the initial drawdown, Stage 3 is characterized by erosion at the margins and deposition of the Upper Unit evaporites indicating both lateral and temporal variation in water level. In the Eastern Mediterranean, a long-lived base level is recorded at -600 m below global sea level potentially related to the depth of the Sicily Sill, but water level might have periodically fluctuated depending on the water budget of the Paratethys and East Mediterranean. In the Western Mediterranean, a multi-precession cycle water level is recorded in the UU of the Valencia Basin at -1100 m, and variable water levels are recorded in erosional terraces of the Alboran Basin. The lower water level in the Western Mediterranean compared to the East can be related to lack of large plausible catchment areas that would have provided extra fresh water to this basin after isolation, and therefore the water level would vary depending on climate and periodic overtopping of the Eastern Basin. Normal marine conditions are established only after reopening of the Atlantic-Mediterranean connection at the Strait of Gibraltar and the reflooding of the Mediterranean.