14

15

Πανεπιστήμιο Πατρών, Τμήμα Γεωλογίας

Fluid Seepage On & Offshore Western Greece

Σταυρούλα Κορδέλλα

Διδακτορική Διατριβή

© 2021 - Με την επιφύλαξη παντός δικαιώματος

16

17

To my parents, Yiannis & Eftychia

18

19

"The approval of the present thesis by the Examining Committee and the Department does not presuppose the acceptance of the author's views." (Law 5343/1932, Article 202)

Advisory Committee Dr. George Papatheodorou, Professor (Supervisor), Department of Geology, University of Patras, Greece Dr. George Ferentinos, Emeritus Professor, Department of Geology, University of Patras, Greece Dr. Nikolaos Lambrakis, Professor, Department of Geology, University of Patras, Greece

Examining Committee Dr. Giuseppe Etiope, Istituto Nazionale di Geofisica e Vulcanologia, Rome, Italy &Faculty of Environmental Science and Engineering, Babes-Bolyai University, Cluj-Napoca, Romania Dr. Maria Geraga, Assοciate Professor, Department of Geology, University of Patras, Greece Dr. Avraam Zelilidis, Professor, Department of Geology, University of Patras, Greece Dr. Dimitris Sakellariou, Research Director, Hellenic Centre for Marine Research, Athens, Greece

20

Aknowledgements I would like to express my gratitude to all the people that supported me in many ways during this voyage. My PhD supervisor Dr. George Papatheodorou, the head of our Lab, was the first to open a new window in to the world for me with a PhD candidate position and a place in one of the best teams, the Laboratory of Marine Geology and Physical Oceanography (LMGPO), which soon became my home. Professor Dr. George Papatheodorou, is an inspiring professor and researcher with a charismatic personality, who has motivated, encouraged, and tirelessly supported me since I started as an MSc student. He offered me the opportunity to participate in numerous multi-objective exhilarating projects that molded my scientific and professional character. I will be always grateful since he was the reason I decided to pursue a career in research and he made my life-long dream to become an oceanographer come true. I am eternaly grateful to Dr. Giuseppe Etiope, who first, believed in me and trusted me with precious geochemical data, and soon after became my “deus ex machina” by kindly helping me overcome any obstacle instantaneously. He kindled the “eternal flame” of my love for science and spurred my interest in the multi-dimensional world of gas geochemistry. The times that he provided me with priceless scientific advice are countless, as well as our insightful discussions throughout the past decade. He is my primary resource for getting my gas geochemistry questions answered and was instrumental in helping me complete this thesis. I will forever be thankful to Em. Prof. Dr. George Ferentinos, the founder of LMGPO, for providing me scientific and moral support, endless inspiration and encouragement. His positive outlook in life and his boundless scientific curiosity will always constitute him a role model for me. I am grateful for having by my side Professor Dr. Maria Geraga; the heart of LMGPO. Professor Maria Geraga is not only a rigorous scientist and inspiring professor, but also the kindest person I know; 21

always willing to offer her scientific and personal support, which I have received countless times over the years. Many thanks go to Professor Dr. Nikolaos Lambrakis, who has been always helpful and willing to show his support providing scientific advice and knowledge throughout the past decade. I am grateful for having in my examining committee Dr. Dimitris Sakellariou an inspiring and excellent researcher who provided me with inspiration as well as helpful scientific advice and suggestions. I am thankful to Professor Dr. Avraam Zelilidis, for his scientific work formed my understanding of the geologic setting of Western Greece in relation to gas geochemistry, as well as for his numerous helpful suggestions and comments that improved this thesis. Many thanks go to Associate Professor Dr. Sotiris Kokkalas from the Department of Geology, University of Patras, who co-authored one of the papers of my thesis, and his contribution significantly improved this work. I am thankful to Dr. Giancarlo Ciotoli and Dr. Giuditta Marinaro, from INGV, Rome, for being excellent collaborators and for co-authoring the scientific papers that were part of my thesis. Each one of them provided added value to this work, and I could not thank them enough. I could not thank enough my colleagues and dear friends, members of the Laboratory of Marine Geology and Physical Oceanography: Dr. Dimitris Christodoulou, Dr. Elias Fakiris, Dr. Margarita Iatrou, Mr. Nikos Georgiou, Mr. Xenophon Dimas, Mr. Michalis Prevenios, Dr. Despina Zoura, and Dr. Dimitrios Eleftherakis. Dr. Dimitris Christodoulou had instrumental role in my thesis’ field work. Dr. Elias Fakiris and Dr. Margarita Iatrou also participated in the field work and together with Dr Christodoulou they provided plentiful help with the software for data elaborations. Dr. Despina Zoura, Mr. Nikos Georgiou, Mr. Xenophon Dimas, Mr. Michalis Prevenios participated in the field work, and Dr. Dimitrios Eleftherakis kindly supported me during the last months of my thesis. All of them were generous and supportive in multiple ways, and most of all, made this journey joyful. 22

Thanks go to the captain of R/S Eirini, Mr. Aggelos Christopoulos and crew, as well as the divers Dr. Sotiris Kyparissis and Mr. Nikos Nikolopoulos for their excellent work and collaboration. I also thank people from the University of Patras who have supported me in different ways over the years. I am thankful to Professor Dr. Kostas Nikolakopoulos from the Department of Geology, who provided me with valuable scientific advice through our collaboration in projects outside my thesis. I am grateful to my dear friend and creative scientist, Associate Professor Dr. Hrissi Karapanagioti from the Department of Chemistry, with whom we collaborated in numerous projects and supported me scientifically and personally through many difficult situations. I am thankful to the staff of the Department of Geology, University of Patras. Especially the Head Secretary Ms. Andriana Lambropoulou and Secretary Ms. Sophia Bakopoulou for always being so helpful and friendly. Also, I would like to thank collaborators from the Hellinic Center for Marine Research, who supported me. Thanks go to Dr. Vasileios Papadopoulos who helped me promptly with MATLAB software experimentations, and Ms. Amalia Venetsanopoulou for her support and encouragement. I get by with a little help from my friends to whom I am always grateful (you know who you are!) and beyond grateful to my closest friends Kyriaki Krali and Eftychia Mavrou, my shining stars Nikolas Paschalis, Magdalini Paschali, and my mentor Charalampos Themistokleous for making my life a little heaven. I would not have made it this far without the support of my family. I thank my brothers Haris and Christos for their support, and my nieces Ioanna and Maya who make me smile without fail. Above all, I thank my parents Yiannis and Eftychia, who sacrificed their lives for our family and provided unconditional love.

23

Funding All offshore surveys were carried out within the framework of the HYPOX project (“In situ monitoring of oxygen depletion in hypoxic ecosystems of coastal and open seas, and landlocked water bodies”, EC Grant 226126, ENV.2008.4.1.2.1, FP7), co-funded by the European Union. The core objective of the EU-FP7 project HYPOX was to better understand the causes of hypoxia formation and improve the predictive capability of models. To this end, HYPOX initiated a comprehensive, state-of-the-art program for monitoring oxygen and related parameters in a variety of European aquatic systems that differ in oxygen status or in their sensitivity to change. HYPOX working areas are subject to a large variety of natural and anthropogenic causes of oxygen depletion and show a wide range in the severity of oxygen deficiency. They are located in coastal and open seas, including the North Atlantic–Arctic Ocean transition (Fram Strait), four contrasting sites in the Black Sea (Bosporus outlet area, Romanian Shelf, Crimean Shelf, Black Sea basin), and two Baltic Sea sites (Gotland Basin and Boknis Eck in Eckernförde Bay). Several lagoons and embayments in the Mediterranean Greek Ionian Sea (Katakolo Bay, Amvrakikos Gulf, and Aetoliko Lagoon) were also studied, as well as fjord systems (Koljoe Fjord in Sweden and Loch Etive in Scotland).

24

Chapter description and publications in scientific journals The present thesis includes 6 main chapters. Chapter 1 describes concisely the overall objectives of this thesis. Chapter 2 consists of an overview of theoretical evaluations and concepts necessary for the comprehension of on and offshore fluid seepage, as well as previous investigations on seabed fluid flows on and offshore Western Greece. Chapter 3 contains the description of the areas that were investigated in this thesis, including a concise description of the geological setting of the general area; Western Greece. In Chapter 4 the methods applied for the accomplishment of this thesis are presented. Chapter 5 consists of the experimental study of this work which took place in Katakolo Bay and Amvrakikos Gulf over the past 10 years. The major part of these studies has been published in scientific journals. Subchapter 5.1.1 presents part of the work published in the journal Chemical Geology (IF: 3.362) “Etiope, G., Christodoulou, D., Kordella, S., Marinaro, G., Papatheodorou, G. (2013). Offshore and onshore seepage of thermogenic gas at Katakolo Bay (Western Greece). Chem Geol, 339, pp. 115-126. https://doi.org/10.1016/j.chemgeo.2012.08.011”. The objective of this work was to investigate the extensive seepage of natural gas on and offshore Katakolo Bay, (Western Greece), through an integrated offshore and onshore survey, which included a marine geophysical survey, underwater exploration by a towed instrumented system, compositional and isotopic analyses, and gas flux measurements. Subchapter 5.1.2 includes the full published work in the journal Applied Geochemistry (IF: 2.903), “Kordella, S., Ciotoli, G., Dimas, X., Papatheodorou, G., Etiope, G. (2020). Increased methane emission from natural gas seepage at Katakolo Harbour (Western Greece). Appl 25

Geochem,116, 104578. https://doi.org/10.1016/j.apgeochem.2020.104578)”. In this paper we reported an about four times sub-decadal increase of methane seepage from one of the largest thermogenic gas seep sites in Europe, Katakolo Harbour, based on gas flux measurements by accumulation chamber performed in 2010 and 2018. This finding is crucial because geological gas seepage is an important natural source of atmospheric methane which generally considered constant over time in methane budget models. In this work multiple lines of evidence suggested that the methane emission increased mainly due to intensification of subsurface gas flow (seepage) after 2010. If similar short-term variations of emission factor occur in large seepage areas worldwide, the global geological methane emission can significantly change, contributing to decadal changes of atmospheric methane budget. In Subchapter 5.2.1 the work presented is published in the journal Geosciences (Site score: 2.1, Q2) “Kordella, S., Christodoulou, D., Fakiris, E., Geraga, M., Kokkalas, S., Marinaro, G., Iatrou, M., Ferentinos, G., Papatheodorou, G. (2020). Gas seepage-induced features in the hypoxic/anoxic, shallow, marine environment of Amfilochia Bay, Amvrakikos Gulf (Western Greece)” (https://doi.org/10.3390/geosciences11010027). The paper includes the integrated study of seabed seepage manifestations and the exploration of possible inter-relationships between shallow gas accumulations and hypoxia in Amfilochia Bay (Eastern Amvrakikos Gulf, Western Greece), a complex marine area affected by tectonism. For this an integrated research methodology that combined geophysical, geochemical, and hydrographic surveys was employed. Marine geophysical and bathymetric surveys led to the discovery of a pockmark group with weak CH4 seepage that slightly increased hypoxia locally, and a protrusion mound, erroneously reported as a submarine “volcano” since 1876, which could be the result of mud volcanism based on this study. 26

Unpublished, yet, experimental studies are presented in: (i) Subchapter 5.1.3 which includes the temporal offshore monitoring study that took place in Katakolo Bay with the employment of a benthic observatory (Gas Monitoring Module) that operated underwater over intense thermogenic gas seepage for 101 days; and (ii) Subchapter 5.2.2 that includes the investigation of Sogono in Amvrakikos Gulf with the same integrated methodological scheme that was applied in Amfilochia Bay. In Chapter 6 the conclusions of this thesis are presented.

27

Table of Contents 1. Objectives ...... 31 2. Theoretical evaluations ...... 34 2.1 What is fluid seepage ...... 34 2.2 Types of Natural Gas from Seepages ...... 36 2.3 Seabed fluid flow ...... 38 2.4 Seepage Geochemistry Applications ...... 47 2.4.1 Seepage as a tool for hydrocarbon exploration ...... 47 2.4.2 Environmental impact ...... 48 2.4.3 Hypoxia and its links to fluid seepage ...... 48 2.4.4 The greenhouse effect and its links to methane seepage 50 2.5 Review of seabed fluid flow manifestations offshore (and onshore) Western Greece ...... 51 References ...... 72 3. Geological setting ...... 91 4. General Geological Setting of Western Greece ...... 91 References ...... 96 5. Material and Methods ...... 100 5.1 Conseptual methodological scheme and survey design .. 100 5.2 Geophysical means ...... 102 5.2.1 Sonar mapping systems - acoustic remote sensing ..... 102 5.2.2 Single-beam echo-sounders ...... 103 5.2.3 Multibeam echo-sounders ...... 104 5.2.4 Side scan sonars ...... 105 5.2.5 Hydroacoustical gas bubble plume detection with the use of side scan sonar. An example from Katakolo Bay ...... 106 5.2.6 Subbottom Profilers ...... 108 28

5.2.7 Visual seafloor observation ...... 110 5.3 Multiparametric instruments developed for offshore seepage detection ...... 111 5.3.1 GMM ...... 112 5.3.2 MEDUSA ...... 112 5.4 Geochemical means ...... 113 5.4.1 Gas Detection Methods on land...... 113 5.4.2 Dissolved gas in seawater and underwater sediment sampling and analysis ...... 114 References ...... 115 6. Experimental study ...... 123 6.1 Katakolo ...... 123 6.1.1 Offshore and onshore seepage of thermogenic gas at Katakolo Bay (Western Greece) ...... 123 6.1.1.1 Introduction...... 123 6.1.1.2 Geologic setting and description of the seepage area 126 6.1.1.2.1 General geology and tectonics ...... 126 6.1.1.2.2 Hydrocarbon potential ...... 128 6.1.1.2.3 Surface seepage ...... 129 6.1.1.3 Methods ...... 130 6.1.1.3.1 Marine Remote Sensing Surveys ...... 130 6.1.1.3.2 Underwater Visual Inspection and Gas Detection .... 132 6.1.1.3.3 Gas sampling and analyses ...... 134 6.1.1.3.4 Gas Flux Measurements Offshore ...... 135 6.1.1.3.5 Gas Flux Measurements Onshore ...... 135 6.1.1.4 Results and Discussion ...... 136 6.1.1.4.1 Areal distribution of the seeps and relationship with local faults 136 29

6.1.1.4.2 Molecular Gas Composition ...... 140 6.1.1.4.3 Isotopic Gas Composition and Gas Origin ...... 141 6.1.1.4.4 Offshore Gas Flux ...... 145 6.1.1.4.5 Onshore gas flux ...... 147 6.1.1.5 Conclusions ...... 148 References ...... 149 6.1.2 Increased methane emission from natural gas seepage at Katakolo Harbour (Western Greece) ...... 157 6.1.2.1 Introduction...... 157 6.1.2.2 Study area and inspection of ground conditions ...... 158 6.1.2.2.1 Geological setting ...... 158 6.1.2.2.2 Description of gas release manifestations ...... 159 6.1.2.3 Materials and methods ...... 161 6.1.2.4 Results ...... 163 6.1.2.4.1 Methane emission from cracks and fissures ...... 163 6.1.2.4.2 Diffuse methane emission from unfractured asphalt 164 6.1.2.5 Discussion ...... 166 6.1.2.5.1 Seepage variation and possible causes ...... 166 6.1.2.5.2 Potential atmospheric impact of seepage variation . 169 6.1.2.6 Summary and conclusions ...... 170 References ...... 171 6.1.3 Long-term Offshore Monitoring in Katakolo Bay. The Gas Monitoring Module ...... 174 6.1.3.1 Materials and methods ...... 175 6.1.3.1.1 The GMM ...... 175 6.1.3.1.2 GMM deployment site ...... 179 6.1.3.1.3 GMM data preparation and processing ...... 181 30

6.1.3.2 Results ...... 182 6.1.3.3 Conclusions ...... 192 References ...... 193 6.2 Amvrakikos Gulf ...... 194 6.2.1 Gas seepage-induced features in the hypoxic/anoxic shallow marine environment of Amfilochia Bay, Amvrakikos Gulf (Western Greece) ...... 194 6.2.1.1 Introduction...... 194 6.2.1.2 Geological and physiographic setting of survey area 196 6.2.1.3 Methodology ...... 199 6.2.1.4 Data Presentation - Results ...... 203 6.2.1.4.1 Multibeam bathymetric data and seafloor morphology 204 6.2.1.4.2 Side Scan Sonar Data ...... 205 6.2.1.4.3 Seismic stratigraphy and gas-bearing acoustic characteristics ...... 207 6.2.1.4.4 The “volcano” of Amfilochia Bay ...... 216 6.2.1.4.5 Physicochemical parameters of the Amfilochia Bay water column ...... 218 6.2.1.4.6 MEDUSA underwater tow multi-parametric platform 220 6.2.1.4.7 Isotopic data from sediment and water samples ..... 222 6.2.1.5 Discussion ...... 222 References ...... 228 6.2.2 Sogono ...... 237 7. Synthesis...... 242 Supplementary Material for Chapter 5.1.2 “Increased Methane Emission from Natural Gas Seepage at Katakolo Harbour (Western Greece)” ...... 245 31

1. Objectives The present thesis is an integrated geological, geochemical, and oceanographic study of the environmental impact of hydrocarbon rich fluid seepage occurring in two areas located in Western Greece, Katakolo Bay and Amvrakikos Gulf. Western Greece is particularly affected by hydrocarbon seepage processes related to the presence of underground oil and gas reservoirs. The seepage is just the natural, upward migration of fluids from these reservoirs. Environmental impacts of hydrocarbon rich fluid seepage include: (a) the emission of greenhouse gases to the atmosphere, (b) hypoxia in aquatic systems (c) degradation of geotechnical properties of ground, soil, and sediments and, (d) geohazard due to explosiveness and toxicity of gas. This work is focused on points (a), (b) and (c).

(a) Methane (CH4), the primary component of natural gas, is a powerful greenhouse gas and its seepage to the surface is an important natural source of greenhouse gas for the atmosphere. The seepage may occur on land (on shore) or in the marine environment (offshore), always associated to the presence of tectonic faults, as documented in this work. Measuring the flux of methane (generally expressed in mg m- 2 per day, or tonnes per day) is the key task for addressing its atmospheric impact. (b) Hypoxia (i.e., oxygen depletion) in aquatic environments is an important environmental phenomenon with strong implications for life and its biodiversity, and as a result, aquatic ecosystem goods and services. Well oxygenated water bodies result from a fragile balance between production, transport and consumption. Hypoxia can be induced by natural and anthropogenic factors, such as water circulation, high water temperature, stratification of the water column, and anthropogenic pressures such as pollution. Hydrocarbon seepage is an additional natural factor that is yet poorly studied; it is known in theory 32

that oxygen dissolved in water is consumed to oxidize methane released from the seabed. Knowing the conditions that regulate this process is still elusive. (c) Pockmarks are cone shaped deep depressions that are the result of erosion/degradation of the marine sediments caused by escaping fluids from below the seafloor (Amfilochia Bay in this thesis). Pockmarks constitute evidence that may illustrate and justify submarine fluid flow. On land gas seepage may be the cause of degradation of ground conditions, such as asphalt deformation and cracking. This occurs because the asphalt and concrete pavement over a seepage site (Katakolo Harbour in this thesis) acts as a cap for gas that gradually accumulates below. When the gas pressure increases the asphalt and concrete may swell and crack. In this study gas emission and hypoxia are investigated through geochemical and oceanographic methods, also using modern instrumentation such as benthic observatories. Katakolo Bay and Amvrakikos Gulf represent two different environments, hosting different type of seepage, geology, and seawater current circulation: Katakolo Bay is an open sea environment, but it includes a semi-enclosed harbor area where water circulation is limited. Amvrakikos Gulf is a naturally semi-enclosed basin receiving limited input from the open sea. The specific objectives of this study were: • Assessment of seepage distribution in the two studied areas • Estimate overall emission of methane into the atmosphere • Monitor over time oxygen concentration in seawater in relation to seepage • Determine the mechanism of hypoxia evolution due to seepage The thesis is divided in two parts: -theoretical evaluation, describing general theory of hydrocarbon seepage and its environmental impact, with specific reference to hypoxia in seawater; 33

-experimental study reporting original, experimental data, some of them acquired in the framework of an EU funded project, HYPOX. The data include marine geophysical data, methane flux from the ground and from the seawater/seabed, methane and oxygen concentration in seawater from areal surveys and from a temporal monitoring in a selected site within Katakolo Port. A descriptive model of the seepage/hypoxia relationship is provided for the first time and can be used to study hypoxia in other marine environments.

34

2. Theoretical evaluations

2.1 What is fluid seepage Fluid seepage is the upward migration of hydrocarbon (mostly methane) rich fluids (i.e., gases and liquids) along faults from the Earth’s crust to the Earth’s surface. In geological sciences, seepage refers to fluids that consist either exclusively (gas phases) or partially (gas and groundwater, crude oil or mud) of natural gas seeping from geological sources.

Natural gas mainly consists of methane (CH4), with lesser amounts of light alkanes (e.g., C2H6, C3H8, C4H10), whilst CO2, N2, He, and H2S may also be present. Natural gas is produced in the Earth’s crust in petroleum-prone sedimentary basins, through microbial or thermal conversions of organic matter in source rocks, generally shales or limestones (Etiope, 2015; Hunt, 1996).

Methane (CH4) is the simplest of alkanes, consisting of one carbon atom (as it stated by its name based on the nomenclature of alkanes) and it has four covalent bounds with an equivalent number of hydrogen atoms. Methane is the most abundant gas accumulated in sedimentary rocks of the Earth’s crust, accounting for more than 85% of the natural gas. Liquid phases (two-phase release) in fluid seepage mainly refer to crude oil and groundwater or pore-water that contains dissolved gas and/or dispersed gas bubbles from subsurface sources to Earth’s surface (Etiope, 2015). Three-phase releases include fluidized mud with dissolved gas expelled from mud volcanoes. By reaching the Earth’s surface, natural gas is injected either continuously or intermittently from the lithosphere into the atmosphere or hydrosphere (underwater seepage).Natural gas, as all hydrocarbon gas accumulations, is mostly biotic in origin, as its principal precursors organic matter (lipids, carbohydrates), which are also the building blocks of petroleum and, of course, life. Marine and terrestrial organic matter is deposited in aquatic environments and then slowly 35

decomposed in sedimentary rocks in the Earth’s crust producing natural gas and oil. Apart from hydrocarbon gases, also non-hydrocarbon gases, such as CO2, N2, He, and H2S, may be present in small concentrations in fluid seepages, originating by both organic and inorganic processes (Hunt, 1996; Etiope, 2015). Natural gas has a totally different composition from geothermal or volcanic gas, where

H2O or CO2 are the major components and hydrocarbons have minimal presence (Etiope, 2015). The natural gas that seeps at the surface of the Earth’s crust has traveled several kilometers before reaching the Earth’s surface. Advection, which is the movement of fluids due to pressure gradients and gravitational energy and is expressed by Darcy’s Law, is the leading mechanism for gas seeps and micro-seepage (Hunt, 1996; Etiope, 2015). Darcy’s Law has been adapted to calculate the actual velocity that the fluids are moving in the subsurface in units such as distance traveled per year. This is called “seepage velocity” (VS) and is calculated by dividing the Darcy Velocity (VD) by the actual open pore area where the fluid is flowing, the “effective porosity” (ne) 푄 푉 = 퐷 퐴 Where:

VD = Darcy Velocity: flow per unit area (units of distance per time) Q = Flow rate (Volume of fluid flow per time, such as m3/yr) A = Cross sectional area perpendicular to groundwater flow (distance, such as m2)

푉퐷 푉푆 = ( ) 푛푒 Where:

VS = seepage velocity, units: distance per time

VD = Darcy Velocity: flow per unit area (units of distance per time) ne = Effective porosity (dimensionless) 36

The velocity of the gas movement (VS) is controlled by the permeability of the rock from which the fluid passes through and the pressure gradients induced by gas pressure at the starting point of migration. The study of a seepage emission can, therefore, reveal the permeability of the subsurface rocks and thus provide substantial information for structural geology. Moreover, faults and fractures facilitate fluid upward migration, thus, especially regarding geological methane, produced in deep source rocks, (e.g., Etiope et al., 2013a; Etiope et al., 2006), the presence of seeps is indicative of both thrust and normal faults, as gas chooses the conduits that exhibit the less resistance (preferential pathways) to migrate vertically to the surface of the Earth’s crust. The gas flow may be tectonically triggered, leading to increases in the volume of seeping gas and/or activation of seeps after intense tectonic activity, i.e., earthquakes (Christodoulou et al., 2009; Kelley et al., 1994; Hasiotis et al., 1996). Therefore, seeps are commonly present in proximity to active and permeable faults (e.g., Etiope, 2015; Etiope et al., 2013a; Klerkx et al., 2006; Etiope et al., 2007; Kutas et al., 2004; Etiope et al., 2006; Christodoulou et al., 2009).

2.2 Types of Natural Gas from Seepages Based upon the molecular and isotopic composition of methane, ethane, propane, and butane (Stahl, 1977; Bernard et al., 1978; Schoell, 1980; Chung et al., 1988; Schoell, 1988; Hunt, 1996; Whiticar, 1999; Milkov, 2011), geological natural gas can be categorized in three main classes microbial (e.g., Thiagarajan et al., 2020) thermogenic (e.g., Etiope et al., 2013a), and abiotic (e.g., Etiope et al., 2013b). Microbial or bacterial gas is produced by Archaea microbial communities at relatively low temperatures (typically up to 60-80 oC) (Hunt, 1996). Microbial gas, in most cases, is produced in relatively shallow (<2000 m) gas source rocks, and is very dry (i.e., depleted in

C2+ components). Thermogenic gas is produced in deeper source rocks 37

by the thermal cracking of organic matter or oil, normally up to 190- 200oC (Hunt, 1996) and consists of high quantities of methane with variable amounts of alkanes with longer chains (i.e., ethane propane, butane, etc.). Both of the abovementioned gases are termed as biotic, as it is stated previously, since they are produced by organic compounds of biotic origin. Bacterial methane is distinguished from thermogenic methane by its ‘light’ carbon isotope (higher 12C/13C ratio). The third type of natural gas is termed as abiotic since it is produced without any direct involvement of organic matter (Etiope and Schoell, 2014). This is a methane-rich gas that is formed in igneous and metamorphic formations (Etiope and Sherwood Lollar, 2013). Abiotic methane seepages have been discovered at the late 1980s (Abrajano et al., 1988) and have been reported ever since, in an accelerating pace throughout the world, in seeps and hyperalkaline springs located in ophiolites and peridotite massifs characterized by low temperature continental serpentinisation (e.g., Etiope et al., 2013b; Etiope et al., 2011; Etiope and Schoell, 2014; Fig. 2.2.1).

Fig. 2.2.1 One of the biggest onshore abiotic gas seep on Earth, “Chimaera” located in the ophiolitic outcrops of the Tekirova Unit in Turkey (Çirali, Antalya). Photo: Stavroula Kordella. 38

The abiotic synthesis of methane is attributed to metal-catalyzed, low temperature Fischer–Tropsch type reactions between carbon atoms in respective compounds and H2, such as the Sabatier reaction also known as methanation or hydrogenation of CO2 (i.e., 4H2 + CO2→ CH4

+ 2H2O; Sabatier and Senderens, 1902), and less commonly direct derivation from olivine hydration (serpentinization), magmatic sources and fluid inclusions (Etiope and Whiticar, 2019). Hydrogen (H2) is produced by serpentinization of ultramafic rocks that is the hydration of olivine and pyroxene minerals by seawater in benthic marine environment (e.g., Wanget al., 2018) or meteoric water on land (e.g., Marques et al., 2018). It should be clarified, in the present thesis, as in petroleum geology, the methane implicated in natural gas seepage is geological (fossil) that is completely different from modern, bio-chemically originating methane, which is produced by recent microbial activity (e.g., Iqbal et al., 2019). Geological methane is fossil, produced in deep source rocks and is radiocarbon (14C) free, which means that C is older than 50,000 years BP, and it is an energy resource for generating electricity and heating, powering transportation, and manufacturing products (Etiope, 2015). Fluid seepage carries messages from the Earth’s crust and its geochemical and geophysical analysis may provide knowledge on localization of hydrocarbon reservoirs and identification of shallow and deep source rocks. Geophysical and geotechnical studies of fluid seepage zones may be employed to assess marine geohazard risk (Gerivani et al., 2020; Hovland and Judd, 1988; Hovland et al., 2002; Ostanin et al., 2013; Etiope et al., 2012).

2.3 Seabed fluid flow Seabed fluid flow includes injection of natural gas and liquid phases from marine sediments to the seawater (Judd and Hovland, 2007). Several different types of fluids that escape from the seabed have been 39

surveyed and studied: (i) hydrothermal fluids (e.g., Tao et al., 2020); (ii) geological hydrocarbon, methane-rich gas of microbial, thermogenic or mixed (e.g., Thiagarajan et al., 2020) and abiotic origin (Sciarra et al., 2019); (iii) modern microbial (e.g., Deville et al., 2020); (iii) freshwater discharges (Christodoulou et al., 2003; Whiticar, 2002); (iv) porewater escape (Harrington, 1985); (v) gas and oil saturated water and mud discharges from mud volcanoes (e.g., Chen et al., 2020). At sea as on land, the term fluid seepage refers to methane-rich geological gas, as described in Chapter 1.1.1. Thus, the marine fluid flow cases discussed in this thesis relate only to submarine natural gas seeps and seepage-induced geological features, observed off the coast of Western Greece. The temperatures and the flow rates of hydrocarbon gas seeps on the seabed are lower in comparison to those observed in hydrothermal vents, for this reason they are also knows as “cold seeps”. Geological gas seeps are further distinct from hydrothermal vents, since the first are associated to consolidated sedimentary basins while the latter occur over young oceanic crust (Etiope, 2015). Gas seepage was initially studied extensively only in the marine environment (Judd, 2000; Etiope, 2015) except for several terrestrial mud volcanoes (e.g., Gorkun and Siryk, 1968). The extensive study of offshore gas seepage initiated due to concerns of the offshore oil and gas industry for accidents, such as blowouts due to pressure build up and explosions during drilling operations, which were not uncommon on offshore platforms (Narula, 2019; Etiope et al., 2015). Since then, numerous active submarine seeps have been reported and studied, in a wide variety of geological environments. Methane seeps have been discovered in continental shelves, the deep ocean-floor, the intervening slope (Hovland and Judd, 1988), in in open and semi-enclosed shallow gulfs and lagoons (Hasiotis et al., 1996, Rogers et al., 2006), in mid- latitude estuaries (called rias in the Iberian margin) (Garcia-Garcia et al., 1999; Ussler et al., 2003), and even at the deepest part of the ocean; the Marianna Trench (Kojima and Watanabe, 2015). Natural gas 40

and other related fluids accumulations on the marine sediments may alter locally the morphology of the seabed, forming features like submarine mud volcanoes (Chen et al., 2020; Pierre et al., 2014; Neurauter and Roberts, 1994; Milkov, 2000), carbonate mounds (Roberts and Ballard, 2001; Roberts et al., 1990) and pockmarks (e.g., Gullapalli et al., 2019; Christodoulou et al., 2009; Whiticar, 2002; Hasiotis et al., 1996; Papatheodorou et al., 1993). Gas hydrates are ice- like crystalline solids (Milkov, 2000). They form when methane and water is combined under conditions of high pressure, low temperature and high methane concentrations. Gas hydrates occur naturally in marine sediments affected by gas seepage when the abovementioned conditions are met (e.g., Dumke et al., 2016; Vadakkepuliyambatta et al., 2013; Davy, 2010; Westbrook et al., 2009; Bohrmann and Torres, 2006). Gas hydrates lie beneath numerous submarine slopes and have contributed to triggering submarine slides when decomposing, e.g., the Cape Fear slide (Popenoe et al., 1993). Other natural gas seepage geomorphological features are the columnar and acoustically transparent sections referred to as ‘gas chimneys’ (e.g., Vadakkepuliyambatta et al., 2013; Plaza-Faverola et al., 2011; Loncke and Mascle, 2004). Fluid seepage is associated with geohazards such as slope failures and landslides (e.g., Gerivani et al., 2020; Deville et al., 2020, Sahling et al., 2008; Prior and Coleman, 1984) that may pose a threat to human activities (Sultan et al., 2004, Elverhøi et al., 2002). The Storegga slide, probably the largest submarine sediment slide, was most probably initiated by the melting of gas hydrates that lie under the slide scour (Sultan et al., 2004). Papatheodorou and Ferentinos (1997) also reported that liquefaction, which was the dominant instability mechanism of earthquake-induced sediment failures in Holocene fan deltas in the Gulf of Corinth, was probably enhanced by the expulsion of gas from deeper layers. Gas seepage may become apparent without the presence of geomorphological features on the seabed. Gas bubbles may be issuing from the seafloor creating bubble plumes (e.g., Gullapalli et al., 2019) in 41

the water column (macro-seepage) that may break into the surface when in shallow depth (e.g., Etiope et al., 2013a) or columns of colored or shimmering water, minimal discharge of microscopic bubbles or hydrocarbons in solution detectable only by geochemical instruments (micro-seepage; Hovland and Judd, 1988). Strong and continuous, gas bubbling from the seabed is indicative of thermogenic origin of the gas, since it implies strong pressure gradients and source potential (Etiope, 2015; Hornafius et al., 1999; Etiope et al., 2013a).

Methane in marine sediments is oxidised and converted to CO2 (anaerobic oxidation of methane, AOM) with the mediation of bacterial consortia constituted by archaeal methanotrophs and sulphate-reducing bacteria (e.g., Boetius et al., 2000). When weak, methane seepage is rapidly, and due to low flux sometimes completely, consumed by AOM and hence the seabed is biological sink for methane (Etiope et al., 2015). As methane travels through marine sediments it is oxidized by anaerobic microbial activity and in conjunction with sulfate reduction (SR), hydrogen sulfide is produced:

−2 − − 퐶퐻4 + 푆푂4 → 퐻퐶푂3 + 퐻푆 + 퐻2푂

Sulphide, which is one of the products of AOM and SR, accumulates to corresponding concentrations of the consumed sulphate (Boetius et al., 2000), thus explaining the occurrence of sulphide-based Beggiatoa spp. white bacterial mats (e.g., Etiope et al., 2013a and Chapter 4.1.1; Girard et al., 2020; Macelloni et al., 2013; Lloyd et al., 2010;) around the seeps. The bicarbonate produced by methane- oxidation leads to submarine cementation and formation of authigenic carbonate crusts/slabs (Judd et al., 2020; Macelloni et al., 2013) and mounds (e.g., Etiope et al; 2013 and Chapter 4.1.1; Lutken et al., 2011). The abovementioned process that takes place in cold seep environments creates reduced conditions in the marine sediments are intimately associated with complex chemosynthetic communities (Levin 42

et al., 2016; Boetius et al., 2000; Sibuet and Olu, 1998). Microbial processes contribute to the development and stability of chemosynthetic communities by providing required H2S. The reduced gases that are released from cold seeps, although toxic to most of the higher organisms (Bagarinao, 1992; Beauchamp, et al., 1984; Truong et al., 2006; Wang and Chapman, 1999) are favorable for chemosynthesis-based organisms, which have high ecological importance, as they are at the base of the food chain (Levin et al., 2016; Sibuet and Olu, 1998; Åström, 2017), and below the euphotic zone, seeps produce an oasis effect, as they stand out from the surrounding bare seabed (Fig., 2.3.1; Levin et al., 2016; Åström, 2017). Thus, the significance of seabed fluid seepage lies beyond geological sciences, since they are highly important to marine biology and chemistry and, they represent globally an important greenhouse gas source for the atmosphere (e.g., Etiope, 2004; Etiope and Milkov, 2004). Methane, the major component of cold seeps, is a high potent greenhouse gas, when it is expelled from the hydrosphere into the atmosphere it may contribute significantly to the atmospheric carbon budget (Judd et al., 2002, Etiope, 2015).

Fig. 2.3.1 Image of methane bubble streams issuing from the seafloor from a cold seep site offshore Virginia north of Washington Canyon. Quill worms, anemones, patches of 43

bacterial mat, pandalid shrimp, and a large red crab, are visible within the seepage area. (Image ID: expn0686, Voyage to Inner Space - Exploring the Seas with NOAA Collect, Location: Virginia, north of Washington Canyon Photo Date: 20130531T140638Z Credit: NOAA OKEANOS Explorer Program, 2013 ROV, License: creativecommons.org/licenses/by/2.0).

Methane flux from individual submarine seeps has been estimated from several authors (Judd et al., 1997; Hornafius et al., 1999; Dimitrov, 2002; García-Gil, 2002; Clark et al., 2010; Brunskil et al., 2011), whilst the estimations of global methane fluxes from the seabed to the atmosphere vary between 5 to 12 Tg yr−1 (Etiope et al., 2019). However, most studies have assessed methane flux at the seabed and not the flux that actually reaches the atmosphere (Judd, 2004). Some methane from shallow (>100-300 m), hence mainly coastal and continental shelf, water seeps escapes to the atmosphere, whilst methane from deep-water seeps may be lost to the hydrosphere (Judd, 2004; Leifer and Patro 2002). The fraction of methane in bubbles issuing from the seabed that finally enters the atmosphere is predominately dependent on the initial water depth, bubble size, temperature, salinity, and methane concentration of the water (Leifer and Patro, 2002). Methane ascending in bubbles from the seabed may never (or not completely) reach the sea surface, due to gas exchange between water and bubbles (see Chapter 4.3) and different types of oxidation (see Chapter 2.5) depending on methane flux and water depth. Kvenvolden et al., 2001, estimated that 40 % of submarine seeps reach the atmosphere. The same work suggested ~20 Tg y-1, which was the average of two comparable values estimated by two different approaches, as a realistic value of methane flux from submarine seeps reaching the atmosphere. This value was extensively used, until Etiope et al. 2019 challenged this estimation, since the two separate estimates of 10 and 30 Tg y-1 provided by Kvenvolden et al., 2001 had no indication of their uncertainties, and the values were purely based on process-based modelling. Etiope et al. 2019 suggested the more 44

realistic estimation of submarine methane emission range of 5-12 Tg y- 1, that is previously mentioned. The presence of seepage induced positive- or negative- relief morphological features over gassy marine sediments, depends on: (i) location and geometry of underlying reservoirs, (ii) the suitability of sediments, e.g., clayey silts or silty clays which are favorable for pockmarks (Papatheodorou et al., 1993), (iii) the presence of mobilized shales in the subsurface for mud volcanoes (Milkov, 2000), (iv) the fluid seep activities and dynamics, and (v) unnegotiably, on the existence of a fault, which acts as a gas (pockmarks) or fluidized methane-rich sediment (mud volcanoes) pathway (Hovland et al., 2002; Roy et al., 2016; Ligtenberg, 2005; Gay et al., 2007; Judd and Hovland, 2007; Chand et al., 2012; Ostanin et al., 2013). Other seafloor gas seepage manifestations with absence of related morphological features (Etiope, 2015) include sediment holes or fractures, bacterial (Beggiatoa spp.) mats (e.g., Etiope et al., 2013a; Naudts et al., 2008) and gas charged sediments, which can be detected by marine geophysical surveys with the application of subbottom profilers (e.g., Papatheodorou et al., 2003; Naudts et al., 2008; Isola et al., 2020). The prolific Coal Oil Point Seep, which results from the seaward extension of the oil-rich Ventura Basin in Southern California, is probably one of the first offshore oil and gas seeps that the early Spanish explorers observed (Hunt, 1996; Hornafius et al., 1999). The first ever reported written records, date back to 1792, when Captain Cook’s navigator George Vancouver wrote about the offshore oil and gas seeps in the Santa Barbara Channel: The surface of the sea, which was perfectly smooth and tranquil, was covered with a thick, slimy substance, which when separated or disturbed by a little agitation, became very luminous, whilst the light breeze, which came principally from the shore, brought with it a strong smell of tar, or some such resinous substance” (Fig. 2.3.2; Imray, 1868; Yerkes et al., 1969). 45

Fig.2.3.2 James Imray book “Sailing directions for the West Coast” that was published in 1868, reported G. Vancouver’s description of the effects of the Coil Oil Point Seep.

There is no doubt that among the seafloor fluid seepage induced features, pockmarks, even today and despite their meticulous worldwide study, still bear some mystery. King and McLean in the most recent 1970, were the first scientists to publish a scientific paper regarding the discovery of the first pockmark. The pockmark that was identified in side scan sonar sonographs collected during surveys in 1963-64 from Emerald Basin of the Scotian Shelf in Nova Scotia, Canada, was described as a crater-like feature (Hovland and Judd, 1988). Pockmarks were initially approachedsimply as a geophysical phenomenon. Later, it was confirmed that these cone-shaped depressions were formed by vigorous fluid flow (Hovland and Judd, 1988; Hovland et al., 1987), which constitutes the forcing mechanism for pockmark formation through winnowing and erosion (Hovland and Judd, 1988). Thus, pockmarks started to be handled as indicators of past or present, continuous or episodic fluid seepage activity (Hovland et al., 2002). The discovery of the first pockmark in 1970 could never be possible without 46

the newfound (then) technology of the side scan sonar in the mid-60s. The side scan sonar was introduced as a seabed survey tool and combined with the development of high-resolution seismic profilers and towed cameras and ROVs, made possible the integrated, high resolution seabed and sub-seabed mapping. Until today the side scan sonar is used extensively for pockmark (as well as all geomorphological expressions) mapping and imaging (Fig. 2.3.3). Since then, pockmarks have been discovered in a variety of continental margin settings all over the world (Hovland and Judd, 1988; Judd and Hovland, 2007). Pockmark fields may occur near deltas (Nelson et al.,1979) and paleo-deltas (Naudts et al., 2008), at petroleum-generating regions (Hovland et al., 1987; Uchupi et al., 1996), in mid-latitude estuaries (Garcia-Garcia et al., 1999; Ussler et al., 2003), in formerly glaciated regions (Whiticar, 2002), areas of groundwater flows (Bussmann and Suess, 1994; Christodoulou et al., 2003), lakes with or without hydrothermal activity (Pickrill, 1993), and lagoons (Papatheodorou et al., 2001). Shelf basins (Fader, 1991), continental slopes and rises (Orange et al., 1999; Paull et al., 2002) also host fields of pockmarks.

water column water column

seepage 50 m pockmark 50 m pockmark seepage Fig. 2.3.3 Uncorrected side scan sonar echograms of active gas seepage from a pockmark in the Patras Gulf pockmark field (Christodoulou et al., 2009).

Whilst pockmarks can be induced by both contemporary (late Pleistocene and Holocene) gas, another iconic morphological feature of methane seepage: the mud volcano, is a result of fossil gas migration (e.g., Loncke et al., 2004; Davies and Stewart, 2005; Deville et al., 2006). Mud volcanoes are conical, positive relief features that occur 47

both on and offshore over faults, in petroleum bearing sedimentary basins (Etiope, 2015). Their formation mechanisms which are thoroughly described by Milkov, 2000, are associated with a seafloor- piercing shale diapirs and upward migration of fluidized sediments along active faults. Mud volcanoes have been reported from many submarine sites around the world including the Gulf of Mexico continental shelf and slope (Neurauter and Roberts, 1994; Sassen et al., 1993; Kohl and Roberts, 1994), offshore Barbados Island (Zhao et al., 1986; Le Pichon et al., 1990), Indonesia (Breen et al., 1986), the Caspian Sea (Hedberg, 1974), the Beaufort Sea (Pelletier, 1980), Norwegian Sea (Vogt et al., 1997), the Arabian Gulf (Hovland and Judd, 1988), the northeast shelf of New Zealand (Nelson and Healy, 1984), offshore Greece (Cita et al., 1981; Cronin et al., 1997; this work).

2.4 Seepage Geochemistry Applications

2.4.1 Seepage as a tool for hydrocarbon exploration Analyses of the molecular and isotopic composition of seeping gas is a key task for understanding subsurface hydrocarbon potential, genesis, and quality (e.g., by discriminating shallow microbial gas from deeper thermogenic accumulations, and suggesting the presence of oil and undesirable non-hydrocarbon gases such as CO2, N2, and H2S). Gas seeps can also indicate subsurface petroleum biodegradation, which has an important impact on hydrocarbon quality and may influence exploration and production strategies. Thus, gas geochemistry can contribute to assessing prior to or without drilling, a petroleum system, which is particularly useful in frontier or unexplored areas (Etiope, 2015). 48

2.4.2 Environmental impact Seeps may become a geohazard for humans, buildings, and industry: explosions and sudden flames may occur in gas-rich environments (boreholes, soil), if methane concentrations reach explosive levels of 5- 10 % in the presence of air. When methane is accompanied by hydrogen sulfide (H2S, e.g., in salt diapirism zones), seeps can be toxic or even lethal under some circumstances (Etiope et al., 2006; Etiope et al., 2013). Seeps plumbing can then damage buildings and infrastructures by gas-pressure build-up in the subsoil or by general degradation of geotechnical properties of soil foundations. The seeps of

Katakolo, studied in this thesis, have considerable amounts of H2S and actually represent hazard for local activities. Hydrocarbon seepage represents, then, a significant global natural source of greenhouse gas (Etiope et al., 2008; Etiope, 2015). Onshore and offshore seeps (together with diffuse microseepage) are estimated to be the second most important natural source of atmospheric methane (more in Subchapter 2.4.4), after wetlands, on global scale. The offshore seeps are important for the atmosphere only if they occur in shallow seafloor, generally not deeper than 200-300 m (Schmale et al., 2005; Mc Ginnis et al., 2006) otherwise most of methane is dissolved and oxidized along the water column. The seeps studied in this thesis are in onshore and in coastal, very shallow seafloor zones.

2.4.3 Hypoxia and its links to fluid seepage A specific environmental impact of seepage, studied in this thesis, is the hypoxia in seawater, i.e., the low concentration of oxygen in solution in water (dissolved oxygen). Although dissolved oxygen in the Earth’s hydrosphere is crucial not only for marine life but also for terrestrial life, since the oceans produce more than the 80 % of oxygen of the atmosphere through the photosynthesis of the phytoplankton, oxygen concentrations in the open ocean and in coastal areas are decreasing 49

worldwide (Diaz, 2001; Diaz and Rosenberg, 2008; Rabalais and Gilbert, 2009). Oxygen depletion and “hypoxia” develop wherever the consumption of oxygen by organisms or chemical processes exceeds the provision that takes place through water exchange/circulation, the atmosphere, and from photosynthesis of the phytoplankton (Friedrich et al., 2014). By the term hypoxia in water bodies, we refer to the depletion of dissolved oxygen in the water, at concentrations below 60 μM (or <2 μg/L) whereas the normal concentration of atmospheric air saturated water is <200 μM (or <7 μg/L). Hypoxia or oxygen depletion in water bodies may be caused by both natural and anthropogenic drivers. Particular hydrodynamic conditions; i.e., enclosed and semi-enclosed water bodies with limited water exchange, fjords, lagoons, and lakes, etc., may cause a natural susceptibility to hypoxia. Moreover, the decline of oxygen often results from the oxidation of organic matter and reduced substances due to stratification and/or the intrusion of oxygen-poor waters from deeper layers or adjacent water bodies. In some cases, hypoxia or anoxia may be geologically driven by submarine gas seepage. Reducing gases such as CH4 and H2S can induce hypoxia since they can be rapidly oxidized by microbial communities resulting in oxygen consumption and thus depletion the seawater (see Subchapters 5.1.3 and 5.2.1). Hydrogen sulfide is produced by seafloor microbial sulfate reduction of methane (see Subchapter 2.3). Moreover, as gas bubbles ascend in the water column, gas exchange between the bubbles (CH4) and seawater (i.e.,

O2, N2) takes place, resulting in dissolved oxygen depletion. Although geologically driven hypoxia is less common, and hence has not been extensively studied to date, this phenomenon has been reported in African lakes, the Cariaco Basin (Venezuela), from offshore Namibia, in the Gulf of Mexico (Nayar, 2009; Ward et al., 1987; Kessler et al., 2005, 2011; Monteiro et al., 2006) and in petroliferous areas that were studied in this thesis, in Western Greece (Friedrich et al., 2014; Ferentinos et al., 2010). 50

Hypoxia may have detrimental effects upon all components of marine ecosystem functioning, such as biodiversity, and ecosystem services such as fisheries and tourism (Pearson and Rosenberg, 1978; Nilsson and Rosenberg, 2000).

2.4.4 The greenhouse effect and its links to methane seepage Methane is a potent greenhouse gas that is emitted to the atmosphere through both natural and anthropogenic sources (e.g., Etiope, 2015; Saunois et al., 2016). Twenty per cent of the radiative forcing from all long-lived greenhouse gases (i.e., CO2, CH4, N2O, and fluorinated gases) is attributed to methane, while its concentration to the atmosphere has increased ~2.5 times since the pre-industrial era (1750); placing it as the second more powerful and prevalent human induced greenhouse gas after carbon dioxide (Etiope, et al., 2019; Saunois et al., 2016). Methane builds up in the atmosphere with the contribution of multiple sources, both natural and anthropogenic, making it difficult to assess the global atmospheric methane budget and to determine the contribution of each source. Almost a decade of stable CH4 concentrations (late 1990s to 2007) was followed by a period of increasing CH4 concentrations that initiated in 2007, whose drivers are debated (Ciais et al., 2013). Regarding geologically originating methane, although at first its contribution to the global atmospheric methane budget was considered insignificant, it was later proved through a series of published scientific researches (Lacroix, 1993; Klusman et al., 1998; Etiope and Klusman, 2002; Judd et al., 2002; Kvenvolden and Rogers, 2005; Etiope et al., 2008; Etiope, 2009, 2012) that it represents the second more dominant natural source of methane to the atmosphere after wetlands (Etiope, 2012; Etiope, 2009; Etiope

–1 and Ciccioli, 2009). With a global emission of 60 Tg (CH4) yr , as indicated by bottom‐up emission estimates, methane corresponds to 51

~10% of the total methane sources (Etiope, 2012). By including natural geological CH4 emissions, the fossil component of the total CH4 emissions (i.e., anthropogenic emissions related to leaks in the fossil fuel industry and natural geological leaks) is now estimated to account for ~30 % of the total CH4 emissions, as opposed to 50-65 % that it was considered in initial assessments (Ciais et al., 2013). There are still questions that need to be answered regarding the contribution of sources from the fossil fuel industry and from natural geological seepage to the total methane emissions (Schwietzke et al., 2016), some could probably be answered by a closer investigation to the numerous seeps which account for ~12,500 but have been partly closely investigated (Etiope, 2015). The recent years Earth is facing the challenge of climate change, which poses conspicuous environmental risks. The assessment of the contribution of natural and anthropogenic sources to the global CH4 budget is of crucial importance for the development of effective management plans for climate change mitigation (Saunois et al., 2016). The recognition of geologically originating methane, as an important component of the atmospheric methane budget as well as the inclusion to the official methane emission inventories (Ciais et al., 2013) was achieved, among others, through the acquisition of a large number of methane flux data for various seepage types in numerous sites and countries (Etiope, 2015).

2.5 Review of seabed fluid flow manifestations offshore (and onshore) Western Greece Over more than 4 decades multiple marine remote sensing surveys have been carried out in Aegean and Ionian Archipelagos by the Laboratory of Marine Geology and Physical Oceanography (Department of Geology, University of Patras), the National Centre of Marine Research (HCMR), the Institute of Geology and Mineral Research 52

(IGME) and collaborators for a variety of objectives (e.g., Stanley and Perissoratis, 1977; Ferentinos et al., 1988; Papatheodorou et al., 1993; Sakellariou et al., 2007; Ferentinos et al., 2020). Plenty of these surveys have revealed acoustic anomalies indicative of shallow gas accumulations, gas seepage, and hydrothermal fluids, suggesting that these common phenomena in the Greek seas (Fig. 2.5.1). Gas accumulations in Greece have been reported in Pleistocene and Holocene sediments in fiords-like environments, deltaic environments, lakes, lagoons, and open sea environments (Lykousis, 1989; Chronis et al., 1991, Papatheodorou et al., 1993) and in pre-Quaternary coastal and open sea environments related to faults and salt doming indicating a deep, thermogenic origin (Papatheodorou et al., 1993) (Fig. 2.5.1). The back-arc volcanism of the Hellenic arc has been triggered the hydrothermal seabed fluids in the Aegean Sea (Dando et al., 1995a,b) (Fig. 2.5.1).

53

Fig. 2.5.1 General map of Greece presenting the areas where gas seepage manifestations and hydrothermal vents have been obtained.

The first pockmark discovery in the Greek seas was offshore Thasos Island, N. Aegean Sea by Newton et al.,1980 (Fig. 2.5.1). Hasiotis et al. (1996) discovered and mapped the largest so far Greek pockmark field in Patras Gulf (Fig. 2.5.1). Papatheodorou et al. (1993) documented the presence of ‘cold’ seeps and gas-charged sediments in lakes, deltaic, open sea, and fjord-like environments based on the acoustic characters and morphological expressions that were detected in high resolution seismic profiles and side scan sonar echograms (Fig. 2.5.1). A high reflectivity horizon has been observed at the Pleistocene/Holocene boundary in Amvrakikos Gulf, a west- northwestern trending elongated graben, that was attributed to gas accumulation (Fig. 2.5.2a, b; Papatheodorou et al., 1993). In the central basin of Amvrakikos Gulf buried pockmarks were observed on the gas accumulative horizon (Fig. 2.5.2a; Papatheodorou et al., 1993). Gas plumes, gas pockets, and low relief intra-sedimentary doming in the Holocene sediments accompanied with seabed doming and gas plumes in the water column may indicate a continuous fluid seepage (Fig. 2.5.2a, b; Papatheodorou et al., 1993).

54

Fig. 2.5.2 (a) & (b) 3.5 kHz profiles of the Gulf of Amvrakikos showing the Pleistocene/Holocene boundary (HOL/PL); intra-sedimentary and seabed doming (D); enhanced reflectors (ER); gas pockets (GP) and gas plumes (GPI); buried pockmarks (BPM) developed at the Pleistocene/Holocene interface and Fault (F) (from Papatheodorou et al., 1993); (c) 3.5 kHz profile of Lake Trichonis showing gas charged sediments (ATZ=Acoustic Turbid Zone; GPI= Gas Plume) (from Papatheodorou et al., 1993); (d) side scan sonar of Aetoliko Lagoon showing pockmarks some of which occur in strings (from Papatheodorou et al., 2001).

Fluid-charged sediments have been recorded in Lake of Trichonis as indicated by an acoustic turbid zone throughout the lake occurring ~1-2 m below the lake floor (Fig. 2.5.2c). A large number of small sized pockmarks have been detected in the southeastern part of the deep (~30 m) lagoon of Aetoliko located at the northern end of the Mesolonghi – Aetoliko complex (Papatheodorou et al., 2001) (Fig. 2.5.2d). It was ambiguous whether the fluid induced expressions were associated with the catastrophic events of anoxia that have occurred from time to time in the lagoon. Gassy sediments have also been recorded in many present-day deltaic environments such as the Mornos, Evinos, and Piros deltas (Lykousis, 1989; Papatheodorou et al., 1993; Fig. 2.5.1). Although, escaping gas has not been geochemically analyzed (outside Amvrakikos, Patraikos Gulfs and Aetoliko) in many of the abovementioned seabed fluid flows, the seismic data imply that the gas is of microbial origin. Papatheodorou et al., (1993) suggested a thermogenic origin of gas seepage from on the Corfu shelf, located on the NW end of the Hellenic Arc (Fig. 2.5.1). The suggestion was based on the seismic data, which 55

showed that gas was seeping from a Triassic salt diapir which pierces the seabed. Since the early 90’s in Greece, the seismic surveys were the only mean to speculate the origin of the gas from seabed fluid flows. In more recent researches, more sophisticated techniques were used in combination with marine acoustic methods in order to investigate submarine gas manifestations. Etiope et al. (2005; 2006) reported that the offshore gas seepage at Katakolo Harbor, Western Greece, is thermogenic methane accumulated in Mesozoic limestones and migrated through faults (Fig. 2.5.3a). Hydrogen sulfide is also present, since it is produced by thermal sulfate reduction and/or decomposition of sulfur compounds in kerogen or oil.

(a)

(b)

Fig. 2.5.3 (a) Underwater photo showing bacterial mat (Beggiatoa sp.) and bubbles rising from the seafloor offshore Katakolo; (b) Side scan sonar mosaic in Elaiona Bay showing pockmarks on the seafloor (from Christodoulou et al., 2003). 56

Hydrothremal seabed fluid flows are common around Greek islands located at the Hellenic volcanic arc (Fig. 2.5.1). Dando et al. (1995a) surveyed marine areas with fluid flows around Milos Island, which is one of several places of significant activity. Many individual spots showing fluid flows were measured, showing that CO2 is the dominant gas with hydrogen sulphide and methane as important minor components. Dando et al. (1995b) presented a documented event of fluid flow triggered by seismic activity. Repeated studies using echo sounding showed that two earthquakes of 5.4 M and 5.0 M (20.4.1992) southern of Milos Isl. had triggered many separate fluid flows on the seabed at ~10mwater depth. One hour after the first earthquake, 99 gas flares were recorded, declining rapidly in the following hours. The number of gas plumes finally reduced to 60, and this number was the same when the marine area was surveyed again 3 months later. Fresh submarine groundwater discharge is known since manycenturies. Kohout (1966) stated that the Roman geographer Strabo (63 BC - 21 AD) mentioned a submarine spring offshore Syria (Mediterranean), from which fresh waterwas collected with a boat and transported to the city. Christodoulou et al., (2003) performed a multi- discipline surveyover a pockmark field in Elaionas Bay in the Gulf of Corinth (Fig. 2.5.3b and d). Low methane concentrations (47–51 nM) and freshwater discharge were detected in and over the pockmarks. Based on the results, the main mechanism for the formation of the Elaionas’ pockmarks was considered the freshwater discharge from the seabed. The Patraikos Gulf Pockmark Field One of the largest pockmark fields discovered so far in Greece, is the Patraikos Gulfpockmark field, located near Patras Port between ~10 and 45 m water depth. Some of the pockmarks (250 m in diameter and 20 m in depth) appear just 500 m away from the shoreline. The Patraikos Gulf pockmark field has a long record of marine geophysical and geochemical research. There is strong evidence that the pockmark field was activated during a 5.4 M earthquake (July 14th, 1993). The 57

Patraikos pockmark field is mentioned in the book of Judd and Hovland “Seafloor fluid flow” published in 2007 as: “some of the most spectacular and best-documented events in active pockmarks” (pp. 230-231). The Gulf of Patras is a semiclosed gulf open into Ionian Sea on the west and is linked via the Rio-Antirio narrows to the Gulf of Corinth on the east (Fig. 2.5.4). Based on offshore shallow (3.5 kHz) and deep (Air-gun) seismic data Ferentinos et al. (1985) thePatraikos Gulf bathymetry is formed by active WNW-ESE trending faults (Fig. 2.5.4). Normal faults cause a seafloor displacement from ~5 to ~25 m.

Fig. 2.5.4 Bathymetric contours (in meters) and offshore and onshore faults in the Gulf of Patras (modified from Ferentinos et al., 1985; Doutsos and Poulimenos, 1992; Flotte et al., 2005), (AT.F= Agia Triada fault).

Ferentinos et al. (1985) suggested that the Patras trough is an asymmetric graben with a north dipping major listric fault. The Patras graben seems to terminate against of ENE-WSW normal faulting of the Pleistocene sequences outcropping onshore in the SE part of the Gulf (Fig. 2.5.4). The Patraikos Gulf pockmark field, first observed by 58

Hasiotis et al. (1996), lays at the intersection of the WNW-ESE and ENE-WSW fault zones (Fig. 2.5.4). High resolution seismic profiles showed two layered sequences throughout the Gulf of Patras (Fig. 2.5.5). The upper almost transparent sequence distinguished into three units (C, B and A). The lower sequence contains strong, thin, parallel reflectors (Fig. 2.5.5). The high reflective surface of the lower sequence passes laterally towards the margin of Patraikos Gulf into an unconformity. Chronis et al. (1991) referred to this surface, as the basal unconformity of the Gulf which is associated to Last Glacial Maximum (18 ka) while Ferentinos et al. (1985) proposed that the unconformity corresponds to the Pleistocene/Holocene boundary. Chronis et al. (1991) suggested that during the isotopic stage 2 the entire Gulf of Patras was sub-aerially exposed, and the central graben muddy flood plain sediments accumulated, with an axial river channel leading to the Ionian Sea. The age at the top of unit C was estimated at ~14 ka and was interpreted as a marine sedimentation phase during the earlier stage of marine transgression. Units B and A represent the complex deltaic progradation of Acheloos and Evinos rivers.

Fig. 2.5.5 3.5 kHz profile across the central graben of the Gulf of Patras showing the basal unconformity and the three units A, B and C in the upper transparent sediment sequence (figure from Chronis et al., 1991).

The Patras Gulf pockmark field has been extensively surveyed by the Laboratory of Marine Geology and Physical Oceanography with the collaboration of INGV and HCMRsince more than 20 years. During 1997 thirteen 3m-long cores were collected inside and outside of the 59

pockmarksfor grain size, organic carbon content, calcium, and heavy metals analysis. The third survey was conducted in five periods: May, September, October and November 2002 and in July 2003, in the framework of the ASSEM Project (Array of Sensors for long-term Seabed Monitoring of geohazards) funded by the EU (EVK-CT2001-00051). The main objective of these surveys was the selection of the optimumsite in the Patras pockmark field for the deployment of the Gas Monitoring Module (GMM; Marinaro et al, 2006; see Subchapter 5.1.3). For the purposes of that survey, geophysical (3.5 kHz subbottom profiler, EG&G side scan sonar) and oceanographic (methane underwater sensor, temperature/salinity/DO probe) data were collected coupled with visual inspection (ROV and divers). After that, the surveys were continued in three periods: 2005, 2006 and 2007, in the framework of PYTHAGORAS II project (National funds). The aim of that survey was a detailed hydro-acoustic and seismic study of the northern sector of the already discovered pockmark field. For these surveys a 3.5 kHz Geopulse subbotttom profiler, an Odom Echotrack echo-sounder, and an EG&G 270 side scan sonar were used. A Differential GPS model Trimble 4000 II with RMS accuracy of 1-2m was used for data positioning. Side scan sonar lanes covered completely the pockmark field. The echo-sounding lines had an interval spacing of 5m. Approximately 70x102 of echo-sounding data were acquired and led to the construction of a detailed bathymetric map. The pockmark field consists of 72 pockmarks and extends over an area of about 2.5 km2 (Figs. 2.5.6, 7). The pockmarks are divided into three morphological classes: (i) unit pockmarks measuring from 20 to 40 m in diameter and up to 5m water depth, (ii) normal pockmarks which are mostly circular or oval in top view, ranging from 50 to 150m in diameter and up to 15m in depth, (iii) complex or composite pockmarks which are an amalgamation of normal pockmarks (Figs. 2.5.6, 7). 60

Pockmark distribution demonstrates a spatial pattern that separates the pockmark field in two sectors (Figs. 2.5.6, 7). The northern sector contains 57 pockmarks within an area of 1 km2 while the southern sector shows lower density: 20 pockmarks per km2. The northern sector of the field contains all the classes of pockmarks (Figs. 2.5.6, 7), whilst the southern sector mainly consists of unit and normal pockmarks without any specific trends in their shapes and sizes (Figs. 2.5.6,7). One composite pockmark (named P4) was detected at the southern end of this sector. It has a diameter of about 250 m, a maximum depth of 40 m and is about 25 m deeper than the surrounding area. On the northern sector two pockmark strings were detected appearing to follow linear trends of WSW-ENE and WNW-ESE orientations (Figs. 2.5.6, 7). Mitsis et al. (2002) studied the morphometric parameters (total area, average slope, relative depth, eccentricity, and direction of long axis) of the pockmarks of Patras using multivariate statistics. These authors showed that the long axes of the pockmarks are oriented in a direction ranging from 110o to 140o (ESE-WNW to SE-NW).

Fig. 2.5.6 Single-beam bathymetric map showing the pockmarks of Patras Gulf. Pockmark P4 was selected for long-term monitoring (Papatheodorou et al., 2007). 61

The long axe is almost perpendicular to the depth contours and to the shoreline. The elongated shape of a pockmark results from a continuous gas release accompanied by strong bottom currents transporting sediments and preventing their deposition in the central pockmark area. In these cases, the long axes of the pockmarks represent the effective bottom current direction. However, this is not the case of the Patras Gulf pockmarks, where the weak bottom currents have NNE-SSW direction, almost parallel to the depth contours. This fact together with the linear alignment of the two pockmark groups in the northern sector of the field probably suggest that these characteristics of the pockmarks are related to the escape of fluids through faults from the soft sediments of the Gulf (Hovland, 1983; Acosta et al., 2001).

Fig. 2.5.7 Side scan sonarmosaic of the Patras pockmark field (Papatheodorou et al., 2007). 62

Interpretation of high-resolution seismic profiles from the pockmark field have shown the existence of an acoustically transparent sequence with few, weak internal reflections overlying a highly reflective surface that usually blocks further seismic penetration (Fig. 2.5.8). This highly reflective surface represents the Pleistocene/Holocene boundary (Chronis et al., 1991; Papatheodorou et al., 1993). The strong acoustic character of this boundary can be attributed to sharp contrast in sediment texture (size, shape, and arrangement of grains) or gas charged sediments (Fig. 2.5.8a). Five shallow borings which have been drilled within the pockmark field showed that the highly reflective Pleistocene/Holocene boundary is a result of a change in the geotechnical properties of the sediments. Hasiotis et al. (1996) suggested that the Pleistocene/Holocene boundary can be considered as a gas accumulative horizon, based on the sharp geotechnical interface in combination with the presence of seismic indicators of shallow gas.

Fig. 2.5.8 3.5 kHz profiles of the Patras pockmark field showing: (a) the Pleistocene/Holocene boundary (HOL/PL), (b) acoustic turbid zones (ATZ) and enhanced reflectors (ER) within the acoustically transparent Holocene sequence and (c) intra- sedimentary and seabed doming (D). (F= fault, GP= gas pocket) (Papatheodorou et al., 2007).

The high-resolution seismic surveys within the pockmark field, revealed acoustic anomalies (i.e., acoustic turbid zones, enhanced 63

reflectors, doming, etc.) in the uppermost 25 m of the seabed sediments which are related to the presence of gas. The very high amplitude reflectors are interpreted to be related to the presence of gas (Fig. 2.5.8b). The enchanced reflectors have been detected within the Holocene sequence, also suggesting the upward migration of gas and its entrapment in porous horizons within the transparent sequence (Fig. 2.5.8b). In 3.5 kHz seismic data, pockmarks showed some features which are diagnostic for the mechanism of their formation and activation. Seismic profiles across the pockmark field show the normal pockmarks as U or V-shaped and/or dish shaped incisions truncating the weak internal reflectors of the upper acoustically transparent sequence (Fig. 2.5.9). Some of them have flat floors and are like truncated cones (Fig. 2.5.9). In many cases the pockmarks are recorded with side echoes that overlap to form “bow-tie” structures which indicate that the side walls of the pockmarks are too steep to insonified by single beam techniques.

Fig. 2.5.9 3.5 kHz profiles showing: (a) acoustic turbid zone (ATZ) underneath the floor of a pockmark, (b), (c) chimney like zone of transparent acoustic character (TR) below the floor and the side walls of the pockmarks, and (d) detached block (DB) resting on the pockmark floor (from Hasiotis et al., 1996) (HOL/PL=Pleistocene/Holocene boundary, ER=enhanced reflectors, F=fault) (Papatheodorou et al., 2007). 64

The floor of the largest pockmarks usually reaches the Pleistocene/Holocene interface. No pockmark penetrates through the Pleistocene/Holocene boundary (Fig. 2.5.9). Many of the pockmarks are characterized by the existence of an acoustically turbid zone which is located under the floor of the pockmarks (Fig. 2.5.9a). This suggests that: (i) there is a continuous supply of gas from below towards the floor of the pockmarks and (ii) the sediments below the floor and within the gas migration path are gas- charged. A number of pockmarks are characterized by a chimney like zone of transparent acoustic character (Fig. 2.5.9b, c). This zone is not restricted below the floor of the pockmarks but also extends below the side walls. The presence of this reflection-free chimney like zone suggests that: (i) the gas venting took place all over the surface of the pockmark and was not limited to the floor, (ii) the fluid flow was strong enough to disturb the texture of the sediments and (iii) the sediments within the migration path below the floor are currently gas free due to the loss of gas through recent out-gassing. The side walls of some complex and normal pockmarks are affected by gravitative mass movements (Fig. 2.5.9d). Rotational slides affect the sidewall of some pockmarks and detached blocks seem to cap the floor of some others. The presence of sediment movements covering the floor of the pockmarks suggests that gas venting can be temporally blocked, which may either lead to a new migration path or to a more abrupt gas emission (Hasiotis et al., 1996). In 3.5 kHz records no pockmarks with rims have been detected, meaning that there were no episodic gas eruptions (Hovland and Judd, 1988). At the northern sector of the field, the northernmost pockmark string is aligned along the base of a terrace slope which drops from ~4 to 10 m (Figs. 2.5.6, 7). This 6-7 craters pockmark string is almost straight and is oriented at WSW-ENE for about 1km (Figs. 2.5.6, 7). Seismic profiles across this slope display the trace of a WSW-ENE 65

trending normal fault which dips to the south with a 5m throw (Figs. 2.5.10a, b). This fault seems to be the northern limit of the pockmark field. This WSW-ENE trending fault is the offshore continuation of Ag. Triada fault (Figs. 2.5.4, 6). Some microseepage flux measurements on land in correspondence with this fault (Etiope, Papatheodorou, Christodolou, unpublished data) showed the existence of weak methane seepage from the ground, suggesting that the fault acts as gas migration pathway.

Fig. 2.5.10 (a), (b) 3.5 kHz profiles showing a small depression and a pockmark (Pc) formed along the Agia Triada fault (AT.F). (HOL/PL=Pleistocene/Holocene boundary) (Papatheodorou et al., 2007).

The 1993 M=5.4R Patras earthquake: Acoustic evidence of pockmark activation One of the most well documented pockmark activation events was recorded within the Patras’ pockmark field. At 15:32 on July 14th, 1993 an earthquake of 5.4 M was recorded in Patras Gulf; one of the largest earthquakes of the 20th century in that area. By chance, a current-meter and a temperature/salinity probe had been placed in the coastal zone of Patras city, 3 m above the seabed. During the 24 hour period prior to the earthquake the bottom temperature anomalously increased trice: (i) from 16.8 C to 19.3 C for 1h 20 min, (ii) from 16.8 C to 23 C for 5h 25 min, and (iii) from 17 C to 22 C for 5 h. A marine remote sensing survey was carried outin the area after the main shock of the earthquake using a 3.5 kHz subbottom profiler and a side scan sonar. The pockmark field of the Patras Gulf was 66

detected in the hydro-acoustic data along with water-column targets which were interpreted as gas flares representing bubble plumes and sediment, rising from the pockmarks. For several days after the earthquake several pockmarks was active during the remote sensing survey as indicated by the sharp-edged “dark cloud” (high reflective patches) rising from the pockmarks (Fig. 2.5.11). The most well- developed gas bubble plumes were recorded in the north along the WSW-ENE trending pockmark string which is related to the Ag. Triada fault (Figs. 2.5.4, 6, 11). The pockmarks activation and the three sharp temperature increases in the water column before the earthquake were probably the result of gas bubbles rising from the pockmarks. The detection of gas- bubbles plumes over several pockmarks, suggests that the 3 temperature rises were not a local phenomenon, but were an event of wider occurrence, within the northern sector of the pockmark field. The earthquake was considered as the triggering mechanism for activation of the pockmarks. It is further suggested that the gas discharge was also caused by the reduction in the pore volume of the sediments, resulting from changes in the stress regime prior to the earthquake. The stronger gas releases were detected from the pockmark strings related to the WSW-ENE trending fault system, indicating that the fault is the preferencial gas pathway. 67

Fig. 2.5.11 Side scan sonar mosaic from the northern part of the Patras’ pockmark field, acquired a couple of days after the 1993 M=5.4R Patras earthquake. White arrows point to bubble plumes (dark shades) rising from some the pockmarks (Papatheodorou et al., 2007).

Gas flow monitoring Short-term monitoring surveys were carried out using a methane sensor (METS; Christodoulou et al., 2003) in 14 pockmarks mapped by the geophysical survey. The background value of dissolved methane was ~1-99 nM, depending on sea conditions and the river outflow into the Patras Gulf. The overall concentration of methane in the pockmark field was between ~3-230 nM at the sea surface and from 33-270 nM just above the seabed. The most interesting results came from a complex pockmark in the southern sector of the field, namely P4. Methane concentration above this pockmark was found to vary from ~3- 99 nM in May, ~22-707 nM in September, ~96-370 nM in October, and ~101-105 nM in November, at sea surface and just above the seabed 68

respectively. Data collected in July 2003, one day after a 4.7 M earthquake,60 km east of the field, showed higher dissolved methane concentration— from ~420 nM at the sea surface to ~1,476 nM near the seabed over pockmark P4. During this period the overall methane concentration in the pockmark field was ~130–210 nM at sea surface and ~480–578 nM close to the seabed. Pockmarks which were activated by the 1993 Patras earthquake (Hasiotis et al., 1996), and which are in the northern sector of the field, showed low methane concentration (31–93 nM). Leifer et al. (2004) studied hydrocarbon gas emissions at an intense, 20-m-deep seep in Santa Barbara, California using a network of three turbine seep-tents and repeated seabed mapping. These funnel installations allow long-term spatial and temporal observations of gas discharges. This new approach was the only long-term monitoring of submarine gas seeps until the initiation of the ASSEM project in the Patras pockmark field. A new seafloor cabled observatory, the Gas Monitoring Module (GMM), was developed in the framework of the European Commission ASSEM project (Array of Sensors for long-term Seabed Monitoring of geohazards). GMM was based on a multiparametric approach in which the detection of gases is associated with that of key physicochemical factors, i.e. temperature, pressure and conductivity (Fig. 2.5.13a, b).

Gas detection was based on the use of CH4and H2S sensors. All sensors have a unique time reference and are controlled by a dedicated data-acquisition system. In the Patras pockmark field experiment, the

GMM included three semi-conductor CH4 sensors (Capsum METS), a

H2S electrode (AMT Microsensor), a CTD (SBE 37-SI Microcat) for temperature, salinity, pressure recording, as well as interfaces available for other sensors (Marinaro et al., 2006). GMM was linked to a shore station through an underwater cable (Fig. 2.5.13c), through which all data files recorded were periodically downloaded and it was also possible to switch sensors on/off and reboot the system.

69

Fig. 2.5.12 (a) The gas monitoring module (GMM) after final assemblage (from Marinaro et al., 2006); (b) the GMM during deployment in the pockmark P4 of the Patras pockmark field; (c) sketch of the GMM within the pockmark P4 and the link to the onshore station (from Marinaro et al., 2006); (d) 3.5 kHz profile showing the complex pockmark P4; and

(e) example of T–P decrease associated with positive CH4–H2S peaks at pockmark P4 (from Marinaro et al., 2006).

GMM was deployed in Patrakos Gulf pockmark field, over a composite, methane bearing pockmark (P4) at water depth of ~42 m

(Christodoulou et al., 2003; Fig. 2.5.12d). The CH4 sensor revealed the presence of high-concentration dissolved methane (up to 1,476 nM) in the water column just above the pockmark. The monitoring which was amounted to 201 days was carried out in two sites within the same pockmark (in total, 4,824 h of data acquisition). This GMM monitoring survey represents the first long-term (>6 months) monitoring ever done 70

on gas-bearing pockmarks with CH4, H2S, temperature, and pressure sensors. The most prominentresult of the survey was frequent sharp decreases in temperature and pressure. Over 60 T-P drops were recorded during the 6.5 months GMM monitoring period in P4 pockmark of Patraikos Gulf. For both monitoring sites inside the pockmark, all sharpT-P drops were associated with sharp increases in H2S and CH4 concentrations (Fig. 2.5.12e). This associated with T-P drops seepage pulsation could be: (i) driven by gas pressure build-up in the pockmark sediments and/or (ii) driven by oceanographic and morphological factors. Regarding the latter, Marinaro et al., (2006) suggested that bottom currents cascade into the pockmark with pressure drops and consequent gas “extraction” and/or pore-water burst from the sediments. The GMM results in P4seem to be in contrast with the direct correlation between T increase and gas escape as suggested by Hasiotis et al., (1996), which was probably related to a different fluid. No relation to seismicity was detected since no significant earthquakes occurred in Patras Gulf during the GMM monitoring period. The intensive surveying of Patraikos pockmarks suggested that the field was formed by gas venting, which episodically and for short duration is enhanced due to earthquakes. ROV observations The objective of the Remotely Operated Vehicle (ROV) survey was to identify the gas flaresdetected hydro-acoustically and explore the seabed for chemical, biological, and lithological evidence of active fluid flow. It is established that seafloor seepage sites may berecognized by the presence of chemosynthetic biological communities and facies characterized by authigenic mineralization (methane-bearing carbonates, sulfide minerals). An ROV based visual inspection inside the pockmark P4 revealed a uniform muddy seafloor (Fig. 2.5.13). The seafloor was textured with a slight relief and small holes, presumably associated with 71

infanual bioturbation (Fig. 2.5.13). No visual evidence for active fluid flow was reported during this survey.

Fig. 2.5.13 R.O.V. photos showing:(a) the muddy seafloor of the pockmark P4; (b), (c) the presence of small holes 5–10 cm in diameter(arrows) in the muddy cover of pockmark P4; and (d) the methane sensor (METS) above the floor of the pockmark P4 (Papatheodorou et al., 2007)..

Other sendimentological evidence Ravasopoulos et al. (2002) analyzed sediment cores collected inside and outside pockmarks of the Patras field. Grain-size, organic carbon, and heavy metal content were investigated, and the data were treated by multivariate statistics. The dominant process that appears to characterize the area was attributed to pollution due to the domestic sewage of the City of Patras. This process does not occur in the center of the most active, northern sector of the pockmark field. The authors attributed the lack of the sewage pollution to pockmark gas venting activity and proposed that this methodology couldrepresent anunconventional method for the detection of active pockmarks. 72

References Abrajano, T.A., Sturchio, N.C. Bohlke, J.K. Lyon, G.L. Poreda, R.J. Stevens, C.M. (1988). Methane-hydrogen gas seeps, Zambales Ophiolite, Philippines: Deep or shallow origin? Chem Geol, 71(1–3), pp. 211-222. https://doi.org/10.1016/0009-2541(88)90116-7 Acosta, J., Muñoz, A., Herranz, P., Palomo, C., Ballesteros, M., Vaquero, M., Uchupi, E., (2001). Pockmarks in Ibiza Channel and western end of the Balearic Promontory (Western Mediterranean) revealed by multibeam mapping. Geo-Mar Lett, 21, pp. 123–130. Åström, E.K.L., Oliver, P.G., Carroll, M.L. (2017). A new genus and two new species of Thyasiridae associated with methane seeps off Svalbard, Arctic Ocean, pp. 402-416. https://doi.org/10.1080/17451000.2016.1272699 Bagarinao, T. (1992). Sulfide as an environmental factor and toxicant: tolerance and adaptations in aquatic organisms. Aquat Toxicol, 24, pp. 21–62. Beauchamp, R.O. Jr., Bus, J.S., Popp, J.A., Boreiko, C.J., Andjelkovich, D.A. (1984). A critical review of the literature on hydrogen sulfide toxicity. Crit Rev Toxicol, 13(1), pp. 25-97. https://doi.org/10.3109/10408448409029321 Bernard, B.B., Brooks, J.M., Sackett, W.M. (1978). Light hydrocarbons in recent Texas continental shelf and slope sediments. J Geophys Res, 83, pp. 4053–4061. Boetius, A., Ravenschlag, K., Schubert, C., et al. (2000). A marine microbial consortium apparently mediating anaerobic oxidation of methane. Nature, 407, pp. 623–626 https://doi.org/10.1038/35036572 Bohrmann, G., Torres, M.E. (2006). Gas Hydrates in Marine Sediments. In: Schulz H.D., Zabel M. (Eds.) Marine Geochemistry. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-32144-6_14 Breen, N.A., Silver, E.A., Hussong, D.M. (1986). Structural styles of an accretionary wedge south of the island of Sumba, Indonesia revealed by Sea Marc II side scan sonar. Geol Soc Am Bull, 97(10), pp. 1,250- 1,261. 73

https://doi.org/10.1130/0016-7606(1986)97<1250:SSOAAW>2.0.CO;2 Brunskill, G.J., Burns, K.A., Zagorskis, I. (2011). Natural flux of greenhouse methane from the Timor Sea to the atmosphere. J Geophys Res, 116, G02024. https://doi.org/10.1029/2010JG001444 Bussmann, I., Suess, E. (1994). Groundwater seepage in Eckernforder Bay (western Baltic Sea): effect on methane and salinity distribution of the water column. Cont Shelf Res, 18, pp. 1,795–806. Chand, S., Thorsnes, T., Rise, L., Brunstad, H., Stoddart, D., Bøe, R., Lågstad, P., Svolsbru, T. (2012). Multiple episodes of fluid flow in the SW Barents Sea (Loppa High) evidenced by gas flares, pockmarks and gas hydrate accumulation. Earth Planet Sci Lett, 331, pp. 305–314. http://dx.doi.org/10.1016/j.epsl.2012.03.021 Chen, N., Yang, T.F., Hong, W. et al. (2020). Discharge of deeply rooted fluids from submarine mud volcanism in the Taiwan accretionary prism. Sci Rep 10, 381. https://doi.org/10.1038/s41598-019-57250-9 Christodoulou D., Papatheodorou G., Ferentinos G., Masson M. (2003). Active seepage in two contrasting pockmark fields in Patras and Corinth Gulfs, Greece. Geo-Marine Letters 23, pp. 194-199. Christodoulou, D., Papatheodorou, G., Fakiris E., Etiope G., Ferentinos G., (2009). The activation of the Patras Gulf pockmark field triggered by the June 8, 2008 earthquake. 9th Symposioum on Oceanography and Fisheries 2009 - Proceedings, Vol. Ι, pp. 3-8. Chronis, G., Piper, D.J.W., Anagnostou, G., 1991. Late quaternary evolution of the Gulf of Patras, Greece: tectonism, deltaic sedimentation and sea level change. Mar Geol, 97, pp. 191–209. Chung, H.M., Gormly, J.R., Squires, R.M. (1988). Origin of gases hydrocarbons in subsurface environments: theoretical considerations of carbon isotope distribution. Chem Geol, 71, pp. 97–104. Ciais, P., Sabine, C., Bala, G., et al. (2013). Carbon and Other Biogeochemical Cycles, in: Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of IPCC, edited by: Stocker, T. F., Qin, D., Plattner, G.-K., 74

Tignor, M., Allen, S. K., Boschung, J., Nauels, A., Xia, Y., Bex, V., and Midgley, P. M., Cambridge University Press, Cambridge, UK. Cita, M.B., Ryan, W.B.F., Paggi, L. (1981). Prometheus mud breccia: An example of shale diapirism in the western Mediterranean Ridge. Ann Geol Pays HelIen, 30, pp. 543-570. Clark, J.F., Washburn, L., Schwager, K. (2010). Variability of gas composition and flux intensity in natural marine hydrocarbon seeps. Geo-Mar Lett, 30, pp. 379–388. http://dx.doi.org/10.1007/s00367-009- 0167-1 Coleman, D.F., Ballard, R.D. (2001). A highly concentrated region of cold hydrocarbon seeps in the southeastern Mediterranean Sea. Geo- Mar Lett, 21, pp. 162-167. Cronin, B.T., Ivanov, M.K., Limonov, A.F., Egorov, A., Akhmanov, G.G., Akhmetjanov, A.M., Kozlova, E., Shipboard Scientific Party TTR-5. New discoveries of mud volcanoes on the Eastern Mediterranean Ridge. J Geol Soc, 154, pp. 173-182. https://doi.org/10.1144/gsjgs.154.2.0173 Dando, P.R., Hughes, J.A., Leahy, Y., Niven, S.J., Taylor, L.J, Smith, C. (1995a). Gas venting rates from submarine hydrothermal areas around the island of Milos, Hellenic volcanic arc. Cont Shelf Res, 15, pp. 913- 929. Dando, P.R., Hughes, J.A., Leahy, Y., Taylor, L.J., Zivanovic, S. (1995). Earthquakes increase hydrothermal venting and nutrient inputs into the Aegean. Cont Shelf Res, 15 (6), pp. 655-662. Davies, R.J., Stewart, S.A. (2005). Emplacement of giant mud volcanoes in the south Caspian Basin; 3D seismic reflection imaging of their root zones. J Geol Soc Lond, 162(1), pp. 1–4. Deville, E., Guerlais, S.-H., Callec, Y., Griboulard, R., Huyghe, P., Lallemant, S., Mascle, A., Noble, M., Schmitz, J. (2006). Liquefied vs stratified sediment mobilization processes: insight from the south of the Barbados accretionary prism. Tectonophysics, 428(1/4), pp. 33–47. Deville, E., Guerlais, S.H. (2009). Cyclic activity of mud volcanoes: evidences from Trinidad (SE Caribbean). Mar Petrol Geol, 26(9), pp. 1681–1691. 75

Deville, E., Scalabrin, C., Jouet, G., et al., (2020). Fluid seepage associated with slope destabilization along the Zambezi margin (Mozambique). Mar Geol, 428,106275, https://doi.org/10.1016/j.margeo.2020.106275 Diaz, R. J. (2001). Overview of hypoxia around the world. J Environ Qual, 30(2), pp. 275-281. https://doi.org/10.2134/jeq2001.302275x Diaz, R., Rosenberg, R. (2008). Spreading dead zones and consequences for marine ecosystems, Science, 321, pp. 926–929. https://doi.org/10.1126/science.1156401. Dimitrov, L. (2002). Contribution to atmospheric methane by natural gas seepages on the Bulgarian continental shelf. Contin Shelf Res, 22, pp. 2429–2442. Doutsos, T., Poulimenos, G., (1992). Geometry and kinematics of active faults and their seismotectonic significance in the western Corinth– Patras rift (Greece). J Struct Geol, 14 (6), pp. 689– 699. Elverhøi, A., DeBlasio, F.V., Butt, F.A., Issler, D., Harbitz, C.B., Engvik, L., Solheim, A., Marr, J. (2002). Submarine mass-wasting on glacially influenced continental slopes — processes and dynamics. J.A. Dowdeswell, C. O'Cofaigh Eds. Glacier-influenced Sedimentation on high Latitude Continental Margins, Geological Society of London Special Publication, 203, pp. 73-87. Etiope G., Papatheodorou G., Christodoulou D., Favali P., Ferentinos G., Sokos, E. (2006). Methane and hydrogen sulfide seepage in the NW Peloponnesus petroliferous basin (Greece): origin and geohazard. AAPG Bulletin, 90(5), pp. 701-713. Etiope, G. (2004). GEM – Geologic Emissions of Methane, the missing source in the atmospheric methane budget. Atmospheric Environment, 38, 19, pp. 3,099-3,100. Etiope, G. (2012). Methane uncovered. Nat Geosci, 5, pp. 373–374. Etiope, G. (2015). Natural Gas Seepage. The Earth’s hydrocarbon degassing. Springer, Switzerland, 199 pp. Etiope, G., Christodoulou, D., Kordella, S., Marinaro, G., Papatheodorou, G. (2013a). Offshore and onshore seepage of 76

thermogenic gas at Katakolo Bay (Western Greece). Chem Geol, 339, pp. 115–126. https://doi.org/10.1016/j.chemgeo.2012.08.011 Etiope, G., Ciotoli, G., Schwietzke, S., Schoell, M. (2019). Gridded maps of geological methane emissions and their isotopic signature. Earth Syst Sci Data, 11, pp. 1–22. https://doi.org/10.5194/essd-11-1- 2019 Etiope, G., Feyzullayev, A., Baciu, C.L. (2009). Terrestrial methane seeps and mud volcanoes: a global perspective of gas origin. Mar Petrol Geol, 26, pp. 333–344. Etiope, G., Lassey, K. R., Klusman, R. W., Boschi, E. (2008). Reappraisal of the fossil methane budget and related emission from geologic sources, Geophys Res Lett, 35, L09307. https://doi.org/10.1029/2008GL033623 Etiope, G., Martinelli, G., Caracausi, A., Italiano, F. (2007). Methane seeps and mud volcanoes in Italy: Gas origin, fractionation and emission to the atmosphere Geophys Res Lett, 34, L14303. https://doi.org/10.1029/2007GL030341 Etiope, G., Milkov, A.V., (2004). A new estimate of global methane flux from onshore and shallow submarine mud volcanoes to the atmosphere. Environ Geol, 46, pp. 997-1002. Etiope, G., Nakada, R., Tanaka, K., Yoshida, N., (2011a). Gas seepage from Tokamachi mud volcanoes, onshore Niigata Basin (Japan): origin, post-genetic alterations and CH4\CO2 fluxes. Applied Geochemistry 26, pp. 348–359. Etiope, G., Papatheodorou, G., Christodoulou, D., Favali, P., Ferentinos, G. (2005). Gas Hazard Induced by Methane and Hydrogen Sulfide Seepage in the NW Peloponnesus Petroliferous Basin (Greece). TAO, 16(4), pp. 897-908. https://doi.org/10.3319/TAO.2005.16.4.897(GIG) Etiope, G., Schoell, (2014). Abiotic gas: Atypical, but not rare. Elements, 10, pp. 291–296. https://doi.org/10.2113/gselements.10.4.291 77

Etiope, G., Schoell, M., Hosgormez, H. (2011b). Abiotic methane flux from the Chimaera seep and Tekirova ophiolites (Turkey): understanding gas exhalation from low temperature serpentinization and implications for Mars. Earth Planet Sci Lett, 310, pp. 96–104. Etiope, G., Sherwood Lollar, B. (2013b). Abiotic methane on Earth. Rev Geophys. 51, pp. 276– 299. https://doi.org/10.1002/rog.20011 Etiope, G., Whiticar, M.J. (2019). Abiotic methane in continental ultramafic rock systems: Towards a genetic model. Appl Geochem, 102, pp. 139–152. Fader, G.B.J. (1991). Gas‐related sedimentary features from the eastern Canadian continental shelf. Cont Shelf Res, 11, pp. 1,123– 1,153. https://doi.org/10.1016/0278‐4343(91)90094‐M Ferentinos, F., Papatheodorou, G., Geraga, M., Iatrou, M., Fakiris, E., Christodoulou, D., Dimitriou, E., Koutsikopoulos, C. (2010). Fjord water circulation and dysoxic/anoxic conditions in a Mediterranean semi- enclosed embayment in Amvrakikos Gulf, Greece. Estuar Coast Shelf S, 88, pp. 473–481. Ferentinos, G., Brooks, M., Doutsos, T. (1985). Quaternary tectonics in the Gulf of Patras, Western Greece. J Struct Geol. 7(6), pp. 713– 717. Ferentinos, G., Fakiris, E., Christodoulou, D., Geraga, M., Dimas, X., Georgiou, N., Kordella, S., Papatheodorou, G., Prevenios, M., Sotiropoulos, M., (2020).Optimal side scan sonar and subbottom profiler surveying of ancient wrecks: The ‘Fiskardo’ wreck, Kefallinia Island, Ionian Sea,Journal of Archaeological Science, 113,105032, https://doi.org/10.1016/j.jas.2019.105032 Ferentinos, G., Papatheodorou, G., Collins, M.B. (1988), Sediment Transport processes on an active submarine fault escarpment: Gulf of Corinth, Greece, Marine Geology, 83(1–4), pp. 43-61. https://doi.org/10.1016/0025-3227(88)90051-5 Flotte, N., Sorel, D., Muller, C., Tensi, J. (2005). Along strike changes in the structural evolution over a brittle detachment fault: Example of the Pleistocene Corinth–Patras rift (Greece). Tectonophysics, 403, pp. 77– 94. 78

formed during petroleum biodegradation in conventional reservoirs. Org Geochem, 42, pp. 184–207. Friedrich, J., Janssen, F., Aleynik, D. et al., (2014). Investigating hypoxia in aquatic environments: diverse approaches to addressing a complex phenomenon. Biogeosciences, 11, pp. 1215–1259. https://doi.org/10.5194/bg-11-1215-2014 Garcia-Garcia, A., Vilas, F., Garcia-Gil, S. (1999). A seeping seafloor in a Ria environment: Ria de Vigo (NW Spain). Environ Geol, 38, pp. 296– 300. Gay, A., Lopez, M., Ondreas, H., Charlou, J.-L., Sermondadaz, G., Cochonat, P. (2006). Seafloor facies related to upward methane flux within a Giant Pockmark of the Lower Congo Basin. Mar Geol, 226(1– 2), pp. 81-95, https://doi.org/10.1016/j.margeo.2005.09.011 Géli, L., Cagatay, N., Henry, P., Gasperini, L., Favali, P., Cifci, G. (2011). Towards permanent, multi‐disciplinary seafloor observatories in the Sea of Marmara: Results from the Marmara Demonstration Mission of ESONET/NoE, paper presented at IEEE Symposium/Workshop on Scientific Use of Submarine Cables and Related Technologies, Tokyo. Gerivani, H., Putans, V.A, Merklin, L.R., Modarres, M.H. (2020). Characteristics of features formed by gas hydrate and free gas in the continental slope and abyssal plain of the Middle Caspian Sea, Mar Georesources & Geotechnol. https://doi.org/10.1080/1064119X.2019.1709585 Girard, F., Sarrazin, J., Olu, K. (2020). Impacts of an eruption on cold- seep microbial and faunal dynamics at a mud volcano. Front Mar Sci, 7, pp. 241. https://doi.org/10.3389/fmars.2020.00241 Gorkun, V.N., Siryk, I.M. (1968). Calculating depth of deposition and volume of gas expelled during eruptions of mud volcanoes in southern Sakhalin. Int Geol Rev, 10(1), pp. 4-12. https://doi.org/10.1080/00206816809474846 Greinert, J., Bialas, J., Lewis, K., Suess, E. (2010). Methane seeps at the Hikurangi Margin, New Zealand. Marine Geology. 272 (1–4), pp. 1- 3. https://doi.org/10.1016/j.margeo.2010.02.018 79

Gullapalli, S., Dewangan, P., Kumar, A., Dakara, G., Mishra, C.K. (2019). Seismic evidence of free gas migration through the gas hydrate stability zone (GHSZ) and active methane seep in Krishna-Godavari offshore basin. Marine and Petroleum Geology, 110, pp. 695-705. Harrington, P.K. 1985. Formation of pockmarks by porewater escape. Geo-Mar Lett, 5, pp. 193-197. https://doi.org/10.1007/BF02281638 Hasiotis, T., Papatheodorou, G., Kastanos, N. and Ferentinos, G. (1996): A pockmark field in the Patras Gulf (Greece) and its activation during the 14/7/1993 seismic event. Mar Geol, 130, pp. 333-344. https://doi.org/10.1016/0025-3227(95)00131-X Hedberg, H.D. (1974). Relation of methane generation to undercompacted shales, shale diapirs, and mud volcanoes. Am Assoc Petrol Geol Bull. 58, pp. 661–673. https://doi.org/10.1306/83D91466- 16C7-11D7-8645000102C1865D Hornafius, J.S., Quigley, D., Luyendyk, B.P. (1999). The world’s most spectacular marine hydrocarbon seeps (Coal Oil Point, Santa Barbara Channel, California): quantification of emissions. J Geophys Res, 104(C9), pp. 20703– 20711. https://doi.org/10.1029/1999JC900148 Hovland, M. (1983). Elongated depressions associated with pockmarks in the western slope of the Norwegian Trench. Mar Geol, 51, pp. 35–46. Hovland, M., Gardner, J.V., Judd, A.G. (2002). The significance of pockmarks to understanding fluid flow processes and geohazards. Geofluids, 2, pp. 127–136. Hovland, M., Judd, A., (1988). Seabed pockmarks and seepages: Impact on geology, biology and the marine environment, Graham & Trotman, London. https://doi.org/10.13140/RG.2.1.1414.1286 Hovland, M., Talbot, M.R., Quale, H., Olaussen, S., Aasberg, L. (1987). Methane-related carbonate cements in pockmarks of the North Sea. J Sediment Petrol, 57, pp. 881–92. Hunt, J.M. (1996). Petroleum geochemistry and geology. 2nd ed. W.H. Freeman and Co, New York, USA, pp. 743. hydrates. Mar Geol, 167, pp. 29–42. 80

Imray, J. (1868). Sailing Directions for the West Coast of North America, Part 1. James Imray, London. Iqbal, A., Shang, Z., Rehman, M.L.U., Ju, M., Rehman, M.M.U., Rafiq, M.K., Ayub, N., Bai, Y., (2019). Pattern of microbial community composition and functional gene repertoire associated with methane emission from Zoige wetlands, China - A review. Sci Total Environ, 694,133675. Isola, J.I., Ormazabal, J.P, Flores, G., Arismendi, S., Druet, M., Muñoz, A., del Río, J.L., Etienot, S.D.A., Gomez Ballesteros, M.P., Principi, S., Bolatti, N.D., Tassone, A.A. (2020). Tectonic and structural controls on Neogene fluid release in the Patagonian Continental Margin. Mar Pet Geol, 115, 104246. https://doi.org/10.1016/j.marpetgeo.2020.104246 Judd, A., Hovland, M. (2007). Seabed Fluid Flow. The impact on Geology, Biology and the Marine Environment, Cambridge University Press, Cambridge CB2 2RU, UK, pp. 475. Judd, A., Noble-James, T., Golding, N. et al. (2020). The Croker Carbonate Slabs: extensive methane-derived authigenic carbonate in the Irish Sea—nature, origin, longevity and environmental significance. Geo-Mar Lett, 40, pp. 423–438. https://doi.org/10.1007/s00367-019- 00584-0 Judd, A.G, (2000). Geological sources of methane. In Atmospheric methane: Its role in the global environment. Khalil, M.A.K. Ed., Springer, Berlin, Springer, Berlin, pp. 280–303. Judd, A.G. (2004). Natural seabed gas seeps as sources of atmospheric methane. Env Geol, 46, pp. 988–996. https://doi.org/10.1007/s00254-004-1083-3 Judd, A.G., Davies, J., Wilson, J., Holmes, R., Baron, G., Bryden, I. (1997). Contributions to atmospheric methane by natural seepages on the UK continental shelf. Mar Geol, 137, pp. 165–189. Judd, A.G., Hovland, M., Dimitrov, L.I., Garcia Gil, S., Jukes, V. (2002). The geological methane budget at continental margins and its influence on climate change. Geofluids, 2, pp. 109–126. 81

Kelley, J.T., Dickson, S.M., Belknap, D.F., Barnhardt, W.A., Henderson, M., (1994). Giant sea-bed pockmarks: Evidence for gas escape from Belfast Bay, Maine. Geology, 22, pp. 59-62. Kessler, J. D., Reeburgh, W. S., Southon, J., Varela, R. (2005). Fossil methane source dominates Cariaco Basin water column methane geochemistry. Geophys Res Lett, 32, L12609. King, L. H., McLean, B. (1970). Pockmarks on the Scotian shelf. Geological Society of America, 81, pp. 3141-8. Klerkx, J. De Batist, M. Poort, J., Hus R., Van Rensbergen, P., Khlystov, O., Granin, N. (2006). Tectonically controlled methane escape in Lake Baikal. In: Lombardi S., Altunina L., Beaubien S. (Eds) Advances in the Geological Storage of Carbon Dioxide. Nato Science Series: IV: Earth and Environmental Sciences. Vol. 65. Springer, Dordrecht. Klusman, R.W., Jakel, M.E., LeRoy, M.P. (1998). Does microseepage of methane and light hydrocarbons contribute to the atmospheric budget of methane and to global climate change? Assoc Petrol Geochem Explor Bull, 11, pp. 1–55. Kohl, B., Roberts, H.H. (1994). Fossil foraminifera from four active mud volcanoes in the Gulf of Mexico. Geo-Mar Lett, 14, pp. 126-134. Kohout, F.A. (1966). Submarine springs: a neglected phenomenon of coastal hydrology. Hydrology, 26, pp. 391–413. Kojima S., Watanabe H. (2015) Vent Fauna in the Mariana Trough. In: Ishibashi J., Okino K., Sunamura M. (Eds) Subseafloor Biosphere Linked to Hydrothermal Systems. Springer, Tokyo. https://doi.org/10.1007/978- 4-431-54865-2_25 Kutas, R.I., Paliy, S.I., Rusakov, O.M. (2004). Deep faults, heat flow and gas leakage in the Northern Black Sea. Geo-Mar Lett, 24, pp. 163. https://doi.org/10.1007/s00367-004-0172-3 Kvenvolden, K.A., Lorenson, T.D., Reeburgh, W. (2001). Attention turns to naturally occurring methane seepage. EOS, 82, pp. 457. Kvenvolden, K.A., Rogers, B.W. (2005). Gaia's breath global methane exhalations. Mar Petrol Geol, 22, pp. 579–590. 82

Lacroix, A.V. (1993). Unaccounted-for sources of fossil and isotopically enriched methane and their contribution to the emissions inventory: a review and synthesis. Chemosphere, 26, pp. 507–557. Le Pichon, X., Henry, P., Lallemant, S. (1990). Water flow in the Barbados accretionary complex. J Geophys Res, 95, pp. 8,945-8,967. Leifer, I., Patro, R.K. (2002). The bubble mechanism for methane transport from the shallow seabed to the surface: a review and sensitivity study. Cont Shelf Res, 22, pp. 2,409–2,428. Levin, L.A., Baco, A.R., Bowden, D.A., et al. (2016). Hydrothermal Vents and Methane Seeps: Rethinking the Sphere of Influence. Front Mar Sci, 3, 72. https://doi.org/10.3389/fmars.2016.00072 Ligtenberg, J.H. (2005). Detection of fluid migration pathways in seismic data: implications for fault seal analysis. Basin Res, 17, pp. 141-153. https://doi.org/10.1111/j.1365-2117.2005.00258.x Lloyd, K.G., Albert, D.B., Biddle, J.F., Chanton, J.P., Pizarro, O., et al. (2010). Spatial Structure and Activity of Sedimentary Microbial Communities Underlying a Beggiatoa spp. Mat in a Gulf of Mexico Hydrocarbon Seep. PLoS ONE, 5(1), e8738. https://doi.org/10.1371/journal.pone.0008738 Lutken, B., Simonetti, A., Ingrassia, M., Macelloni, L., Knapp, H. Fisher, C., Caruso, S., D’Emidio, M. (2011). Biogeophysical classification of seafloor seeps at a carbonate–hydrate mound, northern Gulf of Mexico. In: AAPG International Conference and Exhibition, Milan, Italy, October 23–27, 2011. Lykousis, V. (1989). The echo-character and geotechnical properties of the deltaic deposits of the Mornos, Piros and Evinos rivers in the Patras and Corinth gulfs. Ph.D. Thesis, Patras Univ. Macelloni, L., Brunner, C.A., Caruso, S., Lutken, C.B., D'Emidio, M., Lapham, L.L. (2013). Spatial distribution of seafloor bio-geological and geochemical processes as proxies of fluid flux regime and evolution of a carbonate/hydrates mound, northern Gulf of Mexico, Deep Sea Research Part I: Oceanographic Research Papers. 74, pp. 25-38. https://doi.org/10.1016/j.dsr.2012.12.006 83

Marinaro, G., Etiope, G., Gasparoni, F., Furlan, F., Bruni, F. (2011).Gas seepage detection and monitoring at seafloor.10th Offshore Mediterranean Conference and Exhibition, Ravenna, Italy, March 23– 25, 2011, “Frontier exploration” session. Marinaro, G., Etiope, G., LoBue, N., Favali, P., Papatheodorou, G., Christodoulou, D., Furlan, F., Gasparoni, F., Ferentinos, G., Masson, M., Rolin, J.F. (2006). Monitoring of a methane-seeping pockmark by cabled benthic observatory (Patras Gulf, Greece). Geo-Mar Letters. https://doi.org/10.1007/s00367-006-0040-4 Marques, J.M., Etiope, G., Neves, M.O., Carreira, P.M., Rocha, C., Vance, S.D., Christensen, L., Miller, A.Z., Suzuki, S. (2018). Linking serpentinization, hyperalkaline mineral waters and abiotic methane production in continental peridotites: An integrated hydrogeological-bio- geochemical model from the Cabeço de Vide CH4-rich aquifer (Portugal). Appl Geochem, 96, pp. 287–301. Mc Ginnis, D.F., Greinert, J., Artemov, Y., Beaubien, S.E., Wüest, A. (2006). Fate of rising methane bubbles in stratified waters: how much methane reaches the atmosphere? J Geophys Res, 111, C09007. methane seepage. EOS. 82, p 457. https://doi.org/10.1029/01EO00275 Milkov, A.V. (2000). Worldwide distribution of submarine mud volcanoes and associated gas hydrates. Mar Geol, 167(1-2), pp. 29-42. https://doi.org/10.1016/S0025-3227(00)00022-0 Milkov, A.V. (2011). Worldwide distribution and significance of secondary microbial methane formed during petroleum biodegradation in conventional reservoirs. Org Geochem, 42, pp. 184–207. Mitsis, C., Papatheodorou, G., Ravasopoulos, J., Christodoulou, D., Ferentinos, G. (2002). A multivariate statistical approach to the investigation of pockmarks growth and activity. An example from a pockmark field in the Gulf of Patras (W. Greece) (in Greek). Proc. of 6th Pan-Hellenic Geographical Conference, Thessalonica, 3-6 October 2002, Vol. Ι, pp 194-201. Monteiro, P.M.S., van der Plas, A., Mohrholz, V., Mabille, E., Pascall, A., Joubert, W. (2006). Variability of natural hypoxia and methane in a 84

coastal upwelling system: oceanic physics or shelf biology? Geophys Res Lett, 33, L16614. https://doi.org/10.1029/2006GL026234 Narula, K. (2019). Oceans as a Source of Hydrocarbon Energy. In: The Maritime Dimension of Sustainable Energy Security. Lecture Notes in Energy. Vol. 68. Springer, Singapore. https://doi.org/10.1007/978-981- 13-1589-3_7 Nayar, A. (2009). A lakeful of trouble. Nature, 460, pp. 321–323. https://doi.org/10.1038/460321a Nelson, C.S., Healy, T.R. (1984). Pockmark‐like structures on the Poverty Bay seabed—Possible evidence for submarine mud volcanism. New Zealand J Geol Geoph, 27, pp. 225–230. Nelson, H., Thor, D.R., Sandstrom, M.W., Kvenvolden, K.A. (1979). Modern biogenic gas-generated craters (sea-floor ‘pockmarks’) on the Bering Shelf. Alaska. Geol Soc Am Bull, 90, pp. 1144-1152. Neurauter, T.W., Roberts, H.H. (1994). Three generations of mud volcanoes on the Louisiana continental slope. Geo-Mar Lett, 14, pp. 120–125. https://doi.org/10.1007/BF01203723 Newton, R., Cunningham, R., Schubert, C. (1980). Mudvolcanoes and pockmarks: seafioor engineering hazards orgeological curiosities? Offshore Technol. Conf. (Houston, Texas). Pap. 3729, pp. 425-435. Nilsson, H. C., Rosenberg, R. (2000). Succession in marine benthic habitats and fauna in response to oxygen deficiency: analysed by sediment profile-imaging and by grab samples. Mar Ecol -Prog Ser. 197, pp. 139–149. Orange, D.L., Greene, H.G., Reed, D., Martin, J.B., McHugh, C.M., Ryan, W.B.F., Maher, N., Stakes, D., Barry, J. (1999). Widespread fluid expulsion on a translational continental margin: mud volcanoes, fault zones, headless canyons, and aquifers in Monterey Bay, California. Geol Soc Am Bull, 111, pp. 992–1009. Ostanin, I., Anka, Z., di Primio, R., Bernal, A., (2013). Hydrocarbon plumbing systems above the Snøhvit gas field: Structural control and implications for thermogenic methane leakage in the Hammerfest 85

Basin, SW Barents Sea. Mar Petrol Geol.43, pp. 127-146. https://doi.org/10.1016/j.marpetgeo.2013.02.012 Papatheodorou G., Ferentinos G. (1997). Submarine and coastal sediment failure triggered by the 1995, Ms=6.1R Aegion earthquake, Gulf of Corinth, Greece. Mar Geol, 137, pp. 287-304. Papatheodorou G., Hasiotis T. and Ferentinos G. (1993). Gas charged sediments in the Aegean and Ionian Seas, Greece. Mar Geol, 112, pp. 171-184. Papatheodorou, G., Demopoulou, G., Lambrakis, N. (2006). A long- term study of temporal hydrochemical data in a shallow lake using multivariate statistical techniques. Ecol Modell, 193, pp. 759–776. Papatheodorou, G., Lavrentaki, M., Mourelatos, P., Voutsinas, K, Xenos, K. (2001). Pockmarks on the seabed of the Aetoliko Lagoon, Greece (in Greek). Alieftika Nea (Fishing News), 238, pp. 73–87. Papatheodorou, G., Christodoulou, D., Geraga, M., Etiope, G., Ferentinos, G. (2007). “The pockmark field of the Gulf of Patras: An ideal natural laboratory for studying seabed fluid flow” In: ZELILIDIS, A., PAPATHEODOROU, G., and GERAGA, M. (Eds.): Sedimentology of western and central Greece from recent to Triassic. 25th IAS Meeting of Sedimentology, Patras, 2007, Field Trip Guidebook, pp. 43-62. Paull, C., Ussler III, W. Maher, N. Greene, H.G. Rehder, G. Lorenson, T. Lee H. (2002). Pockmarks off Big Sur, California. Mar Geol, 181, pp. 323-335. Pearson, T.H., Rosenberg, R. (1978). Macrobenthic succession in rela- tion to organic enrichment and pollution of the marine environment, Oceanogr Mar Biol, 16, pp. 229–311. Pelletier, B.R. (1980). Geology and Physiography. In: The Lancaster Sound Regional Study, Background Report, Indian and Nothern Affairs, Canada, pp. 104. Pickrill, R.A. (1993). Shallow seismic stratigraphy and pockmarks of a hydrothermally influenced lake, Lake Rotoiti, New Zealand. Sedimentology, 40, pp. 813-828. https://doi.org/10.1111/j.1365- 3091.1993.tb01363.x 86

Pierre, C., Bayon, G., Blanc-Valleron, M. et al. (2014). Authigenic carbonates related to active seepage of methane-rich hot brines at the Cheops mud volcano, Menes caldera (Nile deep-sea fan, eastern Mediterranean Sea). Geo-Mar Lett, 34, pp. 253–267. https://doi.org/10.1007/s00367-014-0362-6 Plaza-Faverola, A., Bünz, S., Mienert, J. (2011). Repeated fluid expulsion through sub-seabed chimneys offshore Norway in response to glacial cycles. Earth Planet Sci Lett, 305(3–4), pp. 297–308. Prior, D.B., Coleman, J.B. (1984). Submarine slope instability. Brunsden, D., Prior, D.B. (Eds.), Slope Instability, Wiley, New York, pp. 419-455. Rabalais, N.N., Gilbert, D. (2008). Distribution and consequences of hypoxia. Urban, E., Sundby, B., Malanotte-Rizzoli, P., Melillo, J.M. (Eds.), Watersheds, Bays and Bounded Seas, Island Press, Washington DC, USA, pp. 209-226. Ravasopoulos, J., Papatheodorou, G., Kapolos, J., Geraga, M., Koliadima, A., Xenos., K. (2002). A promising methodology for the detection of pockmark activation in nearshore sediments, Gulf of Patras (W. Greece). Instrum Sci Technol, 30, 2, pp. 139-155. Roberts, H.H., Aharon, P., Carney, R., Larkin, J., Sassen, R. (1990). Seafloor response to hydrocarbon seeps, Louisiana continental slope. Geo-mar Lett, 10, pp. 232-243. Rogers, J., Kelley, J., Belknap, D.F., Gontz, A., Barnhardt, W.A. (2006). Shallow-water pockmark formation in temperate estuaries: a consideration of origins in the western Gulf of Maine with special focus on Belfast Bay. Mar Geol, 225, pp. 45– 62.https://doi.org/10.1016/j.margeo.2005.07.011 Roy, N., Molson, J., Lemieux, J.M., Van Stempvoort, D., Nowamooz, A. (2016). Three‐dimensional numerical simulations of methane gas migration from decommissioned hydrocarbon production wells into shallow aquifers. Water Resour Res, 52, pp. 5,598–5,618. https://doi.org/10.1002/2016WR018686 87

Sabatier, P., Senderens, J.-B. (1902). Nouvelles syntheses du methane. Comptes Rendus Des Séances De L‘Académie Des Sciences. Section VI – Chimie, 134, pp. 514-516. Sahling, H., Masson, D. G., Ranero, C. R., Hühnerbach, V., Weinrebe, W., Klaucke, I., Bürk, D., Brückmann, W., and Suess, E. (2008), Fluid seepage at the continental margin offshore Costa Rica and southern Nicaragua. Geochem Geophys Geosyst, 9, Q05S05. https://doi.org/10.1029/2008GC001978 Sakellariou, D., Lykousis, V., Alexandri, S., Kaberi, H., Rousakis, G., Nomikou, P., Georgiou, P., Ballas, D. (2007). Faulting, seismic- stratigraphic architecture and late quaternary evolution of the Gulf of Alkyonides basin-east Gulf of Corinth, central Greece. Basin Res, 19(2), pp. 273-295. https://doi.org/10.1111/j.1365-2117.2007.00322.x Sassen, R., Roberts, H.H., Aharon, P., Larkin, J., Chinn, E.W., Carney, R. (1993) Chemosynthetic bacterial mats at cold hydrocarbon seeps, Gulf of Mexico continental slope. Org Geochem, 20, pp. 77-89. https://doi.org/10.1016/0146-6380(93)90083-N Saunois, M., Bousquet, P., Poulter, B., et al. (2016). The global methane budget 2000–2012. Earth Syst Sci Data, 8(2), pp. 697-751. https://doi.org/10.5194/essd-8-697-2016 Schmale, O., Greinert, J., Rehder, G., 2005. Methane emission from high-intensity marine gas seeps in the Black Sea into the atmosphere. Geophys Res Lett, 32, L07609. Schoell, M. (1980). The hydrogen and carbon isotopic composition of methane from natural gases of various origins. Geochim Cosmochim Acta, 44, pp. 649–661. Schwietzke, S., Sherwood, O. A., Bruhwiler, L. M. P., Miller, J. B., Etiope, G., Dlugokencky, E. J., Michel, S. E., Arling, V. A., Vaughn, B. H., White, J. W. C., Tan, P.P. (2016). Upward revision of global fossil fuel methane emissions based on isotope database. Nature, 538, pp. 88–91. Sciarra, A., Saroni, A., Etiope, G., Coltorti, M., Mazzarini, F., Lott, C., Grassa, F., Italiano, F. (2019). Shallow submarine seep of abiotic 88

methane from serpentinized peridotite off the Island of Elba, Italy. Appl Geochem, 100, pp. 1-7. https://doi.org/ 10.1016/j.apgeochem.2018.10.025 Sibuet, M., Olu, K. (1998). Biogeography, biodiversity and fluid dependence of deep-sea cold-seep communities at active and passive margins. Deep Sea Res Part II Top Stud Oceanogr, 45, pp. 517–567. Stahl, W.J. (1977). Carbon and nitrogen isotopes in hydrocarbon research and exploration. Chem Geol, 20, pp. 121–149. Stanley, D.J., Perissoratis, C. (1977). Aegean Sea ridge barrier-and- basin sedimentation patterns, Mar Geol, 24 (2), pp. 97-107. https://doi.org/10.1016/0025-3227(77)90003-2 Sultan, N., Cochonat, P., Foucher, J.-P., Mienert, J. (2004). Effect of gas hydrates melting on seafloor slope instability. Mar Geol, 213 (1-4), pp. 379-401. Tao, C., Seyfried, W.E., Lowell, R.P. et al. (2020). Deep high- temperature hydrothermal circulation in a detachment faulting system on the ultra-slow spreading ridge. Nat Commun, 11, 1300. https://doi.org/10.1038/s41467-020-15062-w Thiagarajan, N., Kitchen, N., Xie, H., Ponton, C., Lawson, M., Formolo, M., Eiler, J. (2020). Identifying thermogenic and microbial methane in deep water Gulf of Mexico Reservoirs. Geochim Cosmochim Acta, 275, pp. 188-208. Truong, D.H., Eghbal, M.A., Hindmarsh, W., Roth, S.H., O’Brien, P.J. (2006). Molecular mechanisms of hydrogen sulfide toxicity. Drug Metab Rev, 38, pp. 733–744. Uchupi, E., Swift, S.A., Ross D.A. (1996). Gas venting and late Quaternary sedimentation in the Persian (Arabian) Gulf. Mar Geol, 129, pp. 237-269. Ussler, W., III, Paull, C.K. Boucher, J. Friederich, G.E., Thomas, D.J. (2003). Submarine pockmarks: a case study from Belfast Bay, Maine. Mar Geol, pp. 3390 1-18. Vadakkepuliyambatta, S., Bünz, S., Mienert, J., Chand S. (2013). Distribution of subsurface fluid‐flow systems in the SW Barents Sea. 89

Mar Pet Geol, 43, pp. 208–221. https://doi.org/10.1016/j.marpetgeo.2013.02.007 Vogt, P.R., Cherkashev, G.A., Ginsburg, G.D., KIvanov, M., Milkov, A.V., Crane, K., Lein, A., Sundvor, E., Pimenov, N., Egorov, A. (1997). Haakon Mosby mud volcano provides unusual example of venting. EOS. Transactions of the American Geophysical Union, 78, 549, pp. 556-557. Wang, D.T., Reeves, E.P., McDermott, J.M., Seewald, J.S., Ono, S. (2018). Clumped isotopologue constraints on the origin of methane at seafloor hot springs. Geochim Cosmochim Acta, 223, pp. 141–158 Wang, F., Chapman, P.M. (1999). Biological implications of sulfide in sediment—a review focusing on sediment toxicity. Environ Toxicol Chem, 18, pp. 2526–2532. Ward, B.B., Kilpatrick, K.A., Novelli, P. C., Scranton, M.I. (1987). Methane oxidation and methane fluxes in the ocean surface layer and deep anoxic waters. Nature, 327, pp. 226–229. Westbrook, G. K., Thatcher, K.E., Rohlinget, E.J., et al. (2009). Escape of methane gas from the seabed along the West Spitsbergen continental margin. Geophys Res Lett. 36, L15608, https://doi.org/10.1029/2009GL039191 Whiticar, M.J. (1999) Carbon and hydrogen isotope systematics of bacterial formation and oxidation of methane. Chem Geol, 161(1/3), pp. 291–314. https://doi.org/10.1016/S0009- 2541(99)00092-3 Whiticar, M.J. (2002). Diagenetic relationships of methanogenesis, nutrients, acoustic turbidity, pockmarks and freshwaterseepage in Eckernforder Bay. Mar Geol, 182, pp. 29-53. Yerkes, R.F., Wagner, H.C., Yenne, K.A. (1969). Petroleum development in the Santa Barbara channel region. In: US Geological Survey Professional Paper 679, 1969. Geology, Petroleum Development, and Seismicity of the Santa Barbara Channel Region, California.Superintendent of Documents, U.S. Government Printing Office, Washington, D.C. 90

Zhao, W.L., Davis, D.M., Dahlen, F.A., Suppe, J. (1986). Οrigin of convex accretionary wedges: Evidence for Barbados. J Geophys Res, 91, pp. 10,246-10,258.

91

3. Geological setting

4. General Geological Setting of Western Greece Western Greece is largely dominated by the most external zones of the Hellenide fold-and-thrust belt which was formed during the early Eocene (Zelilidis et al., 2015). From west to east these are the pre- Apulian (or Paxoi), Ionian (where the study sites included in this thesis are located), and Gavrovo geotectonic zones (Fig. 3.1; Zelilidis et al., 2015; Karakitsios and Rigakis, 2007; Zelilidis et al., 2003; Karakitsios, 1995).

Fig. 3.1. Simplified geological map of Western Greece, illustrating the principal tectonostratigraphic zones of the external Hellenides fold-and-thrust belt: Pindos, Gavrovo, Ionian and Pre-Apulian (simplified from Zelilidis et al., 2015). The study sites of this thesis Katakolo and Amvrakikos Gulf are also marked.

92

Western Greece was part of the Apulian continental block on the southern passive margin of Tethys from the Triassic to Late Cretaceous (Karakitsios and Rigakis, 2007). During the Late Cretaceous the Dinaride-Albanide-Hellenide continental convergent zone formed a thrust by the subduction of the oceanic crust of the African plate under the Eurasian plate, which started from middle-upper Miocene (e.g., Doutsos et al.,1993; Papanikolaou, 2009). Subduction of basement rocks in the southern Hellenides occurs along the Hellenic trench but a thrust complex extends as far as the Mediterranean Ridge (Fig.3.1; Vassilakis et al., 2011; Finetti, 1982; Kreemer and Chamot-Rooke, 2004). Deposits rich in organic carbon in Western Greece are accossiated with siliceous facies (Karakitsios and Rigakis, 2007). Local source rocks contain marine organic material althoughin siliciclastic sediments terrigenous organic matter has also been found (Karakitsios and Rigakis, 2007). Field observations between the Pre-Apulian and Ionian zone have confirmed that their boundaries are associated with intrusive evaporites between the Ionian and pre-Apulian thrusts and foldseven where the precise location of the thrust is unclear (Karakitsios and Rigakis, 2007; Karakitsios and Pomoni-Papaioannou, 1998). The location of the evaporites in combination with their occurrence in tectonic windows above tectonized flysch, suggests that they correspond to the lowest detachment level of overthrust sheets in the external Hellenides fold-and-thrust belt (Karakitsios and Rigakis, 2007). The evaporites in Western Greece show great thickness locally (>3 km in boreholes in the Ionian Zone; Karakitsios and Rigakis, 2007; BP, 1971; IGRS-IFP, 1966; Zelilidis et al., 2015). The Pre-Apulian (Paxoi) zone The Pre-Apulian or Paxoi zone is the most external zone of the Hellenides fold-and-thrust belt (Fig. 3.1) and consist mainly Triassic to Miocene neritic-pelagic carbonates. It is afairlyuniform Mesozoic to Cenozoic ramp mainly coprised of carbonates,transitional between the 93

shallow-water Apulian Platform and the deeper-water Ionian Basin (Getsos et al., 2007; Zelilidis et al., 2015). Due tointense deformation resulting from active tectonics it has a complex setting (Accordi et al., 1998). The pre-Apulian zone extends to the north up to southwest northwards Albania and up to Zakynthos Isl. on the south (Fig. 3.1). The Ionian zone

Fig. 3.2 Detailed lithostratigraphic column of the Ionian zone, North-Western Greece, from Bourli et al., 2019a.

The Ionian zone consists of three distinct stratigraphic successions sequences: pre-rift, syn-rift and post-rift (Karakitsios, 1995; Zelilidis et al., 2015; Bourli et al., 2019a,b; Bourli et al., 2020; Fig. 3.2): (i) Α pre-rift sequence, represented by the Lower Jurassic “Pantokrator” limestones (over 1,000 m thick) which overlie Upper Triassic “Foustapidima” limestones (50-150 m thick), 94

and Lower to Middle Triassic evaporites at the base (over than 2,000 m thick); (ii) Α syn-rift succession composed of Lower to Upper Jurassic deposits that consist of “Ammonitico Rosso”, “Limestones with filaments”, and “Posidonia beds” (20-200 m thick) at the base, overlain by Lower Jurassic pelagic “Siniais” limestones and “Louros” limestones (20-150 m thick); (iii) A post-rift sequence that starts with the pelagic Vigla Limestones and is composed of Cretaceous to Eocene deposits, is divided into two parts: (a) at the base there are Lower Cretaceous Vigla limestones and Vigla shales (total thickness of 200-600 m); (b) overlain by the Upper Cretaceous Senonian limestones ( 200 to 400 m thickness). The study areas, namely Katakolo and Amvrakikos Gulf, which are included in this thesis are located in the Ionian geotectonic zone (Figs. 3.1, 3). Katakolo (Greek: Κατάκολο) is a seaside town in the municipality of Pyrgos in western Elis, Greece. It is situated on a peninsula on the northwestern coast of Peloponnesus, overlooking the Ionian Sea. The Katakolo Harbour is a popular destination for cruise ships, offering an opportunity for passengers to visit the site of Ancient Olympia. The bay of Katakolo has attracted the interest of scientists due to its extended on- and offshore natural gas seepage (Fig.3.3). It is located in the northwestern Peloponnesus petroliferous basin, in the central Ionian geotectonic zone (Fig. 3.3). It is characterized by active tectonism and it is traversed by active faults at the base of Neogene sequence (Kokinou et al., 2005; Fig. 3.3). Thetectonic regime of the study area is controlled by the diapiric structure of the Triassic evaporites in association with a N-S trending thrust fault close to the western shoreline of the Katakolo peninsula (Kamberis et al., 2000). 95

Fig. 3.3 (a) General map showing the location of Katakolo, one of the study areas and the three external tectonostratigraphic zones of the Hellenides fold-and-thrust belt; (b) Simplified cross section showing the structural evolution of Neogene basins (modified from Nikolaou, 2001), Key: P-PL: Pliopleistocene; Fl: Flysch; J1: Lias; Tr- E: Triassic; E-J: Eocene- Upper Jurasic.

Amvrakikos Gulf (Greek: Αμβρακικός Kόλπος), is located in Northwestern Greece, on the Ionian Greek coast. Amvrakikos is one of the largest enclosed gulfs in Greece (6-15 km wide and 40 km long). Amvrakikos Gulfis characterized as a middle Quaternary E-W graben formed by back-arc extension (Poulos et al., 2008). In the isotopic stages of MIS3 MIS2 (ca. 50 to 11 ka BP; sea level: -55 m compared to present), its eastern part was occupied by a lake (Kapsimalis et al., 2005). Amvrakikos Gulf, which is a seasonal anoxic/hypoxic basin rare Mediterranean fjord, based on its water circulation pattern (Ferentinos et al., 2010; Kountoura and Zacharias, 2011). More details on the local geological settings of the study areas can be found in Chapter 5. Exprimental study, under the respective locations (5.1 Katakolo, 5.2 Amvrakikos Gulf). 96

The Gavrovo zone The Gavrovo zone is a shallow-water carbonate platform from Late Triassic to the Middle-Late Eocene in which no organic matter-rich intervals have been recorded (Karakitsios and Rigakis, 2007; Zelilidis et al., 2015). The upper part of the Gavrovo carbonate sequence consists of Late Cretaceous to Eocene shallow marine carbonates exposed in Varassova andKlokova mountains in south Etoloakarnania as well as in Skolis mountain in northwestPeloponnese (Fleury, 1980). Eocene to earliest Oligocene transitional beds overlie conformably the Ionian Late Eocene pelagic limestones (Fleury, 1980; Kamberis et al., 2000; Sotiropoulos et al., 2003). The detailed geology, stratigraphy, tectonics, and paleogeography of the Gavrovo-Tripoli, Ionian, Pre-Apoulian geotectonic zones, have been studied exhaustively by IFP-IGRS 1966, BP 1971, Monopolis and Bruneton 1982, Karakitsios 1988,1995, Zelilidis et al. 2003, Karakitsios and Rigakis, 2007, Vassilakis et al., 2011; Zelilidis et al., 2015; Pérouse et al. 2017.

References Accordi, G., Federico, C., Pignatti, J. (1998). Depositional history of a Paleogene carbonate ramp (Western Cephalonia, Ionian Islands, Greece). Geol Romana, 34, pp. 131-205. Bourli, N., Maravelis, A.G., Zelilidis, A. (2020). Classification of soft‐sediment deformation in carbonates based on the Lower Cretaceous Vigla Formation, Kastos, Greece. International J Earth Sci, 109, pp. 2,599–2,614. https://doi.org/10.1007/s00531-020-01919-4 Bourli, Ν., Kokkaliari, Μ., Iliopoulos, Ι., Pe-Piper, G., Piper, D.J.W., Maravelis, A.G., Zelilidis, A. (2019a). Mineralogy of siliceous concretions, cretaceous of ionian zone, western Greece: Implication for diagenesis and porosity. Mar Pet Geol, 105 (2019) pp. 45–63. Bourli, N., Pantopoulos, G., Maravelis, A.G., Zoumpoulis, E., Iliopoulos, G., Pomoni-Papaioannou, F., Kostopoulou, S., Zelilidis, A. (2019b). 97

Late cretaceous to early eocene geological history of the eastern Ιonian Βasin, southwestern Greece: A sedimentological approach. Cretac Res, 98, pp. 47-71. https://doi.org/10.1016/j.cretres.2019.01.026 BRITISH PETROLEUM (BP) (1971). The geological results of petroleum exploration in western Greece: IGRS, Athens, 10, pp. 73. Doutsos, T., Piper, G., Boronkay, K., Koukouvelas, I. (1993). Kinematics of the Central Hellenides. Tectonics, 12, pp. 936–953. Ferentinos, G., Papatheodorou, G., Geraga, M., Iatrou, M., Fakiris, E., Christodoulou, D., Dimitriou, E., Koutsikopoulos, C. (2010). Fjord water circulation patterns and dysoxic/anoxic conditions in a Mediterranean semi-enclosed embayment in the Amvrakikos Gulf, Greece. Estuar Coast Shelf Sci, 88(4), pp. 473-481. https://doi.org/10.1016/j.ecss.2010.05.006 Finetti, I. (1982). Structure, stratigraphy and evolution of Central Mediterranean. Boll Geof Teor, Appl., 24, pp. 247-426. Fleury, J.J. (1980). Evolution d’une platforme et d’un basin dans leur cadre alpin: les zones de Gavrovo-Tripolitze et du Pinde-Olonos, Soc Géol du Nord, Special Publication 4, 651. Getsos, K., Pomoni-Papaioannou, F., Zelilidis. A. (2007). A carbonate ramp evolution in the transition from the Apulia platform to the Ionian basin during early to late Cretaceous (NW Greece). Bulletin of the Geological Society of Greece, XXXX, pp. 53-63. Proceedings of the 11th International Congress, Athens, May, 2007. IFP-IGRS (1966). Etude géologique de l' Epire (Grece nord- occidentale). Editions Technip, Paris, France, pp. 306. Kamberis, E., Rigakis, N., Tsaila-Monopolis, S., Ioakim, C., Sotiropoulos, S. (2000). Shallow biogenic gas-accumulations in late Cenozoic sands of Katakolon Peninsula, western Greece. Geological Society of Greece, Special Publication 9, pp. 121–138. Kapsimalis, V., Pavlakis, P., Poulos, S.E., Alexandri, S., Tziavos, C., Sioulas, A., Filippas, D., Lykousis, V. (2005). Internal structure and evolution of the late Quaternary sequence in a shallow embayment: the Amvrakikos Gulf, NW Greece. Mar Geol, pp. 399-418. 98

Karakitsios, V. (1995). The influence of pre-existing structure and halokinesis on organic matter preservation and thrust system evolution in the Ionian basin, Nortwestern Greece. Am Assoc Pet Geol Bull, 79, pp. 960–980. Karakitsios, V., (1988). Sur la différenciation de la zone Ionienne en Epire (Grece nord occidentale): Proc.3th Congress, Athens, 1986, Bulletin of the Geological society of Greece, v. XX/2, pp. 181-196. Karakitsios, V., Koletti L. (1992). Critical revision of the age of the basal Vigla limestones (Ionian zone-Western Greece) based on Nannoplankton. Paleogeographic consequences. Conference: 4th International Nannoplankton Association Conference proceedings, Prague 8-14 September 1991. Nannoplankton Research. HAMRSMID B. & YOUNG J.R. (Eds). Issued by the Moravian Oil Company, Hodonin At: Prague Volume: Knihovnicka ZPN 14a, 1992, Vol.1, pp. 165-177. Karakitsios, V., Pomoni-Papaioannou, F. (1998). Sedimentological study of the Triassic solution-collapse Breccias of the Ionian zone (NW Greece). Carbonates Evaporites, 13, pp. 207–218. https://doi.org/10.1007/BF03176594 Karakitsios, V., Rigakis, N., 2007. Evolution and petroleum potential of Western Greece. J Pet Geol, 30, pp. 197–218. Kokinou, E., Kamberis, E., Vafidis, A., Monopolis, D., Ananiadis, G., Zelilidis, A. (2005). Deep seismic reflection data from offshore western Greece: a new crustal model for the Ionian Sea. J Pet Geol, 28, pp. 185–202. Kountoura, K., Zacharias, I. (2011). Temporal and spatial distribution of hypoxic/seasonal anoxic zone in Amvrakikos Gulf, Western Greece. Estuar Coast Shelf Sci, 94 (2), pp. 123-128. Kreemer, C., Chamot-Rooke, N. (2004). Contemporary kinematics of the southern Aegean and the Mediterranean Ridge. Geophys J Int, 157, pp. 1377–1392. https://doi.org/10.1111/j.1365-246X.2004.02270.x Monopolis, D., Bruneton, A. (1982). Ionian Sea (Western Greece): Its structural outline deduced from drilling and geophysical data. Tectonophysics, 83(3-4), pp. 227-242. 99

Nikolaou, Κ. (2001). Origin and migration mechanism of the main hydrocarbon seeps in western Greece. Bulletin of the Geological Society of Greece, 34(3), pp. 1213-1219. Papanikolaou, D. (2009). Timing of tectonic emplacement of the ophiolites and terrane paleogeography in the Hellenides. Lithos, 108, pp. 262–280. Pérouse, E., Sébrier, M., Braucher, R. et al. (2017). Transition from collision to subduction in Western Greece: the Katouna–Stamna active fault system and regional kinematics. Int J Earth Sci (Geol Rundsch) 106, pp. 967–989. https://doi.org/10.1007/s00531-016-1345-9 Poulos, S.E., Kapsimalis, V., Tziavos, C., Paramana, T. (2008). Origin and distribution of surface sediments and human impacts on recent sedimentary processes. The case of the Amvrakikos Gulf (NE Ionian Sea). Cont Shelf Res, 28(20), pp. 2736-2745. https://doi.org/10.1016/j.csr.2008.09.014 Sotiropoulos S., Kamberis E., Triantaphyllou M., Doutsos T. (2003). Thrust sequences in the central part of the External Hellenides. Geol Mag, 140(6), pp. 661-668. Vassilakis, E., Royden, L., Papanikolaou, D. (2011). Kinematic links between subduction along the Hellenic trench and extension in the Gulf of Corinth, Greece: A multidisciplinary analysis. Earth Planet Sci Lett, 303(1–2), pp. 108-120. https://doi.org/10.1016/j.epsl.2010.12.054 Zelilidis, A., Maravelis, A., Tserolas, P., Konstantopoulos, P.A. (2015). An overview of the petroleum systems in the Ionian Zone, onshore NW Greece and Albania. J Pet Geol, 38, pp. 331-348. https://doi.org/10.1111/jpg.12614 Zelilidis, A., Piper, D.J.W., Vakalas, I., Avramidis, P., Getsos, K. (2003). Oil and gas plays in Albania: Do equivalent plays exist in Greece? J Pet Geol, 26(1), pp. 29-48.

100

5. Material and Methods

5.1 Conseptual methodological scheme and survey design For the purpose of this thesis, a complete spectrum of state-of-the-art equipment for geophysical and geochemical research of on land and offshore fluid seepage has been used. Moreover, a long-term benthic observatory (seafloor lander module) and short-term monitoring techniques were also used, providing a spatio-temporal dimension in the survey. The survey in the study areas was organized into two phases. During the first phase, a systematic survey of the seafloor in Katakolo Bay and Amvrakikos Gulf (Amphilochia Bay and Sogono) using echosounders, side scan sonar and subbottom profilers was carried out. The marine remote sensing survey led to the detection of shallow gas accumulation in the sediments’ interstitials, gas bubbles in the water column, and gas-related seafloor morphological features.

Fig. 4.1.1 Conseptual model of the methodological scheme. (SSS: side scan sonar; MBES: multibeam echosounder; SBP: subbottom profiler; CC: closed chamber; GMM: 101

gas monitoring module; MEDUSA: Module for Environmental Deep-Underwater-Sea Analysis).

The second phase of the survey consisted of seafloor ground truthing planned on the results of the first phase. Seafloor visual inspection, seafloor gas sampling, short- and longterm monitoring of gas seepage were carried out on specific sites selected based on the interpretation of marine remote sensing data. Fig. 4.1.1 shows the conceptual model of the methodology used in the thesis. Fig. 4.1.2 exhibits the survey design and Table 4.1.1 presents the equipment and systems used in each survey area.

Fig. 4.1.2 Survey design flowchart and 3d representation. (SSS: side scan sonar; MBES: multibeam echosounder; ES: echosounder; SBP: subbottom profiler; CC: closed chamber; GMM: gas monitoring module; MEDUSA: Module for Environmental Deep- Underwater-Sea Analysis).

102

Table 4.1.1 Equipment and systems used in each survey area of this thesis. Equipment/ AMVRAKIKOS system KATAKOLO AMFILOCHIA SOGONO Remote Single X sensing beam MARINE Multi X GEOPHYSICS SBP X X SSS X X X Geochemical Gas bottles X X sampling (divers) Visual MEDUSA- X X X inspection Camera Ground Short-term gas MEDUSA X X X truthing measurements platform Long term GMM X monitoring

5.2 Geophysical means

5.2.1 Sonar mapping systems - acoustic remote sensing Acoustic waves have the ability to travel over long distances underwater without attenuating as much as electromagnetic waves (i.e., light). Thus, hydroacoustic methods are the most effective, accurate and often the only means for seabed exploration (Blondel, 2009). The sonar mapping systems can be divided into three categories: single-beam echo-sounders, multibeam echo-sounders, and side scan sonars. All three (separately and combined) have been extensively used in seafloor fluid flow surveys the last 30-40 years (e.g., Greinert et al., 2010). Detection of bubbles rising from the seafloor into the water column (bubble plumes) by employing an echo-sounder, is the principal remotely acquired proof of active seafloor gas seepage acoustic seafloor backscatter analysis,and can be used as a proxy for submarine 103

seepage activity (Naudts et al., 2008). Hydroacoustical gas bubble plume detection is based on the significant contrast in acoustic impedance between water and gas bubbles (Blomberg et al., 2018). The high backscatter hydroacoustic signals in the ecographs’ water column caused by bubble streams are called “flares” (e.g., Egorov et al., 1998; Greinert et al., 2006; Greinert et al., 2010; Nikolovska et al., 2008). Visual observation is required to give exact location (few meter scale) of the actual gas seep on seafloor, since the footprint of the acoustic lobe increases with depth (Greinert et al., 2010) and when bubbles plumes are moved by currents (Schneider von Deimling et al., 2010). The detection of gas flares, especially in “noisy” hydroacoustic data may be elusive, since, for example the gas in the swim bladders of fish shows a similar acoustic target. Visual inspection may shed light on this issue, but what is more, when surveying for submarine gas seeps, it is easier to investigate the seabed for seepage induced geomorphological features, such as pockmarks, mud volcanoes, etc. (Judd and Hovland, 1992) than the seeps per se (Judd, 2004). Of particular importance, is the combination of a hydro-acoustic and a seismic survey; the only way to detect gas-charged sediments and get a clearer overview of seabed (and sub-seabed) features on a large scale. Finally, hydroacoustic methods are particularly suitable for exploring the occurrence and specific characteristics of bubbles and bubble plumes released from cold seeps without causing perturbations (Greinert and Nützel, 2004).

5.2.2 Single-beam echo-sounders Single-beam eco-sounders are simpler to use than the other two instruments and nowadays theyare an integral part of every vessel’s equipment. As their name states, they only transmit a single beam 104

oriented toward the ship’s nadir, mapping the seabed only directly below the survey vessel. They generally use low-frequency signals (<20 kHz) transmitted in short pulses (<2 ms) from a single transducer. Single-beam echosounders have been used for almost 30 years in surveys investigating seabed seepage and methane flux (Merewether et al., 1985, Hornafius et al., 1999, Greinert et al., 2006, Artemov et al., 2007, Greinert et al., 2010). Single-beam eco-sounders have been used in multiple researches for detection and localization of sea floor seepage (e.g., Wenau et al., 2020; Geinert et al., 2010), the study of bubble characteristics and gas flow rates (e.g., Higgs et al., 2019), gas flux estimation (Leifer et al., 2017).

5.2.3 Multibeam echo-sounders The multibeam echo-sounders (MBES) transmit several beams (up to 120), covering swath on each side of the research vessel’s track up to 20 times the water depth (Blondel, 2009). MBES systems acquire bathymetry measurements for each beam, producing an incomparablymore detailed outcome than single-beam echo-sounders. MBES have been extensively used for seabed fluid flow surveying, such as detection and localization of bubble plumes/gas flares (Chen et al., 2020; Idczak et al., 2020) and gas flux estimation (Leifer et al., 2017). MBES have been vastly used for the detection, mapping and study of fluid seepage related morphological expressions (Judd and Hovland, 1992), such as pockmarks (Davy et al., 2010; Idczak et al., 2020; Coskun et al., 2016), mud volcanoes (Isola et al., 2020; Chen et al., 2020), methane hydrate mounds (Liu, 2017; Greinert et al., 2010), and carbonate mounds (Loncke et al., 2004). 105

5.2.4 Side scan sonars The discovery of the side scan sonar in the mid-60s revolutionized marine research and one of its first accomplishments was the discovery of the first pockmark (King and McLean, 1970). The side scan sonar may cover an impressively large part of the seabed ranging from a few meters to 60 km. The coverage of the seabed is accomplished by the transmission of one beam on each side of side scan sonar tow-fish. Side scan sonar frequencies mayrange from 6.5 kHz to 1 MHz, reaching resolutions from 60 m up to 1 cm (Blondel, 2009). Thus, the side scan sonar provides satellite-like, high- quality images of the seabed that would be impossible to obtain with other means, due to the interference of the water. As a result, the side scan sonar is one of the most important instruments for high-resolution mapping of the seabed and concomitantly for surveying seabed fluid seepage (Etiope et al., 2013; Naudts et al., 2008). The side scan sonar and subbottom profiling systems (see Subchapters 5.1.1, and 5.2.1-2) besides mapping seabed geomorphology and stratigraphy, they have been successfully used for gas bubble detection via flare imaging in the water column from various depths (De Beukelaer et al., 2003; Klaucke et al., 2005, 2006; Greinert et al., 2006; Géli et al., 2008). Evidence of active or past fluid seepage may be inferred from increased side scan sonar and/or multi-beam backscatter response (Klaucke et al., 2010; Greinert et al., 2010), which may correspond to a seafloor cementation area, combined with negative (pockmarcks) or positive (mud volcanoes, carbonate mounds) morphological expressions. Side scan sonars may be used for gas flare-imaging surveys, where higher backscatter signals are detected in the water column (Etiope et al., 2013; Greinert et al., 2010). 106

5.2.5 Hydroacoustical gas bubble plume detection with the use of side scan sonar. An example from Katakolo Bay An application of the abovementioned hydroacoustic methods, took place in Katakolo Bay (one of the study sites included in this thesis; Christodoulou, 2010; Etiope et al., 2013). Katakolo is characterised by intense and extended natural gas seepage both on and offshore (see Chapter 5.1). Geophysical survey (side scan sonar and subbottom profiler) data were successfully used for bubble detection via flare imaging, demonstrating gas release in the water column from various depths (Fig. 4.2.5.1).

Fig. 4.2.5.1 (a) 3.5 kHz seismic profiles from Katakolo Harbour showing gas flares. (b) Uncorrected side scan sonar records showing the three types of gas flares recorded in Katakolo Harbour (I: individual flares, II: gas flare clusters, III: gas clouds). (c) Map of Katakolo Harbor presenting the location of gas flares recorded from side scan sonar data and the seepage areas as extracted from the above-mentioned data. Figure from Etiope et al., 2013.

The gain of the output and input signals of the subbottom profiling instruments was increased in order to display clearly the return from the gas bubbles in the water column rather than the underlying sediment stratigraphy (Fig. 4.2.5.1). For bubble plume detection via side scan sonar, sonographs with enhanced gain including the column (uncorrected data) were used (Fig. 4.2.5.1-2). High backscatter 107

anomalies from within the water column could represent either gas bubbles (depending on the bubble resonance frequency which increases with depth) or fish schools. For this reason, the presence and location of the bubble plumes detected in side scan sonar records was confirmed with underwater camera inspection (included in the MEDUSA system, see Subchapters 5.1.1 and 5.2.1-2). The 100 kHz side scan sonar that was used in this study, detects bubbles >0.006 cm diameter in 10-20 m water depth, while 3.5 kHz subbottom profiler detects bubbles > 0.10 cm diameter (Guinasso and Schink, 1973). Moreover, the side scan sonar was considered advantageous compared to the subbotom profiler, since it also maps the spatial distribution of the bubbles. Based on visual observations, the bubbles offshore Katakolo Bay have ~0.5–5 cm diameter, thus all were resonant.

Fig. 4.2.5.2 Signal of bubble plumes (acoustic gas flares) in the water column from Katakolo Bay.

The number of the total gas bubble plumes that reached the sea surface (observed visually and detected in side scan records; Fig. 4.2.5.1) in combination with gas flux measurements performed on some of them, resulted in the estimation of the total methane emission 108

from the seafloor to the atmosphere from the offshore Katakolo Bay seepage area. Inside KatakoloHarbour, a 47,500 m2wide area, with high density of bubble streams was recorded radially around the main dock (Fig. 4.2.5.1c). Moreover,4 areas of ~2000–12,000 m2 and many smallerareas with bubble streams were recorded in the Harbour. More than 523 plumes were hydroacoustically detected over 69,200 m2 in total. Based on the calculations of the density of plumes detected from uncorrected side scan sonar records, the total number of plumes in Katakolo Harbour was estimated at 2,700. The detailed methodology of all the extrapolations as well as the results of the survey is described in detail in Etiope et al., 2013.

5.2.6 Subbottom Profilers The subbottom profiler (SBP) is the most effective tool for the identification of the physical properties and imaging of the surface and subbottom layers of the seabed. The SBPs have been widely employed in marine research, since it is a rapid method to collect subbottom data from large areas in a non-intrusive manner. Subbottom profilers usually consist of “transducers” which send sound pulses vertically downwards into the shallow sub-sea floor sediments. The sub-surface layer interfaces reflect the sound pulses, which are then received and measured by the subbottom profiler’s “receivers” (time measurements). Differences in the sound velocity between the two layers are interpreted as differences in acoustic impedance, and thus different densities. Depending on the difference in the impedance, part of the acoustic energy will be reflected. The reflected acoustic energy is analogous to the impedance contrast (Mondol, 2010). Possible candidates causing strong impedance contrast are changes in lithology, porosity, occurrence of fluids and their degree of saturation and diagenesis in the rocks and sediments (Mondol, 2010). Fluids have significantly different acoustic impedance from sediments and a considerable amount of sound energy will be 109

reflected from the boundary between e.g., a “gas front” (gassy sediments) and sediments free of gas (Tóth et al., 2014). “Pinger” and “Chirp” SBPs are the most often used systems for shallow gas accumulation surveys. There are several different types of shallow SBP systems based on sound sources and frequencies that result in different resolutions and penetrations. The 'Pinger' has a single-frequency transducer that ranges from 3.5 to 7 kHz, and may penetrate few meters up to 50 m, with vertical resolution ~0.2 m. The “Chirp” is a more novel technology whichtransmitsa sweeping range of frequencies. The transducer initially vibrates at a low frequency which, during the pulse, is modulated to a high frequency, producing a chirping sound that resembles a bird. The “Chirp” operated on a range of frequencies between 3 to 117 kHz, and has vertical resolution ~0.05 m and penetration <100 m. As in this thesis, SBP systems have extensively been used for exploration of shallow gas accumulations and their features, over the last 50 years (Judd and Hovland, 2007). Shallow fluid/gas accumulation features such as acoustic turbidity, gas plumes, acoustic blanking, acoustic voids, enhanced reflections (flags or bright spots), acoustic chimneys or fluid pipes, columnar disturbances (Papatheodorou et al., 1993; Judd and Hovland, 1992, Schroot et al., 2005; Gay et al., 2006) have been detected with chirp (Coskun et al., 2016; Dondurur et al., 2011), high resolution sparker (Dewangan et al., 2010), pinger (Papatheodorou et al., 1993), and more than one seismic sources operating simultaneously, e.g., pinger, chirp, and boomer (Félix and De Mahiques, 2013). Accordingly, free gas and gas hydrate associated geomorphological features such as mud diapirs (Lüdmann and Wong, 2003), mud volcanoes (Isola et al., 2020; Vogt et al., 1997; Sager et al., 1999; Loncke and Mascle, 2004), active and inactive pockmarks (Dondurur et., al., 2011; Naudts et al., 2008; Gay et al., 2006) have been discovered and surveyed with SBP systems. Plenty of examples 110

of gas charged sediments (Subchapters 5.2.1-2), as well as gas- induced morphological expressions detected with subbotom profiles are presented in this thesis (Chapter 5.2.1). Subbottom profiling survey is necessary rather than complementary to a hydroacoustic survey, since it may help discover and or validate the existence of structures associated with fluid seepage, such as pockmarks, paleo-pockmarks, mud volcanoes, dormant mud volcanoes, diapirs, carbonate or gas hydrate mounds, and may reveal the mechanisms of their formation (e.g., Dewangan et al., 2010). Moreover, as in the hydro-acoustic data, gas flares are recorded as acoustic noise in the water column on the seismic data (Figs. 4.2.5.1a, 4.2.5.2), which could be confirmed also in side scan sonar and MBES data (Naudts et al., 2008). Modern SBPs produce very high-resolution seismic data, which in combination with multibeam backscatter and bathymetry data, as well as side scan sonar echograms, they complete an integrated survey which can give valuable outcomes for fluid seepage exploration.

5.2.7 Visual seafloor observation Visual seafloor observations are crucial for seabed fluid flow surveys, since they complete the puzzle of the investigation of seepage affected marine area, and undoubtedly confirm the finds stemming from the hydro-acoustic and seismic data. Visual seafloor surveys are carried out by using AUVs (autonomous underwater vehicles; Loher et al., 2018), ROVs (remotely operated vehicle; Loher et al., 2018; Römer et al., 2014; Naudts et al., 2008) or simple drop cameras and diver operated cameras (Etiope et al., 2013; Naudts et al., 2008). The candidate diving sites for either instruments or humans can only be derived from the geophysical survey data (see also Subchapters 5.1.1, 5.1.3, 5.2.1). 111

5.3 Multiparametric instruments developed for offshore seepage detection Detection and flux assessment of gas seepage offshore is much more complicated than onshore. Initially it was performed with custom-made “sniffers” for hydrocarbon detection seepage near the seabed or by sediment sampling and analysis, which was a more complicated and time-consuming procedure. The need for multi-parametric surveying and monitoring of the occurrence of hydrocarbons in the seawater, led to the development of light multi-parametric platforms that can be handled by relatively small research vessels, by INGV (Istituto Nazionale di Geofisica e Vulcanologia) and Tecnomare SpAfor the detection of hydrocarbon seepage and other parameters. The multi-parametric platforms are based on the use of new generation of solid-state methane sensors, combined with traditional oceanographic instruments, controlled by a single data acquisitionsystem and time reference, and implemented in benthic stations for long-term monitoring ormodules for spatial surveys. These platforms are useful for investigation of marine areas for gas seepage and may provide valuable information on suitable sites for gas sampling for laboratory analysis to then determinehydrocarbon origin by its molecular and isotopic composition. In this thesis two multi-parametric platforms were used: • the GMM (Gas Monitoring Module; Fig. 4.3.1), for long-term monitoring at the seabed in correspondence with gas seepage, (first mission in the Patras Gulf, Greece in 2004; Marinaro et al., 2011; Christodoulou, 2010) equipped with methane sensors,

CTD, turbidity meter, H2S sensors, oxygen sensor, current meter. • MEDUSA (Module for Environmental Deep-Underwater-Sea Analysis; Fig. 4.3.1), a towed system (first mission in Marmara 112

Sea in 2009; Marinaro et al., 2011) for short-term aerial surveys.

Fig. 4.3.1 The MEDUSA platform (left) and the GMM benthic observatory (right) (Kordella et al., 2012).

5.3.1 GMM The long-term monitoring of gas flux variability on the seabed is a difficult task due to technological and environmental limitations, such as short-term stability, limited operational depth, and time autonomy of gas sensors. Nevertheless, long-term monitoring is essential for understanding the links between seabed fluid flows and the geodynamic (e.g., seismicity), oceanographic (e.g., tides, currents), and atmospheric (e.g., wind, barometric pressure) factors. GMM (Fig. 4.3.1) is a light observatory that was specifically designed for long-term gas monitoring on the seafloor (Marinaro et al., 2006). It was used in Katakolo Harbour (2010) in the framework of HYPOX project, and the results of this survey are a part of this thesis (more in Chapter 4.1.3). More details on the configuration and technical details of the GMM as it was used in Katakolo Harbour are presented in Chapter 5.1.3.

5.3.2 MEDUSA The MEDUSA (Fig. 4.3.1) is a cabled underwater module with a series of instruments, for multi-parametricdown-casts and towed surveys in the 113

water column. It operates through a Surface Control Unit and an umbilical cable providing AC power and Extended Ethernet telemetry to the module (Marinaro et al., 2011). It was used for the first time in an explorative survey in Marmara Sea in 2009. Finally, it was used for aerial surveys in Katakolo Harbour and Amvrakikos Gulf in the framework of the HYPOX project, and the results are presented in this thesis (more in Subchapters 5.1.1, 5.2.1-2.). More technical details on the MEDUSA platform can be found in the same Subchapters 5.1.1, 5.2.1.

5.4 Geochemical means

5.4.1 Gas Detection Methods on land Natural gas flux measurements from the ground to the atmosphere are of crucial importance for (i) the detection of active seepage, (ii) indication of intensity of seepage.For gas flux measurements on land usually the “closed or accumulation chamber” is employed (Etiope, 2015; Fig. 4.4.1.1). With the closed chamber method, the flux is calculated by measuring gas concentration that is accumulated over time inside a chamber that is placed tightly on the ground (Fig. 4.4.1.1). If the gas emission is steady (i.e., the rate of variation of gas concentration is constant) the linear regression can be used in order to calculate the slope of concentration versus time. The slope of the line indicates the gas flux (Etiope, 2015). The flux is generally expressed in units of mg m−2 day−1. Gas can be collected in vials to measure methane concentrations in the lab or on site with portable sensors. The closed chamber method is suitable for spatial gas flux surveys, and for the assessment of methane escaping from the lithosphere into the atmosphere in gas seepage sites (e.g., Kordella et al., 2020). Between 114

consecutive measurements, the chamber and sensors are “cleaned” with atmospheric air (right photo in Fig. 4.4.1.1).

Fig. 4.4.1.1 Gas flux measurement with the closed-chamber method with portable gas sensors in Faros, Katakolo (see Chapter 5.1.1). Photos: Margarita Iatrou.

5.4.2 Dissolved gas in seawater and underwater sediment sampling and analysis Seawater samples can be collected in glass bottles tightly sealed with hydrocarbon-free septa and closed with aluminium caps (Fig. 4.4.2.1).

Through the septa a microbicide (e.g., mercuric chloride, HgCl2) is typically injected in the sample, in order to prevent methane oxidation. Dissolved gases can then be extracted either on site or in the laboratory (e.g., Capasso and Inguaggiato, 1998). Marine sediment samples from submarine seeps can be collected by corers from the research vessel or divers (Etiope, 2015). The samples must be immediatelly handled and stored in a portable fridge (Fig. 4.4.2.1), since gases are volatile, in non-coated metal cans or clear plastic jars for consecutive laboratory analyses. To prevent hydrocarbon oxidation a microbicide should be added before closing the cans/jars. 115

Fig. 4.4.2.1 Sampling, handling and storage of seawater and marine sediments from a pockmark group in Amphilochia Bay (see Chapter 5.2.1). Photos: Maria Geraga.

References Artemov, Y.G., Egorov, V.N., Polikarpov, G.G., Gulin, S.B. (2007). Methane emission to the hydro- and atmosphere by gas bubble streams in the Dnieper paleo-delta, the Black Sea. Mar Ecol J, 6(3), pp. 5–26. Blomberg, A.E.A., Weber, T.C., Austeng, A. (2018). Improved visualization of hydroacoustic plumes using the split-beam aperture coherence. Sensors, 18(7), 2033. https://doi.org/10.3390/s18072033 Blondel, P. (2009). The handbook of side scan sonar. Springer. Praxis Publishing Ltd, Chichester, UK, pp. 316. Capasso, G., Inguaggiato, S. (1998). A simple method for the determination of dissolved gases in natural waters. An application to thermal waters from Vulcano Island. Appl Geochem, 13, pp. 631–642. 116

Chen, N., Yang, T.F., Hong, W. et al. (2020). Discharge of deeply rooted fluids from submarine mud volcanism in the Taiwan accretionary prism. Sci Rep, 10, 381. https://doi.org/10.1038/s41598-019-57250-9 Christodoulou, D., 2010. Geophysical, sedimentological study: remote sensing on pockmarks in seismogenic active areas. PhD Thesis, University of Patras, Patras, 296 pp. Coskun, S., Dondurur, D., Cifci, G., Aydemir, A., Drahor, M.G. (2016). Natural and anthropogenic submarine morphologies revealed by high resolution acoustic data in the Gulf of Izmir, western Turkey. Mar Pet Geol, 71, pp. 211–224. https://doi.org/10.1016/j.marpetgeo.2015.12.012 Davy, B., Pecher, I., Wood, R., Carter, L., Gohl, K. (2010). Gas escape features off New Zealand: Evidence of massive release of methane from hydrates, Geophys Res Lett, 37, L21309. https://doi.org/10.1029/2010GL045184 De Beukelaer, S.M., MacDonald, I.R., Guinnasso, N.L. et al. Distinct side-scan sonar, RADARSAT SAR, and acoustic profiler signatures of gas and oil seeps on the Gulf of Mexico slope. Geo-Mar Lett, 23, 177– 186 (2003). https://doi.org/10.1007/s00367-003-0139-9 Dewangan, P., Ramprasad, T., Ramana, M.V., Mazumdar, A.., Desa, M., Badesab, F.K. (2010). Seabed morphology and gas venting features in the continental slope region of Krishna–Godavari basin, Bay of Bengal: Implications in gas-hydrate exploration. Mar Pet Geol, 27(7), pp. 1628-1641, https://doi.org/10.1016/j.marpetgeo.2010.03.015 Dondurur, D., Cifci, G., Drahor, M.G., Coskun, S. Acoustic evidence of shallow gas accumulations and active pockmarks in the Izmir Gulf, Aegean Sea. Mar Pet Geol, 28 (2011), pp. 1505-1516, https://doi.org/10.1016/j.marpetgeo.2011.05.001 Egorov, V., Luth, U., Luth, C., Gulin, M.B. (1998). Gas seeps in the submarine Dnieper Canyon, Black Sea: Acoustic, video and trawl data, in MEGASEEPS Gas Explorations in the Black Sea, Project Report, edited by U. Luth, C. Luth, and H. Thiel, pp. 11 – 21, Univ. of Hamburg, Hamburg, Germany. Etiope, G., Christodoulou, D., Kordella, S., Marinaro, G., 117

Papatheodorou, G. (2013a). Offshore and onshore seepage of thermogenic gas at Katakolo Bay (Western Greece). Chem Geol, 339, pp. 115–126. https://doi.org/10.1016/j.chemgeo.2012.08.011 Félix, C.A., de Mahiques, M.M. (2013). Late quaternary evolution and shallow gas formation in a tropical estuarine environment: the case of the Bertioga channel, Brazil. Ieee/Oes Acoustics in Underwater Geosciences Symposium, Rio de Janeiro, 2013, pp. 1-8. https://doi.org/10.1109/RIOAcoustics.2013.6684003 Gay, A., Lopez, M., Ondreas, H., Charlou, J.-L., Sermondadaz, G., Cochonat, P. (2006). Seafloor facies related to upward methane flux within a Giant Pockmark of the Lower Congo Basin. Mar Geol, 226(1– 2), pp. 81-95, https://doi.org/10.1016/j.margeo.2005.09.011 Greinert, J., Artemov, Y., Egorov, V., De Batist, M., McGinnis, D. (2006). 1300-m-high rising bubbles from mud volcanoes at 2080 m in the Black Sea: Hydroacoustic characteristics and temporal variability, Earth Planet Sci Lett, 244, pp. 1 – 15. https://doi.org/10.1016/j.epsl.2006.02.011 Greinert, J., Bialas, J., Lewis, K., Suess, E. (2010). Methane seeps at the Hikurangi Margin, New Zealand. Mar Geol, 272 (1–4), pp. 1-3. https://doi.org/10.1016/j.margeo.2010.02.018 Greinert, J., Nützel, B. (2004), Hydroacoustic experiments to establish a method for the determination of methane bubble fluxes at cold seeps. Geo Mar Lett, 24, pp. 75–85. https://doi.org/10.1007/s00367‐003‐0165‐7 Guinasso, N.L., Jr., Schink, D.R. (1973). A simple physicochemical acoustic model of methane bubbles rising in the sea, Tech. Rep 73-15- T, Texas A&M University. Higgs, B., Mountjoy, J.J., Crutchley, G.J. Townend, J., Ladroit, Y., Greinert, J., McGovern, C. (2019). Seep-bubble characteristics and gas flow rates from a shallow-water, high-density seep field on the shelf-to- slope transition of the Hikurangi subduction margin. Mar Geol, 417, 105985. https://doi.org/10.1016/J.MARGEO.2019.105985 118

Hornafius, J.S., Quigley, D., Luyendyk, B.P. (1999). The world's most spectacular marine hydrocarbon seeps (Coal Oil Point, Santa Barbara Channel, California): quantification of emissions. J Geophys Res Oceans Atmospheres, 20 (C9), 20, pp. 20,703–20,711. Idczak, J., Brodecka-Goluch, A., Łukawska-Matuszewska, K., Graca, B., Gorska, N., Klusek, Z., Pezacki, P.D., Bolałek, J. (2020). A geophysical, geochemical and microbiological study of a newly discovered pockmark with active gas seepage and submarine groundwater discharge (MET1-BH, central Gulf of Gdańsk, southern Baltic Sea). Sci Total Environ, 742, 140306. https://doi.org/10.1016/j.scitotenv.2020.140306 insonification: Application to methane bubble flux from seep areas in the outer Laptev Sea. The 250 Cryosphere, 11, pp. 1333-1350. https://doi.org/10.5194/tc-11-1333-2017 Isola, J.I., Ormazabal, J.P, Flores, G., Arismendi, S., Druet, M., Muñoz, A., del Río, J.L., Etienot, S.D.A., Gomez Ballesteros, M.P., Principi, S., Bolatti, N.D., Tassone, A.A. (2020). Tectonic and structural controls on Neogene fluid release in the Patagonian Continental Margin. Mar Pet Geol, 115, 104246. https://doi.org/10.1016/j.marpetgeo.2020.104246 Judd, A. and Hovland, M. (2007). Seabed Fluid Flow. The impact on Geology, Biology and the Marine Environment, Cambridge University Press, Cambridge CB2 2RU, UK, pp. 475. Judd, A.G. (2004). Natural seabed gas seeps as sources of atmospheric methane. Env Geol, 46, pp. 988–996. https://doi.org/10.1007/s00254-004-1083-3 Judd, A.G., Hovland, M. (1992). The evidence of shallow gas in marine sediments. Continent Shelf Res, 12, pp. 1081-1095. Judd, A.G., Hovland, M. (1992). The evidence of shallow gas in marine sediments. Continent Shelf Res, 12, pp. 1081-1095. King, L. H., McLean, B. (1970). Pockmarks on the Scotian shelf. Geol Soc Am, 81, pp. 3141-8. 119

Klaucke, I., Sahling, H., Bűrk, D., Weinrebe, W., Bohrmann, G. (2005). Mapping deepwater gas emissions with sidescan sonar. Eos, 86, pp. 341–343. Klaucke, I., Weinrebe, W., Joerg Petersen, C., Bowden, D. (2010). Temporal variability of gas seeps offshore New Zealand: Multi- frequency geoacoustic imaging of the Wairarapa area, Hikurangi margin. Mar Geol, 272(1–4), pp. 49-58, https://doi.org/10.1016/j.margeo.2009.02.009 Kordella, S., Etiope, G., Christodoulou, D., Papatheodorou, G., Marinaro, G., Lo Bue, N., Geraga, M., Ferentinos, G. (2012). Underwater gas seepage detection and monitoring in Katakolo Bay (Western Greece) – preliminary results. In Proceedings of 10th Panhellenic Symposium of Oceanography and Fisheries, 7-11 May, Athens, in CD format. Kordella, S., Ciotoli, G., Dimas, X., Papatheodorou, G., Etiope, G. (2020). Increased methane emission from natural gas seepage at Katakolo Harbour (Western Greece). Appl Geochem, 116, 104578. https://doi.org/10.1016/j.apgeochem.2020.104578 Leifer, I., Boles, J.R., Luyendyk, B.P., Clark, J.F. (2004). Transient discharges from marine hydrocarbon seeps: Spatial and temporal variability. Environ Geol, 46 (8), pp. 1,038-1,052. Leifer, I., Chernykh, D., Shakhova, N., Semiletov, I. (2017). Sonar gas flux estimation by bubble insonification: Application to methane bubble flux from seep areas in the outer Laptev Sea. Cryosphere, 11, pp. 1,333–1,350. Liu, L., Lu, X., Zhang, X., et al. (2017). Numerical simulations for analyzing deformation characteristics of hydrate-bearing sediments during depressurization. Adv Geo-Energ Res, 1(3) pp. 135-147, https://doi.org/10.26804/ager.2017.03.01 Loher, M., Pape, T., Marcon, Y., et al. (2018). Mud extrusion and ring- fault gas seepage – upward branching fluid discharge at a deep-sea mud volcano. Sci Rep, 8, 6275. https://doi.org/10.1038/s41598-018- 24689-1 120

Loncke, L., Mascle, J., (2004). Mud volcanoes, gas chimneys, pockmarks and mounds in the Nile deep-sea fan (Eastern Mediterranean): geophysical evidences. Mar Pet Geol, 21(6), pp. 669- 689. https://doi.org/10.1016/j.marpetgeo.2004.02.004 Loncke, L., Mascle, J., (2004). Mud volcanoes, gas chimneys, pockmarks and mounds in the Nile deep-sea fan (Eastern Mediterranean): geophysical evidences. Mar Pet Geol, 21(6), pp. 669- 689. https://doi.org/10.1016/j.marpetgeo.2004.02.004 Lüdmann, T., Wong, H.K. (2003). Characteristics of gas hydrate occurrences associated with mud diapirism and gas escape structures in the northwestern Sea of Okhotsk. Mar Geol, 201(4), pp. 269-286. https://doi.org/10.1016/S0025-3227(03)00224-X Marinaro, G., Etiope, G., Gasparoni, F. Furlan F., Bruni, F. (2011). Gas seepage detection and monitoring at seafloor.10th Offshore Mediterranean Conference and Exhibition, Ravenna, Italy, March 23-25, 2011, “Frontier exploration” session. Marinaro, G., Etiope, G., LoBue, N., Favali, P., Papatheodorou, G., Christodoulou, D., Furlan, F., Gasparoni, F., Ferentinos, G., Masson, M., Rolin, J.F. (2006). Monitoring of a methane-seeping pockmark by cabled benthic observatory (Patras Gulf, Greece). Geo-Mar Lett, 26, pp. 297–302. https://doi.org/10.1007/s00367-006-0040-4 Merewether, R., Olsson, M.S., Lonsdale, P. (1985). Acoustically detected hydrocarbon plumes rising from 2‐km depths in Guaymas Basin, Gulf of California. J Geophys Res, 90(NB4), pp. 3075–3085. https://doi.org/10.1029/JB090iB04p03075 Mondol, N.H. (2010). Seismic Exploration. In: Petroleum Geoscience. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-02332- 3_17 Naudts, L., Greinert, J., Artemov, Y., Beaubien, S.E., Borowski, C., De Batist, M. (2008). Anomalous sea-floor backscatter patterns in methane venting areas, Dnepr paleo-delta, NW Black Sea. Mar Geol, 251(3–4), pp. 253-267. https://doi.org/10.1016/j.margeo.2008.03.002 121

Nikolovska, A., Sahling, H., Bohrmann, G. (2008). Hydroacoustic methodology for detection, localization, and quantification of gas bubbles rising from the seafloor at gas seeps from the eastern Black Sea, Geochem Geophys Geosyst, 9, Q10010. https://doi.org/10.1029/2008GC002118 Papatheodorou G., Hasiotis T., Ferentinos G. (1993). Gas charged sediments in the Aegean and Ionian Seas, Greece. Mar Geol, 112, pp. 171-184. Römer, M., Torres, M., Kasten, S., et al. (2014). First evidence of widespread active methane seepage in the Southern Ocean, off the sub-Antarctic island of South Georgia. Earth Planet Sci Lett, 403, pp. 166-177. https://doi.org/10.1016/j.epsl.2014.06.036 Sager, W.W., Lee, C.S., MacDonald, I.R., Schroeder, W.W. (1999). High-frequency near-bottom acoustic reflection signatures of hydrocarbon seeps on the northern Gulf of Mexico continental slope. Geo-Mar Lett,18, pp. 267-276. Schneider von Deimling, J., Greinert, J., Chapman, N.R., Rabbel, W., Linke P. (2010). Acoustic imaging of natural gas seepage in the North Sea: sensing bubbles under control of variable currents. Limnol Oceanogr Methods, 8, pp. 155-171. https://doi.org/10.4319/lom.2010.8.155 Schroot, B.M., Klaver, G.T., Schüttenhelm, R.T.M. (2005). Surface and subsurface expressions of gas seepage to the seabed—examples from the Southern North Sea. Mar Pet Geol, 22(4), pp. 499-515. https://doi.org/10.1016/j.marpetgeo.2004.08.007 Seep-bubble characteristics and gas flow rates from a shallow-water, high-density seep field on the shelf-to-slope transition of the Hikurangi subduction margin, Mar Geol, 417, 105985, https://doi.org/10.1016/j.margeo.2019.105985 Tóth, Z., Spieß, V., Jensen, J. (2014). Seismo-acoustic signatures of shallow free gas in the Bornholm Basin, Baltic Sea. Cont Shelf Res, 88, pp. 228-239. https://doi.org/10.1016/j.csr.2014.08.007 122

Vogt, P.R., Cherkashev, G.A., Ginsburg, G.D., KIvanov, M., Milkov, A.V., Crane, K., Lein, A., Sundvor, E., Pimenov, N., Egorov, A. (1997). Haakon Mosby mud volcano provides unusual example of venting. EOS Transactions of the AGU, 78, 549, pp. 556-557. Wenau, S., Preu, B., Spiess, V. (2020). Geological development of the Limpopo Shelf (southern Mozambique) during the last sealevel cycle. Geo-Mar Lett, 40, pp. 363–377. https://doi.org/10.1007/s00367-020- 00648-6

123

6. Experimental study

6.1 Katakolo

6.1.1 Offshore and onshore seepage of thermogenic gas at Katakolo Bay (Western Greece)

Parts of the data and figures (where indicated) are published in: Etiope, G., Christodoulou, D., Kordella, S., Marinaro, G., Papatheodorou, G. (2013). Offshore and onshore seepage of thermogenic gas at Katakolo Bay (Western Greece). Chem Geol, 339, pp. 115-126, https://doi.org/10.1016/j.chemgeo.2012.08.011.

6.1.1.1 Introduction Natural hydrocarbon seepage, either offshore or onshore, has for many years served petroleum exploration as a direct indicator of gas and/or oil subsurface accumulations (Link, 1952; Jones and Drozd, 1983). Surface macro-seeps (visible gas vents or oil leaks from the soil or rock outcrops) are generally an indication of a fault in an active Petroleum Seepage System (Abrams, 2005) belonging to a Total Petroleum System (Magoon and Schmoker, 2000; Etiope et al., 2009a). The assessment of the origin and flux of the seeping gas, is therefore a key task for understanding, without drilling, the subsurface hydrocarbon potential, genesis, and quality, e.g., the presence of shallow microbial gas, deeper thermogenic accumulations, oil, and non-hydrocarbon

undesirable gases (CO2, N2, H2S). The global number of seeps likely exceeds 10,000 (Etiope et al., 2008), but only a small number have been directly investigated. A global analysis of more than 200 onshore seeps worldwide, revealed that methane is thermogenic in about 80% 124

of the cases, microbial gas is in only 4% of seeps, and mixed gas is in the remaining cases (Etiope et al., 2009a). Gas seeps can also indicate subsurface petroleum biodegradation (Etiope et al., 2009b), which has an important impact on hydrocarbon quality and it may influence exploration and production strategies. Seeps can then represent a geo-hazard for humans, buildings, and industry: explosions and sudden flames may occur in gas-rich environments (boreholes, soil), if methane concentrations reach explosive levels of 5–10% in the presence of air. When methane is accompanied by hydrogen sulfide (H2S, e.g., in salt diapirism zones), seeps can be toxic or even lethal under some circumstances (Etiope et al., 2006).

Fig. 5.1.1.1 Structural map of Katakolo showing onshore and offshore seepage areas, modified after Kamberis et al. (2000) and Etiope et al. (2006). Figure from Etiope et al., 2013.

Seeps plumbing can then damage buildings and infrastructures by a gas-pressure build-up in the subsoil or by general degradation of geotechnical properties of soil foundations. Finally, seeps represent a significant global natural source of greenhouse gas (methane; Etiope et al., 2008; US EPA, 2010), photochemical pollutants, and ozone precursors (ethane and propane; Etiope and Ciccioli, 2009). Onshore and offshore seeps (together with diffuse microseepage) are estimated 125

to be the second most important natural source of atmospheric methane, after wetlands, both on a global and European scale (Etiope, 2009). While the global emission of methane from onshore seepage (~22– 44 Tg y−1; Etiope et al., 2008) was assessed on the basis of a wide experimentaldata-set, including direct flux measurements from many countries and different geologic settings, the global emission into the atmosphere from offshore submarine seeps is still based on a pure theoretical exercise (~10–30 Tg y−1; Kvenvolden et al., 2001). Very few in fact, are the flux data from marine seeps referring to the actual amount of gas entering the atmosphere (Hornafius et al., 1999; Judd and Hovland, 2007; Zhou et al., 2009; Clark et al., 2010; Greinert et al., 2010; Yang et al., 2010; Brunskill et al., 2011; Jessen et al., 2011; Schneider von Deimling et al., 2011) and they may include methane originated in recent, shallow sediments which is not completely “fossil” (radiocarbon-free) as defined by Etiope (2009). It is known, however, that significant amounts of methane can reach the atmosphere only for relatively big and shallow seeps, generally not deeper than 200–300 m (Schmale et al., 2005; Mc Ginnis et al., 2006) otherwise most of methane is dissolved and oxidized within the water column. A wide number of flux data are however needed to assess the marine seepage “emission factors”, i.e., the basic element of the up-scaling procedures and greenhouse gas emission estimates on large scales (EMEP/EEA, 2009). In this work we show the results of a wide investigation, carried out in 2009 and 2010, of one of the biggest offshore–onshore gas seepage areas in Europe, located in the Katakolo Bay, along the Ionian coast of Peloponnesus, Greece (Fig. 5.1.1.1). The origin of gas was preliminarily assessed in 2006 and partial flux measurements were done onshore (Etiope et al., 2006). Thermogenic CH4-dominant gas

(>80 %) includes toxic and lethal amounts of H2S (up to 0.5 vol.%), which induce a severe hazard for the local harbor and touristic activity.

The origin of H2S was suspected to be due to thermochemical sulfate reduction (TSR), based on the large H2S amounts in Jurassic 126

limestones (2,500 m deep) and the wide presence of anhydrite. Here we extend the study by reporting: (a) side scan sonar and 3.5 kHz subbottom profile data, providing a complete “picture” of the offshore seepage area; (b) new molecular and isotopic data of the seeping gas including, for the first time, the isotopic composition of H2S, CO2, ethane

(C2H6), and propane (C3H8); (c) extensive gas flux measurements, both onshore and offshore, based on a closed-chamber system. These data allowed the characterization of the Katakolo seepage in terms of areal distribution and extent, its link with tectonic structures, origin of all gases, including an evaluation of the type and maturity of the source rocks, and the total methane, ethane, and propane emission to the atmosphere from the whole seepage area.

6.1.1.2 Geologic setting and description of the seepage area

6.1.1.2.1 General geology and tectonics Katakolo Bay is located in the northwestern Peloponnesus petroliferous basin. This area belongs to the central Ionian geotectonic zone, which is one of the external units of the Hellenides (Fig. 5.1.1.2). The external Hellenides are part of the active margin of the Eurasian Plate. The Ionian Basin and the shallow-water carbonate platforms on the east and west (the Gavrovo and Apulian platforms respectively) were formed during Early Mesozoic Tethyan opening (Aubouin and Decourt, 1962; Fig. 5.1.1.2). Although the production of platform carbonates persisted through the entire Jurassic in the adjacent Paxos, Pre-Apulian, and Gavrovo Zones, the Ionian Basin became an area of stronger faulting and ensuing subsidence. This paleogeographic configuration continued with minor of- and onlap movements along the basin margins until the late Eocene when orogenic movements and flysch sedimentation began (Rigakis and Karakitsios, 1998). In the Ionian geotectonic zone, the 127

stratigraphic succession consists of Triassic evaporites, Triassic to Eocene carbonates, Tertiary flysch and post-Alpine Neogene marine, and continental clastic sediments (Karakitsios, 1995; Rigakis and Karakitsios, 1998; Kamberis et al., 2000; Zelilidis et al., 2003). Upper Triassic to Middle Jurassic deposits consist of anhydrites and carbonates. Evaporitic sedimentation ceased with the onset of Ionian rifting.

Fig. 5.1.1.2 (a) Map showing the location of Katakolo seepage areas and photos showing (b) bubbles rising from a fissure in the seabed of the harbor covered by white bacterial mats, (c) measurements of gas fluxes by closed-chamber system in front of Port Authorities building in the harbor, (d) Faros seepage area, (e) natural flame at Faros, and (f) bubbles rising from Posidonia oceanica -covered seabed at offshore Katakolo South area. Figure from Etiope et al., 2013.

The Katakolo area is crossed by a series of faults at the base of Neogene sequence (Fig. 5.1.1.1). According to Kamberis et al. (2000) the main diapiric structure of Triassic evaporites is associated with a N– S trending thrust fault close to the western shoreline of the Katakolo Peninsula, which bounds to the west. The above mentioned fault and a smaller one located close to the offshore KA-103 well (Kamberis et al., 2000), have created a typical flower structure. Occasional active diapiric 128

movements, affecting the whole detrital Upper Cenozoic sequence, resulted in the formation of synsedimentary faults with horizontal components of movement, which have been active since Lower Pliocene. The main trends of the faults at the base of Neogene are quite similar to those of the Pliocene–Pleistocene synsedimentary NW to NE-trending extensional structures (Kamberis et al., 2000).

6.1.1.2.2 Hydrocarbon potential In the Western Greece hydrocarbon province, petroleum systems (reservoir, source type andmaturity) arewell defined in the Epirous region (e.g., Rigakis and Karakitsios, 1998; Karakitsios and Rigakis, 2007), but not in the Western Peloponnesus Katakolo zone, where source rocks of the several oil and gas reservoirs are uncertain (Karakitsios, Rigakis and Kamberis, personal communications). In general, the main Ionian Zone source rocks are all Mesozoic andwith Type I and II organic matter (with up to ~29% of total organic carbon), including mainly the Lower Cretaceous Vigla shales (with kerogen carbon isotopic composition δ13C:−26.8‰), the Lower Jurassic Posidonia shales (δ13C:−25 to −27‰), and the deeper Triassic shales intercalated in the evaporites (δ13C: −27.2 to −27.8‰; Karakitsios and Rigakis, 2007). The oil window in the central Ionian Basin occurs between 3,700 m and 5,800 m and the Triassic shales entered the dry gas window (Karakitsios and Rigakis, 2007). Reservoir rocks are in Upper Miocene–Lower Pliocene sandstones, Upper Cretaceous–Eocene limestones, which hostmost of oil in Katakolo and in the deeper Jurassic Pantokrator limestones (Karakitsios and Rigakis, 2007). The cap rocks are mainly formed by post-Alpine flysches and Triassic evaporites, whose diapiric salt tectonics dominate the Ionian zone. Kokinou et al. (2005) reported the occurrence of dry gases (C1/C2+C3 ratios >1000) in Late Cenozoic siliclastics and Mesozoic carbonates and in shallower (b2000 m deep) Neogene clastic 129

sequences; they assumed that the dry gas has a microbial (or mixed) origin only based on the molecular composition.

6.1.1.2.3 Surface seepage The active deep faults, likely related to salt tectonics, act as preferential pathways for the natural gas upward migration, producing seeps on and off the Ionian coast at Katakolo Bay (Etiope et al., 2006). Gas seepage occurs in two offshore and two onshore areas, hereafter named as “offshore Katakolo south”, “offshore Katakolo harbour”, “onshore Katakolo harbour”, and “onshore Faros seep” (Fig. 5.1.1.2). The offshore Katakolo south seepage is distributed throughout approximately 25,000 m2, at depths ranging from 9 to 16 m, along a N– S fault evidenced by seismic profiles (Figs. 5.1.1.2a, f, 5.1.1.3a). In the harbor several bubble plume fields are visible from the wharfs over a wide area (~100 m2), at depths of 7 to 10 m. Onshore seeps in the harbor have penetrated and damaged the asphalt pavement along the main wharf, especially around the Duty Free Shop building (Katakolo is a port of call for tourist cruise ships) and at the entrance of the building of the local port authority (Figs. 5.1.1.2c, 5.1.1.3b). In 1972 a flame blew-out from the pavement of the main wharf destroying a pole (Etiope et al., 2005, 2006). The other onshore Katakolo seepage is located about 0.5 km northward at a hilly site named “Faros” (Figs. 5.1.1.2a, d, e, 5.1.1.3c). It is an 800 m2 wide, elongated dry seep along an active fault. Some vents can be easily ignited with a lighter producing flames about 20 cm in height. Stamatakis et al. (1987) used the term “mud volcano” for this seep; we outline instead, that it is simply a dry gas seep, without water or mud discharge or evidence for breccia release.

130

Fig. 5.1.1.3 (a) Bathymetric map of Katakolo Harbour and offshore Katakolo south areas showing the gas flux measurements offshore and onshore sites and the MEDUSA system tracklines; (b) detailed view of Katakolo Harbour showing the gas flux measurements sites and the location of the offshore gas sample (P.A.: Port Authorities Building; D.F.S.: Duty Free Shop building); (c) detailed map showing the Faros seepage area and the gas flux measurement sites. Figure from Etiope et al., 2013.

6.1.1.3 Methods

6.1.1.3.1 Marine Remote Sensing Surveys The marine remote survey utilized: (i) an EG&G 272 TD side scan sonar tow fish of a dual frequency (100 kHz and 500 kHz) in association with a 4200 Topside processor EDGETECH, digital corrected image recorder and Positional data for the remote sensing with horizontal resolution of ±2 m, was provided by a Hemisphere global positioning system (GPS) with differential corrections from the marine beacon system (Fig. 5.1.1.4). Positional data were downloaded at a 2-s interval using the WGS-84 ellipsoid. The navigation of the vessel was managed by the Delphmap software. The side scan sonar system emits high frequency acoustic pulses which scan the seafloor and provides a plan view seafloor acoustic image. 131

Since ground-truthing showed that the active gas emission in the form of gas bubbles is detectable by acoustic means, as evidenced by anomalies in acoustic images (Hovland and Sommerville, 1985; Hovland and Judd, 1988; Judd et al., 1997; Hornafius et al., 1999), the EG&G side scan sonar records were also used to track ascending bubbles from the seafloor to the surface (Judd et al., 1997). Gas bubbles present in the water column produce acoustic anomalies caused by the impedance contrast between the gas bubbles and the surrounding water. The acoustic anomalies interpreted as gas bubbles in the water column will be referred to ‘gas flares’ as suggested by Greinert et al. (2006) (“hydroacoustic manifestations of bubbles in the water column”), in order to distinguish them from the non acoustic ‘plumes’, that are defined as water column volumes with high concentrations of dissolved gas (Obzhirov et al., 2004).

Fig. 5.1.1.4 The EG&G side scan sonar (top) systems aboard the research vessel (bottom) used for the geophysical survey offshore Katakolo Bay. Photos: Stavroula Kordella (top), Margarita Iatrou (bottom). 132

6.1.1.3.2 Underwater Visual Inspection and Gas Detection The offshore seepage area (Fig. 5.1.1.3a), both inside and outside the harbor (“offshore Katakolo south”), was explored with an underwater system, MEDUSA (Module for Environmental Deep UnderSea Applications). It is a cabled and instrumented aluminum-frame module for casts and towed surveys close to the seabed (Marinaro et al., 2011). MEDUSA was equipped with underwater methane sensors (semiconductor METS by Franatech GmbH and the optical HydroC by CONTROS Systems & Solutions GmbH), a video camera (Multi SeaCam 1060), lights and other standard oceanographic sensors for the measurement of temperature, salinity, pressure, oxygen, water turbidity, and distance from the seabed (Marinaro et al., 2011). Once deployed in seawater, MEDUSA transmitted in real-time all data collected by its sensors toward a Surface Control Unit (Fig. 5.1.1.5) through an umbilical cable.all data collected by the installed payload (see Table 5.1.1.1), were acquired at 1 Hz sampling rate. All measurements and visual recordings had the same time reference given by a GPS to allow for correlations between different parameters and images, including those relevant to the system status, like position and altitude. In the Katakolo Bay, MEDUSA was towed from a small research vessel in order to explore the seabed in the seepage areas previously identified by remote sensing survey. Horizontal “fly-overs” were conducted keeping the module as close as possible to the seafloor (generally within 1–2 m). In this work we report the main results of the methane detection; more complete results of all other MEDUSA sensors will be presented elsewhere. 133

Fig. 5.1.1.5 Left: The MEDUSA Surface Control Unit.Center: The MEDUSA towed from the research vessel. Right: The research vessel (Barbantonis). Photos: Margarita Iatrou, Stavroula Kordella.

Table 5.1.1.1 MEDUSA platform scientific payload.

DESCRIPTION MODEL- RANGE RESOLUT. MANUFACTURER

CH4 sensor (1) K-METS - 0.1-10 µM 2 nM Franatech

CH4 sensor (2) HydroC - 0.1-100 µM 10 nM CONTROS

O2 sensor Optode 3830 - 0-500 µM <1 µM AANDERAA

H2S sensor AMT microsensor 0-50 mg/L 0.1 mg/L

TV camera + Multi SeaCam 1060 light – DEEPSEA 134

DESCRIPTION MODEL- RANGE RESOLUT. MANUFACTURER

POWER&LIGHT

CTD SBE-19plus 0-9 S/m 0,00007 S/m -5;35 °C 0,0001 °C 7000 dbar 0,002% f.s.

Turbidimeter Wet labs ECO-NTU, 0,0024-5 m -1 controlled by CTD

6.1.1.3.3 Gas sampling and analyses One offshore gas vent (bubble plume 7 m deep) was sampled by divers in the Katakolo Harbor (Fig. 5.1.1.3b). Gas bubbles were collected at seabed directly into a 250 mL glass bottle sealed underwater with PTFE (polytetrafluoroethylene) butyl rubber septa and aluminum caps. Other two gas samples were collected from the onshore Faros seep by an inverted funnel connected to a syringe and a silicon tube equipped with a T-valve, in order to reduce atmospheric air contamination; the samples were stored in 100 mL glass tubes equipped with two vacuum stop-cocks. All samples were analyzed at Isotech Labs Inc. (Illinois, USA) for

C1–C6 hydrocarbons, He, H2, Ar, O2, CO2, N2, (Shimadzu 2010 TCD- FID GC; accuracy and precision 2% (1σ)) and isotopic compositions

13 13 13 2 13 34 δ CCH4, δ CC2H6, δ CC3H8, δ HCH4, δ CCO2, and δ SH2S (Finnigan Delta Plus XL mass spectrometer, precision ± 0.3‰ (1σ) for 13C and δ34S, ± 4‰ (1σ) for 2H). Rapid onsite analyses of gas bubbles (at seabed and sea-surface) and gas vents in the harbor were also made for H2S and SO2 by RAE Systems colorimetric tubes and sampling pump (accuracy <10%) and for CH4 by thermal conductivity detector used for flux measurements, described below. 135

6.1.1.3.4 Gas Flux Measurements Offshore A closed-chamber system used onshore (see below) was modified into a floating chamber for measuring gas flux at the sea surface in correspondence with bubble trains (Fig. 5.1.1.6). Floating chambers are widely used to mea- sure gas (CO2, CH4, and N2O) ebullition in aquatic environments, wetlands, and lakes (e.g., Bastviken et al., 2004; Cole et al., 2010); bigger gas capture buoys were used at the Coal Oil Point seep field, offshore California (Washburn et al., 2001; Clark et al., 2010). The aluminum chamber was positioned on a buoyant polystyrene collar forming a net chamber volume of 14 L (Fig. 5.1.1.6). By using a small boat, the floating chamber was positioned on 32 bubble plumes, inside the harbor and in the “off- shore Katakolo south” area (Fig. 5.1.1.6). CH4 concentration build-up in the chamber was quite rapid, as in the onshore vents, reaching orders of several hundred ppmv up to 2000 ppmv, in 30–40s.

a a

b c

Fig. 5.1.1.6 (a) Graphic representation of the floating chamber; (b, c) Flux measurements on submarine gas seeps, with the floating chamber connected to West Systems CH4-CO2 sensors.

6.1.1.3.5 Gas Flux Measurements Onshore Methane fluxes from vents and diffuse seepage (miniseepage) were measured by a widely tested equipment (e.g., Etiope et al., 2011) based on a closed-chamber system (West Systems srl, Italy) equipped with a portable CH4 sensor and wireless data communication to a palm-top 136

computer; gas flux is automatically calculated through a linear regression of the gas concentration build-up in the chamber. The CH4 sensor includes semiconductor (range 0–2000 ppmv; lower detection limit: 1 ppmv; resolution: 1 ppmv), catalytic (range: 2000 ppmv–3 vol. %), and thermal conductivity (3 vol.% to 100 vol.%) detectors. Maximum accumulation time of 15 min allowed the detection of fluxes down to 10 mg m−2 d−1. The chamber (in aluminum, 10 L volume) was equipped with a Nafion® dryer for humidity removal. In the Katakolo harbor, gas flux was measured in 29 vents (37 chamber measurements), corresponding to cracks or fissures in the asphalt pavement, spread over 15,000 m2 (Fig. 5.1.1.3b). At the Faros seep site, 51 flux measurements were made throughout a total area of 5000 m2 (Fig. 5.1.1.3c).

6.1.1.4 Results and Discussion

6.1.1.4.1 Areal distribution of the seeps and relationship with local faults Examination of high resolution seismic profiles and side scan sonar images in the Katakolo Bay, has shown the existence of a variety of anomalous acoustic characters indicative of the presence of gas in the interstices of the sediments (Jones et al., 1986, Papatheodorou et al., 1993, Judd and Hovland, 2007). Moreover, acoustic data contains backscatter signals (gas flares) from sources within the water column considered to represent gas bubbles (e.g., Judd et al., 1997). At the “offshore Katakolo south”, five areas with a total extent of 25,000 m2 and with more than 300 counted bubble streams were recorded based on digital side scan sonar data. Many of the bubble streams, detected hydroacoustically, were also observed by the underwater video of MEDUSA system (Fig. 5.1.1.5). According to visual observation by divers and underwater video, bubbles with a diameter ranging from 0.5 to 1 cm were issuing from the seagrass-covered 137

seabed (Fig. 5.1.1.2f). Bubble streams ascending the water column were often visible on the sea surface at calm seas. The concentration of dissolved CH4 recorded by MEDUSA sensors ranged from 50 to 60 nM in shallow water, to ~ 0.9 μM close to seabed around bubble plumes (Fig. 5.1.1.7). During the horizontal passages approaching some plumes (observed by the camera) CH4 signals increased by 0.2–0.5 μM over 10–40 m (the ship velocity was about 2 knots) (Fig. 5.1.1.7). All bubble streams and associated CH4 anomalies were observed along the fault line and on its western side (footwall), and they were particularly abundant at the northern part of the fault line.

Fig. 5.1.1.7 Side scan sonar mosaic of the offshore Katakolo south area and example of methane detection in seawater performed with the MEDUSA system along the offshore Katakolo fault. White dots represent the gas bubble plumes visually detected by the MEDUSA underwater video camera. Maximum distance between MEDUSA and seabed was 3 m. Figure from Etiope et al., 2013.

Methane concentration signals recorded by MEDUSA were generally higher than those recorded in the southern area outside the 138

harbor, with peaks reaching concentrations of 8 μM. Both MEDUSA and divers have found bubbles up to the order of 10–20 cm in diameter issuing from cracks in the seabed covered by white bacterial mats of Beggiatoa spp. (Fig. 5.1.1.2b). The MEDUSA system allowed the underwater visual inspection of 68 bubble plumes along the offshore Katakolo south area (Figs. 5.1.1.2, 5.1.1.7-8). The MEDUSA showed no significant temperature and salinity variations around the seeps, implying fluid emission from the seeps is free of fresh water. The

MEDUSA system detected variations of CH4 and dissolved O2 (DO) related to the numerous gas seeps and carbonate mounds along the offshore Katakolo south area (Figs. 5.1.1.7-8). The dissolved methane and oxygen measurements performed by MEDUSA showed a clear inverse correlation between methane related to the bubbles and the level of O2 in seawater surrounding the bubbles. Regardless of the open sea conditions, O2 decreases locally (ΔDO up to ~10 μM, from a background of 218-219 μM) close to the main seep fields, where CH4 concentration in seawater is higher (Figs 5.1.1.8-9). Complete compositional analyses including dissolved O2 (in surface bubbles and surrounding seawater) should be made to test the gas-exchange model (McGinnis et al., 2006) and quantify the role of bubbling as driver of dissolved O2 depletion.

Fig. 5.1.1.8 3-D presentations of methane and dissolved oxygen concentrations obtained with the MEDUSA system, over the side scan sonar mosaic of the Katakolo South area. 139

Here it is shown that when there is increased methane seepage (methane concentration line is coloured red) the dissolved oxygen descreases substantially (DO line turns white). Perpendicular short, pale blue lines represent bubble plumes visually detected by the MEDUSA underwater video camera.

Fig. 5.1.1.9 Left: Side scan sonar mosaic of the offshore Katakolo south area and example of methane detection in seawater performed with the MEDUSA system along the offshore Katakolo fault. Purple dots represent the gas bubble plumes visually detected by the MEDUSA underwater video camera. Thin grey lines are the seepage zones, where bubble plumes were hydroacoustically detected. Right: Graphic representation of methane and dissolved oxygen concentrations in seawater, obtained by the MEDUSA system. Over the bubble plumes, methane concentration values are elevated and dissolved oxygen is lower than in when seepage is absent.

At “onshore Katakolo harbour”, seeps are located on the main dock and at the establishments of the harbor, mainly around two buildings (Duty Free Shop and Port Authorities) probably due to the loading of the substrata from the heavy constructions. The position of the seeps was the same, but with some more vents as observed four years before, in 2006 (Etiope et al., 2006). Τhe harbor is located exactly in correspondence with the convergence of the onshore Faros and offshore Katakolo south fault systems (Fig. 5.1.1.1); this would lead to higher fracturing and permeability of the rocks, explaining the diffuse and vigorous seepage. 140

6.1.1.4.2 Molecular Gas Composition

Offshore seeps release gases with dominant CH4 concentrations, and subordinate amounts of N2, CO2, H2S, and C2+ alkanes (Table 5.1.1.2). This gas is similar to that released onshore, as discussed below.

In the gas bubbles sampled offshore, CH4 and H2S concentrations decreased from the seabed (7 m depth) to the sea- surface: CH4 was 85.6 vol.% at the base of a bubble plume (Table

5.1.1.2) and 80 vol.% at the surface. H2S decreased from 0.63 vol.% or 400 ppmv (sample analysed on-site by colorimetric tube) to 0.2 ppmv on the surface. The H2S decrease was probably due to rapid oxidation in seawater, as also suggested by the increase of SO2 from 0 (at the bottom) to 5 ppmv (at the surface, analysed by colorimetric tube). The

CH4 decrease from bottom to the surface is a well-known process due to gas exchange between seawater and bubble (N2 and O2 from seawater replace CH4 in the bubble; e.g., McGinnis et al., 2006). Theoretical calculation based on bubble dissolution models (Leifer and Patro, 2002) suggests that after 7 m of upward rise, a bubble with diameter of about 0.5-1 cm (as observed by divers) loses less than 10% of its initial CH4 amount.

Table 5.1.1.2 Gas composition and isotopic data of Katakolo offshore and on shore seeps. Table from Etiope et al., 2013.

Gas concentration in vol. %; δ13C in ‰ vs VPDB; δD in ‰ vs VSMOW; δ34S in ‰ vs VCTD a from Etiope et al. (2006); b from Kamberis et al. (2000); sp: sulfur precipitation on a seep in the Katakolo harbour; bdl: below detection limit 141

6.1.1.4.3 Isotopic Gas Composition and Gas Origin

Stable carbon and hydrogen isotopic compositions of CH4 show a clear

13 13 thermogenic origin, either onshore (δ C1: -34.3 ‰) or offshore (δ C1: - 36.5 ‰; Fig. 5.1.1.10). The concentration and isotopic composition of

CH4 in the bubble at seabed are similar to those of the gas at the onshore vents, which suggests that the sampled offshore gas was not fractionated and actually represented the original seeping gas. The isotopic data confirm those acquired in 2006 (Etiope et al., 2006), and this suggests that there is no significant variation in the gas source. The

13 stable carbon isotopic ratio of CO2 is C-enriched with respect to typical catagenetic CO2 related to kerogen decarboxilation (generally in the range -15 to -25 ‰); it suggests CO2 reduction in secondary methanogenesis following hydrocarbon biodegradation (Etiope et al., 2009b; Milkov, 2011). The offshore gas seep shows a stronger biodegradation signal, as also indicated by the 13C-enrichment of propane (Fig. 5.1.1.11). Secondary microbial methane is however isotopically indistinguishable from thermogenic methane (Brown, 2011; Milkov, 2011).

142

Fig. 5.1.1.10 Genetic zonation diagrams of methane. (a): δ13C vs δ2H plot; To:

13 thermogenic with oil; Tc: thermogenic with condensate; TD: dry thermogenic. (b): δ C vs

C1(C2 + C3) plot. Figure from Etiope et al., 2013.

13 Fig. 5.1.1.11 Natural gas plot, with the δ C sequence of C1–C3 alkanes of Katakolo offshore (biodegraded) and onshore Faros seep. Figure from Etiope et al., 2013. While thermogenic hydrocarbon (gas and oil) reservoirs in the Katakolo area are known to be in Upper Cretaceous limestones (Kamberis et al., 2000; Karakitsios and Rigakis, 2007), the source rocks are unknown, as discussed in section 2. Candidates source rocks have 143

all marine, Type II, organic matter and are the Lower Cretaceous Vigla shales (δ13C: -26.8 ‰), the Lower Jurassic Posidonia shales (δ13C: -25 to -27‰), and the deeper Triassic shales (δ13C: -27.2 to -27.8 ‰; Karakitsios and Rigakis, 2007). Thermogenic gas formation modeling (Fig. 5.1.1.11); based on isotope modeling by Tang et al., 2000 and conducted with GeoIsochem Corp. GOR Isotope software 1.94 for instantaneous generation of C1-C2 from default Type I, II and III kerogen, suggests that Katakolo gas may be produced by a Type II kerogen with a maturity around 1.9-2.0 Ro (vitrinite reflectance) and temperatures of about 190-200 °C or by a Type III kerogen with maturity of 1.3 Ro. We can exclude the Type III organic matter, as it was never reported in the Ionian Zone (Karakitsios and Rigakis, 2007). Type II kerogen is further confirmed by the maturity plot based on the Berner

13 and Faber (1996) model (Fig. 5.1.1.12): the δ C values of C1 and C2 of Katakolo fall exactly along the maturity line of marine kerogen, which suggests a maturity of 1.6-1.8 Ro for a source rock having δ13C: - 27.8‰, like the Triassic shales or a maturity slightly lower, 1.4-1.6 Ro, for a kerogen with δ13C: -26.8‰, like the Vigla Cretaceous shales. Maturity plot and thermogenic modelling would coincide in the case of the Triassic shales (Fig. 5.1.1.13). This is consistent with the fact that Triassic shales entered the dry gas window (Karakitsios and Rigakis, 2007).

144

Fig. 5.1.1.12 Thermogenic gas formation modeling from default marine (Type II) and terrestrial (Type III) kerogen, calculated using GeoIsochem Corp. GOR software 1.94; heating rate of 5 °C per million year (Tang et al., 2000, Etiope et al., 2011). Vertical arrow indicates the maturity (1.9–2.0 Ro) inferred by assuming that Katakolo gas derives from marine kerogen, as reported by Karakitsios and Rigakis (2007). Figure from Etiope et al., 2013.

Fig. 5.1.1.13 Methane vs. ethane maturity plot of Katakolo source rocks based on the model by Berner and Faber (1996). % Ro: vitrinite reflectance; kerogen isotope compositions are those of the source rocks suggested by Karakitsios and Rigakis (2007). Figure from Etiope et al., 2013. 145

34 The sulfur isotopic composition of H2S (δ SH2S: +2.4 ‰) suggests an origin related to thermochemical sulphate reduction (TSR) or thermal decomposition of sulfur compounds in kerogen or oil, rather than a production mediated by bacteria (Machel et al., 1995). The δ34S values of thermally generated H2S are typically “positive”, similar to those of the precursor oil/kerogen and around 5-10 ‰ lower than those of thermally altered anhydrite (Desrocher et al., 2004; Krouse et al., 1988). There are no available data of the sulphur isotopic composition of the Katakolo anhydrites but similar Triassic rocks have typically

34 δ SH2S values between +10 and +30‰ (Desrocher et al., 2004;

34 Horacek et al., 2010); so, Katakolo δ SH2S would be consistent with a TSR from Triassic evaporites. Etiope et al. (2006) already hypothesized that H2S in Katakolo has a TSR origin, based on the high concentrations in surface seeps (>1 vol %) and in 2,500 m deep Jurassic carbonate reservoirs (>10 %; Kamberis et al., 2000), and on the presence of anhydrite-limestone contacts at 4,000–5,000 m. TSR is, in fact, the only process able to produce large amounts of H2S and is dominant in the presence of evaporites in contact with limestone at temperatures generally above 80 °C (Noth, 1997). Elemental sulphur precipitated at a seep in the Katakolo harbor, is slightly more enriched

34 in S than the H2S (Table 5.1.1.2), as typically observed after oxidation of H2S (Krouse et al., 1988).

6.1.1.4.4 Offshore Gas Flux At the Katakolo harbour and the “offshore Katakolo south” area methane flux was measured with the floating chamber system at 32 bubble plumes, the exact location of which is shown in Fig 5.1.1.3a, b. In the surrounding area of the measured bubble plumes, further 156 bubble trains were counted by visual observation from the sea surface over about 6,000 m2. Methane flux to the atmosphere from individual bubble trains ranged from 0.05 to 0.42 kg d-1, with an average of 0.12

-1 kg d (Table 5.1.1.3). An estimate of the total CH4 emission to the 146

atmosphere has been made by assuming that the 32 measured plumes were representative, in terms of flux, of the counted 156 plume population. For this exercise, the measured bubble plumes were sorted by size based on a frequency histogram (small: <0.08 kg d-1; medium: 0.08 to 0.25 kg d-1; large: >0.25 kg d-1); we also observed that 20% of counted plumes (= 31 plumes; 20 small + 11 medium) were episodic, with an average frequency of activity of 1 min every 5 min (= 20% of continuous activity). Applying the average measured flux (0.12 kg d-1) and considering the average episodic bubbling frequency, 17.1 kg CH4 d-1 (6.3 t y-1) were estimated to reach the atmosphere from 156 bubble plumes. By applying the same method to the 823 bubble plumes counted in the side scan sonographs (see Chapter 4 ) in the harbour (523 bubble plumes) and in the “offshore Katakolo south” area (300 bubble plumes), 90.5 kg d-1 (33 t y-1) are conservatively estimated (Table 5.1.1.3). Finally, considering the potential number of plumes (~2,700) extrapolated as described briefly in Chapter 4 and in Etiope et al., 2013 in detail. For the harbour area having uncertain acoustic data, total CH4 emission into the atmosphere from the whole offshore seepage area (~0.1 km2) may exceed 300 kg d-1 (100 t y-1). These values are comparable to those reported for the Timor Sea seeps (47- 470 t y-1 over 0.7 km2; Brunskill et al., 2011) and are higher than the emission of the Tommeliten seepage in the North Sea (26 t y-1 over 0.12

2 km ; Schneider von Deimling et al., 2011). Note that our CH4 emission estimate refers only to ebullition flux; diffusive CH4 flux from seawater to the atmosphere was not taken into account in this study; models and measurements suggest, however, that diffusive CH4 output is generally very small in comparison with the ebullition one, especially in shallow water conditions (e.g., McGinnis et al., 2006; Greinert et al., 2010; Schneider von Deimling et al., 2011). The emission of ethane and propane can be estimated following the procedure by Etiope and Ciccioli (2009), based on the C1/C2

(methane/ethane) and C1/C3 (methane/propane) compositional ratios (Table 5.1.1.2): these ratios, ~60 and ~580 respectively, are 147

comparable to the global average ratios of thermogenic seeps (Etiope and Ciccioli, 2009); the C2 and C3 emissions are estimated to be 5.4 kg d-1 (2 t y-1) and 0.5 kg d-1 (0.2 t y-1), respectively.

6.1.1.4.5 Onshore gas flux At the harbor, at least 50 cracks or fissures were observed in the asphalt pavement (Figs. 5.1.1.2c, 3b) and all of them represent gas vents, as indicated by rapid exhalation checks with the portable CH4 sensor. The asphalt coverage acts as a cap for widespread gas accumulations at less than 1–2 m below the surface, as shown by explosive levels of CH4 (~5 to 12 vol.%) in some manholes on the main wharf. The risk is particularly worrying because the harbor is used by tourists embarking and disembarking from cruise ships.

CH4 fluxes measured in 29 fissures ranged from 0.001 to 1.12 kg d−1, with a mean of 0.15 kg d−1 (Table 5.1.1.3); only one large fissure- vent, located a few meters from the Duty-Free Shop, emitted 49.3 kg d−1 (18 t y−1). The simple sum of the measured 28 fluxes gives an output of 54 kg d−1 (19.7 t y−1) and assuming the average vent flux for the other

−1 22 non-measured vents, gives a total CH4 emission of about 57.3 kg d (21 t y−1). At the Faros seep, gas exhalation is pervasive throughout a small elongated valley (800 m2 wide). Four main vent zones (macro-

−2 seeps) were detected, where CH4 fluxes are in the order of 105 mg m d−1. The main vent was easily ignited producing a flame about 20 cm

−1 −1 high: 82 kg CH4 d (30 t y ) are released only at this site. The other

−1 −1 vents release in total 5.5 kg CH4 d (2 t y ). Around the vents, throughout the seepage area, diffuse gas fluxes (miniseepage) range from 40 to 82,000 mg m−2 d−1. Natural neighbour interpolation of such miniseepage values (as made in other seep cases; e.g., Etiope et al.,

−1 −1 2011) suggests a total CH4 output of 98.6 kg d (36 t y ). Total methane emission from the Faros seepage zone is then about 186 kg d−1 (68 t y−1). The harbour and Faros emission estimate are now better evaluated with respect to previous incomplete studies, which reported a 148

total output of about 15 t y−1 and were referring to a partial coverage of the seepage area (Etiope et al., 2006; Etiope, 2009); the CH4 emission now assessed is of the same order of magnitude of other large thermogenic gas seeps in Europe (compare with Table 4 in Etiope, 2009). Based on the C1/C2 and C1/C3 compositional ratios (which are similar to those of the offshore seep; Table 2), the emission of ethane and propane is estimated to be 3.4 kg d−1 (1.25 t y−1) and 0.82 kg d−1 (0.3 t y−1), respectively.

Table 5.1.1.3 Summary of CH4 emissions data in the offshore and onshore Katakolo seepage areas. Figure from Etiope et al., 2013.

a value extracted from the mean value (0.12 kg d-1) taking into account that 20 % of plumes were episodic b excluding the large vent releasing 49.3 kg d-1

6.1.1.5 Conclusions Katakolo results to be one of the most prolific thermogenic gas seepage zones in Europe (surely the biggest methane seep ever reported in Greece) and a natural laboratory to study gas bubbling in the sea and onshore seepage processes. The same type of thermogenic gas is seeping offshore and onshore in correspondence with two main normal faults converging to the harbour zone, which shows a spectacular field of gas bubble plumes. Isotopic data, combined with thermogenic gas generation modelling and maturity plots, suggest that the gas is related to a deep Petroleum System characterized by Jurassic carbonate 149

reservoirs, Triassic source rocks, and Triassic evaporites. The contact between carbonates with hydrocarbons and evaporites determines the production of H2S by TSR. Secondary alteration of gas (biodegradation) seems to occur at shallow depths, mainly in correspondence with the offshore seepage. These data represent the missing piece of the “puzzle” of the Katakolo petroleum system, whose isotopic gas composition and source rock were never carefully evaluated, notwithstanding this area may have a great potential for natural gas productivity. This potential is now further confirmed by the relevant amounts of gas continuously migrating to the surface: the location of the individual seeping points did not significantly change over five years and, indeed, they appeared to increase in number. Total offshore CH4 emission into the atmosphere (considering only ebullition flux) may

−1 2 range from 33 to 120 t y , over about 94,200 m . Onshore CH4 output was estimated to be around 89 t y−1 for only 15,800 m2. These values, combined with the isotopic data, suggest the existence of deep pressurized reservoirs and a powerful petroleum seepage system, which may deserve further exploration. The gas flux measurements contributed to improve the global data-set of marine and onshore geologic emissions of methane, ethane, and propane to the atmosphere; the bubble plume output data, in particular, can be used to refine the global statistics and emission factors of marine seeps, still today poorly assessed (US EPA, 2010).

References Abrams, M.A. (2005). Significance of hydrocarbon seepage relative to petroleum generation and entrapment. Mar Pet Geol, 22, pp. 457–477. Aubouin, J., Decourt, J. (1962). Zone preapulienne, zone ionienne et zone du Gavrovo en Peloponnese occidental. Bulletin de la Societe Geologique de France, 4, 7, pp. 85–794. Bastviken, D., Cole, J., Pace, M., Tranvik, L. (2004). Methane emissions from lakes: dependence of lake characteristics, two regional 150

assessments, and a global estimate. Global Biogeochemi Cycles, 18, GB4009. Berner, U., Faber, E. (1996). Empirical carbon isotope/maturity relationships for gases from algal kerogens and terrigenous organic matter, based on dry, open-system pyrolysis. Org Geochem, 24, pp. 947–955. Brown, A. (2011). Identification of source carbon for microbial methane in unconventional gas reservoirs. Am Assoc Pet Geol Bull, 95, pp.1321–1338. Brunskill, G.J., Burns, K.A., Zagorskis, I. (2011). Natural flux of greenhouse methane from the Timor Sea to the atmosphere. J Geophys Res, 116, G02024. Christodoulou, D. (2010). Geophysical, sedimentological study: remote sensing on pockmarks in seismogenic active areas. PhD Thesis, University of Patras, Patras, 296 pp. Clark, J.F., Leifer, I., Washburn, L., Luyendyk, B.P. (2003). Compositional changes in natural gas bubble plumes: observations from the Coal Oil Point seep field. Geo-Mar Lett, 23, pp. 187–193. Clark, J.F., Washburn, L., Schwager, K. (2010). Variability of gas composition and flux intensity in natural marine hydrocarbon seeps. Geo-Mar Lett, 30, pp. 379–388. http://dx.doi.org/ 10.1007/s00367-009- 0167-1 Cole, J.J., Bade, D.L., Bastviken, D., Pace, M.L., Van de Bogert, M. (2010). Multiple approaches to estimating air–water gas exchange in small lakes. Limnology and Oceanography: Methods, 8, pp.285–293. Desrocher, S., Hutcheon, I., Kirste, D., Henderson, C.M. (2004).

Constraints on the generation of H2S and CO2 in the subsurface Triassic, Alberta Basin, Canada. Chem Geol, 204, pp. 237–254. EMEP/EEA, 2009. EMEP/EEA air Pollutant Emission Inventory Guidebook — 2009. European Environment Agency, Copenhagen http://dx.doi.org/10.2800/23924 Etiope, G. (2009). Natural emissions of methane from geological seepage in Europe. Atmos Environ, 43, pp.1430–1443. 151

Etiope, G., Ciotoli, G., Schwietzke, S., Schoell, M. (2019). Gridded maps of geological methane emissions and their isotopic signature. Earth Syst Sci Data, 11, pp. 1–22. https://doi.org/10.5194/essd-11-1- 2019 Etiope, G., Ciccioli, P. (2009a). Earth's degassing — a missing ethane and propane source. Science, 323 (5913), 478. Etiope, G., Papatheodorou, G., Christodoulou, D., Favali, P., Ferentinos, G. (2005). Gas hazard induced by methane and hydrogen sulfide seepage in the NW Peloponnesus Petroliferous Basin (Greece). TAO, 16, pp.897–908. Etiope, G., Papatheodorou, G., Christodoulou, D., Ferentinos, G., Sokos, E., Favali, P. (2006). Methane and hydrogen sulfide seepage in the NW Peloponnesus petroliferous basin (Greece): origin and geohazard. Am Assoc Pet Geol Bull, 90, pp. 701–713. Etiope, G., Lassey, K.R., Klusman, R.W., Boschi, E. (2008). Reappraisal of the fossil methane budget and related emission from geologic sources. Geophys Res Lett, 35, L09307. Etiope, G., Feyzullayev, A., Baciu, C.L. (2009a). Terrestrial methane seeps and mud volcanoes: a global perspective of gas origin. Mar Pet Geol, 26, pp.333–344. Etiope, G., Feyzullayev, A., Milkov, A.V., Waseda, A., Mizobe, K., Sun, C.H. (2009b). Evidence of subsurface anaerobic biodegradation of hydrocarbons and potential secondary methanogenesis in terrestrial mud volcanoes. Mar Pet Geol, 26, pp.1692–1703. Etiope, G., Nakada, R., Tanaka, K., Yoshida, N. (2011). Gas seepage from Tokamachi mud volcanoes, onshore Niigata Basin (Japan): origin, post-genetic alterations and CH4\CO2 fluxes. Appl Geochem, 26, pp.348–359. Géli, L., Henry, P., Zitter, T., Dupré, S., Tryon, M., Çaĝatay, M.N., de Lépinay, B.M., Le Pichon, X., Şengör, A.M.C., Görür, N., Natalin, B., Uçarkuş, G., Özeren, S., Volker, D., Gasperini, L., Burnard, P., Bourlange, S., the Marnaut Scientific Party (2008). Gas emissions and active tectonics within the submerged section of the North Anatolian 152

Fault zone in the Sea of Marmara. Earth Planet Sci Lett, 274, pp. 34– 39. Greinert, J., Artemov, Y., Egorov, V., Batist, M., McGinnis, D. (2006). 1300-m-high rising bubbles from mud volcanoes at 2080 m in the Black Sea: hydroacoustic characteristics and temporal variability. Earth Planet Sci Lett, 244, pp. 1–15. Greinert, J., McGinnis, D.F., Naudts, L., Linke, P., De Batist, M. (2010). Atmospheric methane flux from bubbling seeps: spatially extrapolated quantification from a Black Sea shelf area. J Geophys Res, 115, C01002. Guinasso, N.L., Schink, D.R. (1973). A Simple Physiochemical Acoustic Model of Methane Bubbles Rising in the sea. Texas A&M University, College of Geosciences, Texas. Horacek, M., Brandner, R., Richoz, S., Povoden-Karadeniz, E. (2010).Lower Triassic sulphur isotope curve of marine sulphates from the Dolomites, N-Italy. Palaeogeogr Palaeoclimatol Palaeoecol, 290, pp. 65–70. Hornafius, J.S., Quigley, D., Luyendyk, B.P. (1999). The world's most spectacular marine hydrocarbon seeps (Coal Oil Point, Santa Barbara Channel, California): quantification of emissions. J Geophys Res Oceans Atmospheres, 20 (C9), 20, pp.703–20,711. Hovland, M., Judd, A.G. (1988). Seabed pockmarks and seepages. Impact on Geology, Biology and the Marine Environment. Graham & Trotman, London. Hovland, M., Sommerville, J.H. (1985). Characteristics of two natural gas seepages in the North Sea. Mar Pet Geol, 2, pp. 319–326. Jessen, G.L., Pantoja, S., Gutierrez, M.A., Quiñones, R.A., Gonzalez, R.R., Sellanes, J., Kellermann, M.Y., Hinrichs, K.U. (2011). Methane in shallow cold seeps at Mocha Island off central Chile. Cont Shelf Res, 31, pp. 574–581. Jones, V.T., Drozd, R.J. (1983). Predictions of oil or gas potential by near-surface geochemistry. AAPG (Am Assoc Pet Geol) Bull, 67, pp. 932–952. 153

Jones, G.B., Floodgate, G.D., Bennell, J.D. (1986). Chemical and microbiological aspects of acoustically turbid sediments: preliminary investigations. Marine Geotechnology, 6, pp. 315–332. Judd, A.G., Hovland, M. (1992). The evidence of shallow gas in marine sediments. Cont Shelf Res, 12, pp. 1081–1096. Judd, A.G., Hovland, M. (2007). Submarine Fluid Flow, the Impact on Geology, Biology, and the Marine Environment. Cambridge University Press, Cambridge. Judd, A.G., Davies, G., Wilson, J., Holmes, R., Baron, G., Brydon, I. (1997). Contributions to atmospheric methane by natural gas seepages on the UK continental shelf. Mar Geol, 137, pp.165–189. Kamberis, E., Rigakis, N., Tsaila-Monopolis, S., Ioakim, C., Sotiropoulos, S. (2000). Shallow biogenic gas-accumulations in late Cenozoic sands of Katakolon Peninsula, western Greece. Geol Soc Greece Spec Publ, 9, pp. 121–138. Karakitsios, V., 1995.The influence of pre-existing structure and halokinesis on organic matter preservation and thrust system evolution in the Ionian basin, Nortwestern Greece. AAPG (Am Assoc Pet Geol) Bull, 79, pp. 960–980. Karakitsios, V., Rigakis, N. (2007). Evolution and petroleum potential of Western Greece. J Pet Geol, 30, pp. 197–218. Kokinou, E., Kamberis, E., Vafidis, A., Monopolis, D., Ananiadis, G., Zelilidis, A. (2005). Deep seismic reflection data from offshore western Greece: a new crustal model for the Ionian Sea. J Pet Geol, 28, pp. 185–202. Krouse, H.R., Viau, C.A., Eliuk, L.S., Ueda, A., Halas, S. (1988). Chemical and isotopic evidence of thermochemical sulfate reduction by light hydrocarbon gases in deep carbonate reservoirs. Nature, 333, pp. 415–419. Kvenvolden, K.A., Lorenson, T.D., Reeburgh, W. (2001). Attention turns to naturally occurring methane seepage. Eos, pp. 82, 457. 154

Leifer, I., Patro, R.K. (2002). The bubble mechanism for methane transport from the shallow seabed to the surface: a review and sensitivity study. Cont Shelf Res, 22, pp. 2409–2428. Leifer, I., Kamerling, M.J., Luyendyk, B.P., Wilson, D.S. (2010). Geologic control of natural marine hydrocarbon seep emissions, Coal Oil Point seep field, California. GeoMar Lett, 30, pp. 331–338. Link, W.K. (1952). Significance of oil and gas seeps in world oil exploration. AAPG (Am Assoc Pet Geol) Bull, 36, pp. 1505–1540. Machel, H.G., Krouse, H.R., Sassen, R. (1995). Products and distinguishing criteria of bacterial and thermochemical sulfate reduction. Appl Geochem, 10, pp. 373–389. Magoon, L.B., Schmoker, J.W. (2000). The total petroleum system — the natural fluid network that constraints the assessment units. U.S. Geological Survey World Petroleum Assessment 2000. Description and results: USGS Digital Data Series, 60. World Energy Assessment Team, p. 31. Marinaro, G., Etiope, G., Gasparoni, F., Furlan, F., Bruni, F. (2011).Gas seepage detection and monitoring at seafloor.10th Offshore Mediterranean Conference and Exhibition, Ravenna, Italy, March 23– 25, 2011, “Frontier exploration” session. Mc Ginnis, D.F., Greinert, J., Artemov, Y., Beaubien, S.E., Wüest, A. (2006). Fate of rising methane bubbles in stratified waters: how much methane reaches the atmosphere? J Geophys Res, 111, C09007. Milkov, A.V. (2011). Worldwide distribution and significance of secondary microbial methane formed during petroleum biodegradation in conventional reservoirs. Org Geochem, 42, pp. 184–207. Noth, S. (1997). High H2S contents and other effects of thermochemical sulfate reduction in deeply buried carbonate reservoirs: a review. Geologische Rundschau 86, pp. 275–287. Obzhirov, A., Shakirov, R., Salyuk, A., Suess, E., Biebow, N., Salomatin, A. (2004). Relations between methane venting, geological structure and seismo-tectonics in the Okhotsk Sea. Geo-Mar Lett, 24, pp. 135–139. 155

Papatheodorou, G., Hasiotis, T., Ferentinos, G. (1993). Gas charged sediments in the Aegean and Ionian Seas, Greece. Mar Geol, 112, pp. 171–184. Rigakis, N., Karakitsios, V., 1998. The source rock horizons of the Ionian Basin (NW Greece). Mar Pet Geol, 15, pp. 593–617. Schmale, O., Greinert, J., Rehder, G. (2005). Methane emission from high-intensity marine gas seeps in the Black Sea into the atmosphere. Geophys Res Lett, 32, L07609. Schneider von Deimling, J., Rehder, G., Greinert, J., McGinnnis, D.F., Boetius, A., Linke, P. (2011). Quantification of seep-related methane gas emissions at Tommeliten, North Sea. Continental Shelf Research 31, pp. 867–878. Stamatakis, M.G., Baltatzis, E.G., Skounakis, S.B. (1987). Sulfate minerals from a mud volcano in the Katakolo area, western Peloponnesus, Greece. Am Mineral, 72, pp. 839–841. Tang, Y., Perry, J.K., Jenden, P.D., Schoell, M. (2000). Mathematical modeling of stable carbon isotope ratios in natural gases. Geochimica et Cosmochimica Acta 64, pp. 2673–2687. US EPA (United States Environmental Protection Agency), 2010.Methane and nitrous oxide emissions from natural sources.US Environmental Protection Agency, EPA 430-R-10-001. Office of Atmospheric Programs, Climate Change Division, Washington DC. Washburn, L., Johnson, C., Gotschalk, C.G., Egland, E.T. (2001). A gas capture buoy for measuring bubbling gas flux in oceans and lakes. J Atmos Ocean Technol, 18, pp. 1411–1420. Yang, J., Zhang, G.L., Zheng, L.X., Zhang, F., Zhao, J. (2010). Seasonal variation of fluxes and distributions of dissolved methane in the north Yellow Sea. Cont Shelf Res, 30, pp. 187–192. Zelilidis, A., Piper, D.J.W., Vakalas, I., Avramidis, P., Getsos, K. (2003). Oil and gas plays in Albania: do equivalent plays exist in Greece? J Pet Geol, 26, pp. 29–48. 156

Zhou, H., Yin, X., Yang, Q., Wang, H., Wu, Z., Bao, S. (2009). Distribution, source and flux of methane in the western Pearl River Estuary and northern South China Sea. Mar Chem, 117, pp. 21–31.

157

6.1.2 Increased methane emission from natural gas seepage at Katakolo Harbour (Western Greece)

Data and figures are published in: Kordella, S., Ciotoli, G., Dimas, X., Papatheodorou, G., Etiope, G. (2020). Increased methane emission from natural gas seepage at Katakolo Harbour (Western Greece). Appl Geochem,116, 104578. https://doi.org/10.1016/j.apgeochem.2020.104578

6.1.2.1 Introduction Katakolo is an area of extensive natural gas seepage, located along the Ionian Coast of Peloponnese in Western Greece (Etiope et al., 2006, 2013). Gas seepage occurs both offshore and onshore, especially within the local tourist harbour. The gas, stemming from deep pressurized reservoirs and migrating along a system of neotectonic faults, is mainly composed of thermogenic methane (CH4), with minor amounts of nitrogen (N2), carbon dioxide (CO2), heavier alkanes

(ethane to exhane) and hydrogen sulphide (H2S), which in some sites reaches hazardous levels, exceeding 1 vol% (Etiope et al., 2013). Katakolo Harbour is visited by more than 200 cruise ships per year, from which hundreds of people disembark daily. The ships generally dock at the main pier, where most of gas is seeping. After preliminary gas flux measurements in a few points in 2004 (Etiope et al., 2006), a first detailed CH4 flux survey at Katakolo Harbour was carried out in 2010. Results showed that gas is seeping to the atmosphere from tens of cracks and fissures in the asphalt pavement throughout the main pier and the parking area (Etiope et al., 2013). Potentially explosive levels of

CH4 and toxic concentrations of H2S accumulate in the ground, posing a severe hazard for humans and tourist infrastructures. A total CH4 emission of 57.3 kg d−1 (21 t y−1) was estimated (Etiope et al., 2013). 158

The asphalt and concrete pavement acts as a cap for gas that gradually accumulates underneath, but it may swell and crack when the gas pressure increases. From 2004 to 2010 we did not observe significant changes in the ground conditions, in terms of number and size of cracks and fissures (Etiope et al., 2013). However, after 2010, a progressive increase of the number of cracks was observed, with higher degradation of ground conditions, including asphalt swelling, in different sites of the harbour. This motivated a new gas flux survey that was performed in November 2018 and its results are presented in this work. We show that the methane emission factor (the average CH4 emission related to seepage) varied considerably over almost a decade (2010–2018) and that it reflects natural seepage variability. We then discuss the potential implications that similar emission factor variability, if occurring in other large seepage areas, may have on global geological CH4 emission estimates. We will caution about the notion that geological gas emissions, a major natural methane atmospheric source (e.g., Etiope and Schwietzke, 2019), do not significantly fluctuate on short-term and do not affect decadal changes of atmospheric methane, as assumed in greenhouse gas budget models (e.g., Sapart et al., 2012; Saunois et al., 2017).

6.1.2.2 Study area and inspection of ground conditions

6.1.2.2.1 Geological setting Katakolo is a small village hosting the second most visited harbour for cruise ships in Greece, situated on a homonymous peninsula in Western Peloponnese. Katakolo has attracted scientific interest due to its extensive onshore and offshore natural gas seepage (Etiope et al., 2006, 2013). It is geologically located in the Ionian zone, one of the external tectonostratigraphic zones of the Hellenides fold-and-thrust 159

belt, which dominates Western Greece (Zelilidis et al., 2003). Katakolo is characterized by active tectonics, with neotectonic faults crossing the base of Neogene sequence (Kokinou et al., 2005, Fig. 5.1.2.1). Salt diapirism related to Triassic evaporites and a N–S trending thrust fault system are responsible for brittle deformations and fracturing (Kamberis et al., 2000). The harbour, in particular, is located at the convergence of onshore and offshore fault systems (Christodoulou, 2010). The faults act as pathways for natural gas upward migration, producing seeps of thermogenic methane-dominant gas (>80 %) with toxic to lethal amounts of H2S (up to 0.5 vol% within the harbour, exceeding 1 vol% in a site, called Faros, outside the harbour; Etiope et al., 2013). The gas originates from a deep petroleum system characterized by Jurassic carbonate reservoirs, Triassic source rocks, and Triassic evaporites (Etiope et al., 2013).

Fig. 5.1.2.1 Location of the study area (encircled areas) and faults within KatakoloHarbour (structural data after Zelilidis et al., 2015; Kamberis et al., 2000; Christodoulou, 2010). Figure from Kordella et al., 2020.

6.1.2.2.2 Description of gas release manifestations The asphalt and concrete pavement throughout the main pier, the parking area next to the pier, and the area in front of the Port Authority Building is characterized by widespread fractures and cracks, generally 160

of sub-meter size. Cracks often develop on the top or along the margins of bulges or asphalt swellings. Along the main pier, several meters long linear fissures developed at the contact between cement and asphalt blocks that characterize the structure of the dock. The contacts between different blocks are likely weak zones where fissures open in response to underground gas pressure and corrosion. Examples of cracks on asphalt swellings and other fissures are shown in Fig. 5.1.2.2 and in additional photos in the Supplementary Material. In September 2010, it was verified that cracks and linear fissures are all points of gas release. At that time, 50 cracks-fissures were observed (Etiope et al., 2013). In November 2018, we observed that throughout the harbour the asphalt and concrete pavement was more pervasively degraded, with a higher number of bulges, swellings, cracks, and fissures: we counted 277 cracks and linear fissures (Fig. 5.1.2.2 and Photos S1-8 in Supplementary Material).

Fig. 5.1.2.2 The three main sectors of Katakolo Harbour affected by cracks and linear fissures: the main pier, parking lot and entrance of the Port Authority Building, and the parking area next to the main pier. Satellite image by Google Earth. Photos: S. Kordella, November 2018. Figure from Kordella et al., 2020.

161

6.1.2.3 Materials and methods Detailed observation on the ground conditions regarding the cracks and fissures on the concrete and asphalt of Katakolo Harbour was carried out on November 2018. Then, methane flux to the atmosphere was assessed with 124 measurements corresponding to 100 cracks-fissures and two manholes, throughout the three sectors of the harbour (Fig. 5.1.2.2, Fig. 5.1.2.3; Photos S1-9 and S12 in Supplementary Material). Methane flux was measured by closed-chamber method, using a

Tunable Diode Laser Absorption Spectroscopy (TDLAS) CH4 detector (accuracy 0.1 ppmv, range from 0.1 ppmv to 100 vol%). Data are transmitted via bluetooth to a smartphone. Gas flux was calculated via linear regression of the gas concentration build-up in the chamber (e.g., Etiope et al., 2013; Etiope et al., 2007). The size of each crack and fissure was measured and considered in the estimate of the methane emission from each individual manifestation. For linear fissures, multiple measurements were performed and the overall emission, in kg d−1, was determined by multiplying the emission calculated for the chamber size (i.e., chamber- size emission factor – CEF, referring to 0.07 m2 and 0.3 m in diameter) by the number of 0.3 m segments along the fissure. For example, the emission of a 3 m long fissure was estimated by multiplying the chamber-size emission factor by 10. Similarly, the output from manholes was estimated by multiplying the measured CEF by the number of 0.3 m segments of fissure bordering the manhole cap (see Photo S12 in Supplementary Material). 162

Fig. 5.1.2.3 Location of methane emission points of the 102 measured cracks, fissures and manholes (124 chamber measurements; see also Photo S1 in the Supplementary Material). Figure from Kordella et al., 2020.

Methane emission from non-measured cracks-fissures was estimated applying the average, measured CEF on the size of the crack; for example, if the crack is smaller than the chamber, the average CEF value is associated to that crack; if the non-measured fissure is longer than the chamber diameter, the average CEF is multiplied by the number of 0.3 m segments along the fissure. The uncertainty of the estimates mainly refers to the error of single measurements (repeated measurements showed fluxes within the same order of magnitude) and subjective attribution of the size of cracks-fissures, which in several cases were characterized by non- linear and irregularly shaped fractures. The data obviously do not take into account the short temporal variability of the flux, which may fluctuate on minutes to hours time scale. Accordingly, the derived emissions shall be considered in terms of order of magnitude. Preliminary test measurements were performed placing the flux chamber on the asphalt in absence of visible cracks. No null CH4 fluxes were observed, which suggests that the asphalt is not totally 163

impermeable. Therefore, an additional survey was performed to evaluate the average flux from diffuse seepage (miniseepage emission factor, MEF), from the unfractured asphalt. This investigation included 28 flux measurements performed over an area of 104 m2, in the parking area in front of the Port Authority Building. The total emission from the diffuse miniseepage was derived by Surfer® 17 (Golden Software, LLC), using radial basis function interpolation method.

6.1.2.4 Results

6.1.2.4.1 Methane emission from cracks and fissures The 2018 survey confirmed that all bulges, cracks and fissures in the asphalt and concrete pavement of the harbour are points of methane emission. Measured methane fluxes and derived output from each crack are reported in Supplementary Tables S1–3, with basic statistics in Table S4 and Fig. S1. Methane output from individual cracks-fissures has a wide range of values (7 orders of magnitude): in 101 cracks- fissures and manholes methane emission ranges from ~0.00005 to ~4

−1 −1 kg d (93% of cracks-fissures release less than ~0.3 kg d of CH4, 4%

−1 −1 release ~0.3–0.9 kg d , and 3% release ~1.8–4.6 kg d of CH4; Table S2-Supplementary Material). One 5 m long fissure (numbered as K137) showed a huge output estimated at about 111 kg d−1 (~40.5 t y−1). This fissure is located in the central part of the main pier, in proximity of the contact between two cement blocks (Fig. 3and Photo S1 in the Supplementary Material). Total methane output from the 102 measured cracks, linear fissures and manholes is then around 125 kg d−1 (~45 t y−1). The mean chamber-size emission factor (excluding the large emission of the K137 fissure) is 0.05 kg d−1. Applying this value to the non-measured cracks- fissures (for which the size was measured; Tables S5–6 in Supplementary Material) results in a total emission of about 100 kg d−1 164

(~36.5 t y−1). Therefore, the overall methane emission from cracks and fissures in the harbour exceeds 200 kg d−1. The results are summarized in Table 5.1.2.1, including a comparison with the data acquired in 2010 (Etiope et al., 2013).

Table 5.1.2.1. Summary of CH4 emissions data from cracks and fissures in Katakolo Harbour. Table from Kordella et al., 2020.

Number of vents Mean CH4 Mean Total CH4 Reference

(cracks) output from CH4 emission individual CEF kg d−1(t y−1) Measured Not vents kg d−1 kg measured d−1

102 0.14a 0.05b 125 (45) 2018 (THIS STUDY) 177 ~100 (36)

Total 225 (82)

28 0.15c 54 (19.7) 2010 (Etiope et al., 2013) 22 3.3 (1.2)

Total 57.3 (21) a,b excluding the large K137 fissure. c excluding a large vent releasing 49.3 kg d−1 CEF: chamber-size emission factor.

6.1.2.4.2 Diffuse methane emission from unfractured asphalt On the unfractured asphalt in front of the Port Authority Building, within an area of 157 m2, 16 measurements over a total of 28 (57%) showed

−2 −1 positive CH4 fluxes (miniseepage), up to 453 mg m d . Radial basis function interpolation of the data results in a total methane emission of 26,665 mg d−1 (0.027 kg d−1) (Fig. 5.1.2.4). The average miniseepage emission factor (MEF, i.e., the total emission divided by the area) is then 170 mg m−2 d−1. Applying this emission factor to the total area of the harbour where miniseepage is expected to occur (7,962 m2, i.e., the 165

three sectors: main pier, parking area, area in front of the Port Authority Building, minus the total area of cracks-fissures; Table 5.1.2.2) leads to a potential diffuse methane degassing of 1.4 kg d−1. This is negligible compared to the overall emission from the cracks/fissures, but further measurements should be performed over a wider area, including ground portions outside the harbour facilities.

−2 −1 Fig. 5.1.2.4. Contour map of the CH4 miniseepage (in mg m day ) from unfractured asphalt. The map was obtained by using Radial Basis Function deterministic interpolator. Figure from Kordella et al., 2020.

166

Table 5.1.2.2. Estimation of potential diffuse degassing (miniseepage) in the Katakolo harbour, based on average MEF (miniseepage emission factor) of 170 mg m−2 d−1. Table from Kordella et al., 2020.

Site Area (m2) Assessed Potential diffuse

diffuse CH4 emission −1 CH4 (kg d ) emission (kg d−1)

Port Authority area with 157 0.018 miniseepage measurements

Port Authority 340 0.06

Parking area 320 0.05

Main pier 7,500 1.27

Area occupied by 198 – vents/cracks

Total 7,962 1.4

6.1.2.5 Discussion

6.1.2.5.1 Seepage variation and possible causes The number of cracks and fissures and the total methane emission from Katakolo Harbour increased considerably from 2010 to 2018. Compared to the previous 2010 survey (Etiope et al., 2013), after 8 years the number of gas release manifestations increased 5.5 times and the methane output increased ~4 times. Many cracks and fissures developed on deformed, swelled asphalt (e.g., see Fig. 5.1.2.2 and Photos S2, S9 in Supplementary Material). Pavement swelling and 167

bulging is frequently observed in gas seepage areas (e.g., Etiope et al., 2011; Etiope et al., 2017), and it is typically due to gas pressure build- up below the asphalt cover, induced by a continuous underground gas outflow. Bulges with circular holes or cracks on the summit represent typical gas plumbing systems. The opening of linear fissures, several meters long, at the contact between cement and asphalt blocks, is a further indication of mechanical stress due to underground gas flow or differential subsidence induced by gas occurrence in the sediments below the structure. Because Katakolo seeps do not release water, pavement swelling due to water pressure must be excluded. Asphalt and concrete could however be also deformed by hydration of secondary minerals or dissolution of salts, induced by infiltration of meteoric water, below the pavement layers. These reactions would anyway occur where gas is released (in fact all bulges are gas vents) and therefore they can be considered “proxies” of gas seepage. So, increased number of deformations means increased seepage. The methane output from the fissure K137, the biggest gas vent, increased from 49 (in 2010) to 111 kg d−1 (in 2018). Outside this fissure, the total methane output (measured and estimated) increased from 8 to 114 kg d−1. We also observed that in fissures that did not change over time, those bordering the main manhole, the gas flux increased considerably: for the bigger manhole, the flux of methane leaking through the metal fissures around the cap (see Photo S12) increased from 172,000 mg m−2 day−1 in 2008, to more than 3,000,000 mg m−2 day−1 in 2018. The manhole provides a good indication of the actual subsoil gas flow variability as its structure and permeability did not change over time. The harbour, initially built for commercial activities in 1850, was upgraded becoming a cruise ship harbour in the mid-90s. This development implied building of a major pier and general restructuring of docks and ground facility areas. The new covering of asphalt and concrete blocks initially acted as a seal for gas migrating upwards, but after a few years, cracks appeared on the ground. The seal degraded 168

over time due to gas pressure and corrosion related to a continuously active seepage. However, from 2004 to 2010 there were no significant changes in the ground conditions, only a few more cracks were observed (Etiope et al., 2013); asphalt swelling and cracking increased considerably only after 2010. This suggests that (a) complete breakthrough of gas migration front, with steady seepage through the ground and pier structure, occurred already in 2004 or before; (b) the degradation due to corrosion, constantly operating, is not a main factor determining cracks and fissures. In conclusion, multiple lines of evidence (i) pavement deformation resulting from mechanical stress; (ii) considerable increase of asphalt swelling and opening of linear fissures only after 2010; (iii) increased exhalation with constant fissure condition; (iv) no significant degradation and cracking between 2004 and 2010 (while corrosion was constantly operating), suggest that the measured increase of methane emission at the surface is actually due to intensification of the flow of gas from depth (seepage), and not to progressive corrosion related to H2S. Therefore, although set in an area impacted by human activity, the observed changes in methane emissions reflect natural variability of gas seepage. We note, then, that a large crack developed on a second pier of the harbour, not affected by seepage, during a 6.6 M earthquake occurred on October 26, 2018 (2 weeks before our survey) with epicentre at about 80 km from the harbour (Photos S10-11 in Supplementary Material). We cannot assess whether that seismic event was determinant for the observed seepage increase, but it demonstrates that fracturing in the harbour's ground, and thus an increase of permeability in the subsurface rocks actually occurs due to earthquakes. The Katakolo region is affected by intense seismicity: it was evaluated that 146 seismic events with magnitude ≥4.5 M, occurred from 1995 (when the harbour was constructed in its modern form) to 2018 within a radius of 100 km (Fig. S1 in Supplementary Material). These earthquakes may have contributed to the variability of the flow of gas from the deep subsurface reservoir. Data are however insufficient 169

to assess whether seismic activity has been influential in the observed increase of methane emission.

6.1.2.5.2 Potential atmospheric impact of seepage variation As discussed above, although the large Katakolo gas seep developed within man-made infrastructures, its seepage variability appears to be natural. The site offers therefore a good example of short-term variation of methane seepage into the atmosphere. In atmospheric CH4 budget models, geological gas sources (amounting globally to about 45 Tg y−1, i.e. ~8% of total methane sources; Etiope et al., 2019; Etiope and Schwietzke, 2019) are instead generally considered constant and not affecting atmosphere CH4 decadal scale variability (e.g., Sapart et al., 2012;Saunois et al., 2017). However, geological records and present- day monitoring suggest that, due to variations of subsurface gas pressure gradients and rock permeability, gas seepage is variable over different time scales (e.g., Etiope et al., 2008; Svensen et al., 2004; Etiope and Schwietzke, 2019). Considering the total area of the three Katakolo harbour sectors (15,000 m2), the overall emission factor from crack-fissures increased

−2 −1 from 4000 to 15,000 t CH4 km y in less than 10 years. This is a substantial variation. A simple numerical exercise can be made to estimate the effect that such an emission factor variation may have on global geo-methane emission, if it occurs in larger seeps. Being speculative, the exercise has only the scope of showing the potential impact of seepage CH4 emission factor variability. Areal emission factors are known for two major geological methane sources (Etiope et al., 2019): (a) mud volcanoes (ranging from 102 to 104 t km−2 y−1depending on the mud volcano size and activity, mean value of

−2 −1 2 ~4,100 t km y , over 680 km , leading to a global CH4 output of ~3 Tg y−1, eruptions excluded) and (b) microseepage, the diffuse gas exhalation from the ground that is independent from macro-seeps 170

(median value of ~3 mg m−2 d−1, i.e., ~1,000 t km−2 y−1, leading to a global output of 10–25 Tg y−1). If the emission factor increases four times, as in Katakolo, for only ten large mud volcanoes in Azerbaijan (among the “big emitters” described in Etiope et al., 2019), their total methane emission would increase from ~0.9 to 3.6 Tg y−1, adding 2.7 Tg y−1 to the atmospheric methane budget. A four-fold emission factor increase for all mud volcanoes would add 9 Tg y−1. A similar increase in microseepage emission factor for 10 % of global microseepage area (potentially ~13 million km2) would add 4–10 Tg y−1. Global variation of atmospheric methane budget over the period from 2002 to 2006 to 2008–2012 was assessed at ~22 (16–32) Tg y−1 (Saunois et al., 2017). Then, our numerical exercise suggests that emission factor variations of the same order of magnitude and timing as observed at Katakolo, if occurring in large seepage zones on Earth, have the potential to contribute to decadal atmospheric methane changes.

6.1.2.6 Summary and conclusions Gas flux measurements by accumulation chamber performed in the Katakolo Harbour in 2018 revealed considerable increase of methane seepage compared to a previous study carried out in 2010 at the same site (Etiope et al., 2013). With a five-fold increase in the number of gas manifestation spots (cracks and fissures on the asphalt and concrete pavement throughout the harbour), methane emission increased from 57 to 225 kg d−1 (~4 times), with emission factor changing from ~4,000 to 15,000 t km−2 y−1. Multiple lines of evidence (mechanical deformation and fissuring of asphalt pavement, increased exhalation with constant fissure conditions, no significant cracking with operating corrosion from 2004 to 2010) suggest that the observed emission changes reflect natural, subsoil seepage variability, and are not particularly influenced by progressive chemical corrosion and degradation of the asphalt. 171

It is noted that Katakolo is affected by intense seismicity, which can enhance fracturing in the harbour subsurface rocks, as evidenced by a large crack formed after an earthquake occurred 80 km from the harbour, a few weeks before our survey. In Katakolo, gas is seeping also at the harbour offshore and at a seep named “Faros” located 0.5 km northward from the harbour; measurements performed in 2010 indicated emissions of ~90–329 kg d−1 (~33–120 t y−1) and ~186 kg d−1 (~68 t y−1), respectively (Etiope et al., 2013). It is likely, however, that seepage from these sites also changed over time. Adding the 2018 onshore harbour emission estimate (~226 kg d−1 or ~82 t y−1) to the 2010 seepage in the other sites results in a total methane output of about 502–741 kg d−1 (~183– 270 t y−1). Katakolo is one of the largest natural gas seepage sites in Europe. Katakolo is, then, an example of short-term variation of methane seepage into the atmosphere. We estimated that if similar sub-decadal, four-fold variations of seepage emission factor occur on large seepage areas on Earth, decadal changes of atmospheric methane budget can be affected. The notion of geological emissions as constant source for the atmospheric methane, an assumption often used in atmospheric methane budget models (e.g., Sapart et al., 2012; Saunois et al., 2017), is at least questionable. Accordingly, the emission factors of large seeps reported in the global inventories should be investigated also over time.

References Christodoulou, D. (2010). Geophysical, Sedimentological Study: Remote Sensing on Pockmarks in Seismogenic Active Areas. PhD Thesis.University of Patras, Patras. Etiope, G., Christodoulou, D., Kordella, S., Marinaro, G., Papatheodorou, G. (2013). Offshore and onshore seepage of 172

thermogenic gas at Katakolo Bay (Western Greece). Chem Geol, 339, pp. 115–126. https://doi.org/10.1016/j.chemgeo.2012.08.011. Etiope, G., Ciotoli, G., Schwietzke, S., Schoell, M. (2019). Gridded maps of geological methane emissions and their isotopic signature. Earth Syst Sci Data, 11, pp. 1–22. https://doi.org/10.5194/essd-11-1- 2019 Etiope, G., Doezema, L., Pacheco, C. (2017). Emission of methane and heavier alkanes from the La Brea tar Pits seepage area, Los Angeles. J Geophys Res Atmos, 122 (12) https://doi.org/10.1002/2017JD027675, 008-12,019 Etiope, G., Martinelli, G., Caracausi, A., Italiano, F., 2007. Methane seeps and mud volcanoes in Italy: gas origin, fractionation and emission to the atmosphere. Geophys Res Lett, v34, L14303.https://doi.org/10.1029/2007GL030341 Etiope, G., Milkov, A.V., Derbyshire, E. (2008). Did geologic emissions of methane play any role in Quaternary climate change? Global Planet Change, 61, 79–88. https:// doi.org/10.1016/j.gloplacha.2007.08.008 Etiope, G., Nakada, R., Tanaka, K., Yoshida, N. (2011). Gas seepage from Tokamachi mud volcanoes, onshore Niigata Basin (Japan): origin, post-genetic alterations and CH4-CO2 fluxes. Appl Geochem, 26, 348– 359. https://doi.org/10.1016/j. apgeochem.2010.12.008. Etiope, G., Papatheodorou, G., Christodoulou, D., Ferentinos, G., Sokos, E., Favali, P., 2006. Methane and hydrogen sulfide seepage in the northwest Peloponnesus petroliferous basin (Greece): origin and geohazard. AAPG (Am Assoc Pet Geol) Bull, 90 (5), pp. 701–713. https://doi.org/10.1306/11170505089. Etiope, G., Schwietzke, S., (2019). Global geological methane emissions: an update of topdown and bottom-up estimates. Elem Sci Anth, 7, 47. https://doi.org/10.1525/ elementa.383 Kamberis, E., Rigakis, N., Tsaila-Monopoly, S., Ioakim, C., Sotiropoulos, S. (2000). Shallow biogenic gas-accumulation in Late Cenozoic sands of Katakolon peninsula. Western Greece. Geol Soc Greece Spec Publ, 9, pp. 121–138. 173

Kokinou, E., Kamberis, E., Vafidis, A., Monopolis, D., Ananiadis, G., Zelilidis, A. (2005). Deep seismic reflection data from offshore Western Greece: a new crustal model for the Ionian Sea. J. Petrol. Geol. 28 (2), pp. 185–202. https://doi.org/10.1111/j.1747- 5457.2005.tb00079.x Saunois, M., Bousquet, P., Poulter, B., et al. (2017). Variability and quasi-decadal changes in the methane budget over the period 2000– 2012. Atmos Chem Phys, 17 (11) https://doi.org/10.5194/acp-17-11135- 2017, 135-11,161 Sapart, C.J., Monteil, G., Prokopiou, M., van de Wal, R.S.W., Kaplan, J.O., Sperlich, P., Krumhardt, K.M., van der Veen, C., Houweling, S., Krol, M.C., Blunier, T., Sowers, T., Martinerie, P., Witrant, E., Dahl- Jensen, D., Rockmann, T. (2012). Natural and anthropogenic variations in methane sources during the past two millennia. Nature, 490 (7418), pp. 85–88. https://doi.org/10.1038/nature11461. Svensen, H., Planke, S., Malthe-Sørenssen, A., Jamtveit, B., Myklebust, R., Eidem, T., Rey, S.S. (2004). Release of methane from a volcanic basin as a mechanism for initial Eocene global warming. Nature, 429, 542–545. https://doi.org/10.1038/ nature02566 Zelilidis, A., Piper, D.J.W., Vakalas, I., Avramidis, P., Getsos, K. (2003). Oil and gas plays in Albania: do equivalent plays exist in Greece? J Petrol Geol, 26 (1), pp. 29–48. https://doi.org/10.1111/j.1747- 5457.2003.tb00016.x Zelilidis, A., Maravelis, A.G., Tserolas, P., Konstantopoulos, P.A. (2015). An overview of the petroleum systems in the Ionian zone, onshore NW Greece and Albania. J Petrol Geol, 38, pp. 331–348. https://doi.org/10.1111/jpg.12614

174

6.1.3 Long-term Offshore Monitoring in Katakolo Bay. The Gas Monitoring Module In the framework of EU HYPOX project, a benthic station, named Gas Monitoring Module (GMM; Fig. 5.1.3.1; Marinaro et al., 2006; Marinaro et al., 2011) was used in Katakolo Harbour, that is heavily affected by intense gas seepage (Etiope et al, 2006; Etiope et al., 2013). The objective of this study was to explore the inter-relationships between CH4 and O2, and other environmental parameters, i.e., seawater temperature, turbidity, currents, pressure, meteo data, earthquakes, etc. As far as we know, this was the first long-term monitoring experiment investigating O2 vs CH4 in the seawater in a thermogenic gas seepage field (Etiope et al., 2013) with strong open sea influence.

Fig. 5.1.3.1 Members of INGV and the Laboratory of Marine Geology and Physical Oceanography, University of Patras that participated in HYPOX project aboard the research vessel with the GMM. 175

6.1.3.1 Materials and methods

6.1.3.1.1 The GMM The GMM is based on a multi-parametric approach in which the detection of gases (O2, CH4, H2S) is associated with that of key physicochemical factors, i.e. Temperature, pressure, and conductivity. Gas detection is based on the use of oxygen, methane, and hydrogen sulfide sensors commercially available. A turbidimeter, a CTD, a current meter, and electronic and battery vessels completed the payload. All sensors have a unique time reference and are controlled by a dedicated data-acquisition system (Fig. 5.1.3.2).

Fig. 5.1.3.2 Left: Graphic representation and photo from the GMM deployment in Katakolo Harbour. Right: The GMM payload.

GMM is a state-of-the-art observatory specifically designed by INGV and Tecnomare for long-term gas monitoring on the seafloor. Originally it was developed in the framework of the European Commission ASSEM project (Array of Sensors for long term SEabed Monitoring of geohazards; Marinaro et al. 2006). Since then, GMM has been used in Patras Gulf (2004-2005), offshore Panarea (Aeolian Islands), and after upgrade work for payload extension, in Katakolo Harbour (2010-2011). The GMM accommodates a series of sensors, for the abovementioned gas concentrations and physicochemical parameters, controlled and managed by a data acquisition and control system. GMM 176

main technical characteristics as it was used in this study (Marinaro et al., 2011) are summarized in Table 5.1.3.1. GMM in Katakolo Harbour was configured with the following payload (Fig. 5.1.3.2):

• two semiconductor METhane (CH4), sensors Range: 50 nmol/l to 10 mmol/l, accuracy: 5-15% (METS; Franatech GmbH, Luneburg, Germany) • an oxygen sensor • a turbidimeter

• an H2S electrode microsensor (ATM GmbH, Rostock, Germany) • a CTD SBE-37-SI Microcat (SeaBird) for measurements of conductivity (C), temperature (T) and depth (D, or pressure, P). Range: Temperature: -5 to 35 °C, Conductivity: 0-7 S/m, Depth: 3,500 m, Accuracy: Temperature: 0.002 °C, Conductivity: 0.0003 S/m, Depth: 0.1% of full scale • a current meter • an electronic vessel • a battery vessel GMM acquisition and control systems (Fig. 5.1.3.3) can perform: • data acquisition from all sensors • preparation and continuous update of data messages • management (e.g., data request, system reconfiguration, restart) of the sensors by remote operator • switch on/off of individual sensors • reception/management (system reconfiguration, data request, etc.) of commands • monitor technical parameters (status and power supply, etc.) • back-up of data in internal memory • processing data (quality check, occurrence of events)

177

Fig. 5.1.3.3 The GMM acquisition and control systems unmounted.

Before the deployment of the GMM in Katakolo Bay, INGV and Tecnomare tested the two commercially available underwater methane sensors (Franatech METS and Contros Hydro-C) in controlled conditions. The purpose was to verify their reaction time, response to temperature variations and water turbulence, both in presence and in absence of dissolved methane (Marinaro et al., 2011). Based on the results, a pump and a flow-through chamber were adopted, for constant water flow in contact with the sensitive sensor membrane, so as to avoid bio-fouling and hence increment long-term autonomy. For the Katakolo Bay mission (Fig. 5.1.3.4), the GMM payload was updated (Table 5.1.3.1) from the previous missions (Marinaro et al., 2006) so that it operated in autonomously (battery powered, internal data storage (Fig. 5.1.3.3). Two methane sensors were adopted (see Fig. 5.1.3.2), operating in parallel; one was fitted with pump and flow- through chamber, while the other was directly exposed to the environment. Continuous monitoring with the GMM mainly aimed at detecting gas seepage variations and their impact in the oxygen budget of the seawater. The methane sensors configuration from Katakolo is shown in Fig. 5.1.3.2.

178

Table 5.1.3.1 The GMM main technical characteristics (from Marinaro et al. 2011).

Fig. 5.1.3.4 Left, center: The Benthic station GMM aboard the research vessel (Barbantonis) at Katakolo Bay. Right: The GMM recovery from Katakolo Harbour. Photos: Stavroula Kordella, Margarita Iatrou. 179

6.1.3.1.2 GMM deployment site The GMM was deployed within a site of gas bubble plumes inside Katakolo Harbour, for long-term monitoring, on the 21st of September 2010 (Fig. 5.1.3.4). The monitoring lasted 101 days, until the 31st of December 2010. The GMM deployment site (Fig. 5.1.3.5) was selected based on the results of the geophysical, geochemical and areal survey that was performed offshore Katakolo and is described above in detail (see Chapter 5.1.1; Etiope et al., 2013). Katakolo Bay was chosen for this experiment due to the extensive submarine seepage field located in the Harbour (see Chapter 5.1.1; Etiope et al., 2013), which can be considered as a unique natural laboratory to study O2 versus gas seepage by long term monitoring. In the marine area of Katakolo, natural gas is seeping from shallow (<-11 m depth) seabed while creating bacterial mats (Beggiatoa spp.), and the bubbles issuing from the seabed are up to the order of 10-20 cm in diameter (Fig. 5.1.3.6). The origin of gas was assessed in 2006 and 2010 (see Chapter 5.1.1; Etiope et al., 2013; 2006). It is a thermogenic methane-dominant gas

(>80 %) that includes toxic and lethal amounts of H2S (up to 0.5 vol. %). For the long-term monitoring in Katakolo bay the Gas Monitoring

Module (GMM) was deployed (Fig. 5.1.3.4) equipped with a series of gas and oceanographic sensors (Fig. 5.1.3.2), in order to study the methane release from bubble plumes, its temporal variations, its relationships with oceanographic parameters, meteorological factors, seismic activity, and its effect on oxygen distribution in the seawater (Marinaro et al., 2006).

180

Fig. 5.1.3.5 Location of the GMM deployment site in Katakolo Harbour.

The site was constituted ideal for the GMM deployment, based on its specific charecteristics: -7 m water depth: shallow enough to ensure a safe deployment/recovery, deep enough to monitor oxygen fluctuations and other natural proceedures. -Inside a harbour with open sea influence: ensures protection from extereme weather conditionsbut is also affected by the meteorologic conditions and currents. -Last but not least, the particular site is affected by vigorous gas seepage; several seeps within few square meters at this site release methane continuously (Fig. 5.1.3.6).

Fig. 5.1.3.6 Photos of underwater seeps at the location of GMM deployment in Katakolo Harbour. Photos: Nikos Nikolopoulos. 181

6.1.3.1.3 GMM data preparation and processing The GMM data was examined in detail for variations and correlations among variables derived from each sensor in combination with meteo data, at the different, original time-scales, and at different time-windows (seconds, minutes, hour-s, day-s time window, etc.). For this, diagrams of all data in original time-scales were produced, and all the variations of oxygen and methane, which are the key variables in this study, in conjunction with all other variables, were noted in detail in an excel sheet, for grouping and quantification of the events (Fig.5.1.3.7). The purpose of that procedure was to identify the possible trends and natural processes, as well as reveal new relationships or links between methane seepage and other parameters in the marine environment. This method, although time-consuming and graphic, was considered an important first step so at to identify patterns and trends to be further investigated with multivariate statistics. The abovementioned results were used as a basis for the multivariate statistical analyses and other statistical experiments.

3 4 5 6 7

Fig. 5.1.3.7 Graphic representation of the 24 h GMM, meteo, and current data and the transcription of the variations of oxygen and methane in conjunction with all the other variables in an excel sheet, for grouping and quantification of the events.

The GMM dataset was also analyzed using Principal Component Analysis; a multivariate data analysis tool. R-mode Principal Component Analysis with Varimax Rotation was applied to analyze the 182

inter-relationships within a date set of the 7 selected variables. The selection of variables was determined by the prior graphic analysis. The 7 variables that were selected were the ones that appeared to be interacting throughout the data set. The following criteria were used for the selection of the number of factors: (i) Cattel's scree test (ii) the variance explained criterion emphasizing on the minimum number of factors that explain the variance of the data set (iii) the communalities of each variable (>0.5). As a result, a 4-factor solution was selected to explain the variability of the original data set (Papatheodorou et al., 2006). Finally, data was curated with MATLAB© for plots containing the complete data sets, as well as for investigation of auto-correlations, to examine periodicity, and cross-correlations between variables in different time lags.

6.1.3.2 Results The GMM successfully performed a continuous, long-term (101 days) monitoring in a methane seepage, open-sea-influenced site in Katakolo Harbour, at ~ -7 m water depth. The complete sequence of physico- chemical data recorded by GMM is presented in Fig. 5.1.3.8. The GMM data set provided complex and multi-parametric infromation on seawater hypoxia in relation to intense methane seepage phenomena. A huge amount of data with very high sampling frequency was collected during the three-month period (O2, CH4, H2S: 1 sample/5 sec; CTD: 1 sample/min; currents: 2 Hz). The detailed examination of the variations/correlation among sensors at the different, original temporal scales of each sensor, and at different time windows (seconds, mininutes, hour-s, day-s, etc.), provided a deep understanting of the data trends and the identification of the natural processes behind them. Those results, were used as a basis for the multivariate statistical analyses and other . The results from the correlation using descriptive statistics of the physico-chemical data collected by the GMM of Fig. 183

5.1.3.8 with the wind speed data, that were collected from a local meteo station, are described below. Over the 101 days of monitoring, 8 main oxygen decreasing periods took place (Fig. 5.1.3.9). During these periods, oxygen decreased from 265 μΜ reaching hypoxic (<62.5 μΜ) up to of quasi- anoxic levels (down to 10 μΜ). Short-term oxygen variations occurred whithin these periods, ranging from 265 to 10 μM. The 8 main oxygen decreasing periods were investigated in detail, so as to identify the processes behind the short- and long-term variations. No effect of seismic activity on the GMM dataset was detected. This is because the seismic events during the monitoring period were few and did not exseed 3.5 M, which is probably too weak to invoke any methane flux variations. 184

Fig. 5.1.3.8 Complete sequence of data recorded by GMM within Katakolo Harbour.

185

Fig. 5.1.3.9 Dissolved oxygen concentration during the 101 days monitoring period. The 8 main O2 decreasing periods are marked.

In the 8 long-term oxygen decreasing periods (Fig. 5.1.3.9), 54 major events of hypoxia/anoxia (episodes of significant dissolved oxygen decrease) were recognized in the data sequence. In the same periods,

43 short term CH4 peaks (up to the order of ~2 μM in 1.5 to ~6 h) were identified, which in more than 95% of cases were followed by O2 decrease. Among these 54 hypoxia/anoxia events two main patterns (A & B) occurred, representing contrasting natural processes. Pattern A (interruption of hypoxia/anoxia) (Fig. 5.1.3.10) where sharp peaks of current speed (variations of ~2.5 – 36 cm/s) are coupled with: (i) sharp O2 increase (from hypoxic and quasi-anoxic levels to ~80- 265 μΜ) (ii) turbidity peaks (in 80 % of cases) (iii) Temperature sharp variations (in 60 % of cases). 186

Fig. 4.1.3.10 Examples of the pattern A that was identified in the GMM data sequence.

Sharp peaks of current speed are coupled with: (i) sharp O2 increase, (ii) turbidity peaks (in 80% of cases) (iii) Temperature sharp variations (in 60% of cases).

Pattern B (methane peaks/oxygen depletion) (Fig. 5.1.3.11) is characterized by relatively steady and low current speed (<4 cm/s) for a considerable time (normally 4-6 hours or longer). In pattern B, low current activity is coupled with elevated dissolved CH4 concentration, leading to dissolved O2 concentration dropping and lingering at hypoxic levels, and even reaching quasi-anoxic levels (10 μΜ). The longest period of hypoxia and quasi-anoxia was ~45 h, and it was detected on 20-22 November (Fig. 5.1.3.12). 187

Fig. 5.1.3.11 Example of the pattern B that was identified in the GMM data sequence. In pattern B, the current speed is low and steady for several hours, then CH4 concentration rises and O2 lingers at hypoxic levels, and even reaches quasi-anoxic levels.

From both patterns A (Fig. 5.1.3.10) and B (Fig. 5.1.3.11) detected in the GMM data sequence, it is understood that as long the current speed is low and steady, disolved oxygen concentration drops and lingers in hypoxic, and anoxic levels, even in such a site which is highly influenced by open sea. Due to the site location (open-sea- influcenced), as soon as the current speed increases, there is a simultaneous rise of the disolved oxygen concentration levels, which rapidly returns to the normal levels for open sea environment, even when CH4 concentration is eleveted (Fig. 5.1.3.12). This process is attributed either to the rapid O2 oxidation of reducing gases (i.e., CH4 and H2S) or to gas exchange between methane bubbles and gases disolved in the seawater, namely N2 and O2 (Fig. 5.1.3.13; Mc Ginnis et al., 2006; Etiope, 2015). 188

Fig. 5.1.3.12 The longest period of hypoxia is from 20/11, 08:00 am to 22/11, 05:00, resulting in almost two days of continuous hypoxia (Pattern B). When current speed increased again, O2 also increased (pattern A). 189

Wind speed, especially the east-west direction is, in all cases, directly correlated with current speed and turbidity, making it clear that the currents are wind-driven. Therefore, it is evident that the GMM site is strongly affected by wind-driven currents. When wind/current activity is low, there is no mixing of the water column and thus no input of atmospheric oxygen in the water column. Then, dissolved oxygen in the seawater is removed either by gas exchange (Fig.4.1.3.11) or consumed for CH4 and H2S oxidation, causing enhanced oxygen depletion for as long as the current/wind activity, and thus the fresh oxygenated water supply, remain low.

O2 CH4 (+N2) CH4/O2 gradient

Katakolo fault

Fig. 5.1.3.13 Illustration of the oxygen decreease around a gas bubble plume. Diagram of gas fraction with water depth, based on the by Mc Ginnis et al., 2006 model of gases exchange between seawater and 0.5 cm diameter bubbles. When seepage bubbles ascending from the seabed rise at 70-80 m, they are not comprised of methane anymore. Methane has been gradually replaced by oxygen and nitrogen that were dissolved in the seawater. Figure modified from Etiope, 2015.

Autocorrelation and cross-correlation Linear correlation between different time steps or lags within the same variable, when applied to all parameters (O2, CH4, turbidity, pressure, etc.) showed no periodicity. Exception is the pressure; whose autocorrelation showed the periodicity of tide Fig. 5.1.3.14. Cross- correlation between variables at different time lags showed no important results. As it was expected from the visual graphic examination of the 190

GMM data sequence, the variables do not have any periodicity and also there is no periodic correlation between them.

Fig. 5.1.3.14 Diagrams of oxygen, methane, and pressure autocorrelation with different time lags.

The four-factor solution was selected according to the abovementioned combination of criteria for number of factors selection. These 4 factors 191

explained 72% of the total variance of the data set, and the communalities of the 7 variables were moderate high and high (>0.50). Therefore, this 4-factor model may indicate the dominant trends between the selected 7 variables without losing significant information. The first factor (Table 5.1.3.1), that explains 19% of the total variance, showed a negative correlation between methane concentration (loading -0.844) and temperature (loading 0.783). This is could be since porewater in the sediments, which is colder than the supernatant seawater, is released in the water column due to the mechanical disturbance caused by methane bubbling. As a result, gas seepages are characterized by slightly lower temperature compared to the temperature of the surrounding seawater. The second factor (Table 5.1.3.1), that explains 18% of the total variance, showed a strong positive correlation between turbidity (loading 0.750) and the N-S direction current (loading 0.840), which is the current that mainly affects the monitoring site. This was also observed easily in the data graphs, and it is also self-evident that at the relatively shallow depth of the site (-7 m) when the winds driven currents are strong the seabed sediments are affected (resuspension). The third factor (Table 5.1.3.1), that explains 18% of the total variance, confirmed the abovementioned pattern A. This factor showed positive correlation of dissolved oxygen concentration (loading 0.870) with turbidity (loading 0.428) as indicated by the positive loadings of both variables. The interpretation of this factor, in terms of natural processes is that elevated turbidity is the result of strong winds/currents at the shallow site, resulting also in continuous oxygenation of the water column, making it impossible to be consumed by seeping CH4. The slight negative loading of the CH4 (-0.361) further support the interpretation that the Factor 3 represents the Pattern A. The slight negative loading of the temperature (-0.383) is probably related with the flow of refreshed water in the GMM site due to N-S currents. 192

The fourth factor (17% of total variance) shows that E-W and z direction current speed do not affect the physicochemical parameters of the water column in the GMM site (Table 5.1.3.1).

Table 4.1.3.1. Rotated component matrix of Factor Analysis. Extraction method: Principal Component Analysis. Rotation: Varimax with Kaiser normalization. Rotation converged in 5 iterations. Variables Components 1 2 3 4

O2 0.021 -0.04 0.870 0.140 Temperature 0.783 -0.072 -0.383 0.121 Turbidity -0.069 0.750 0.428 -0.120

CH4 -0.844 -0.057 -0.316 0.008 N-S current 0.049 0.840 -0.248 0.207 E-W current 0.024 0.016 0.052 0.736 Z current 0.054 0.065 0.047 0.764 Variance 19 18 18 17 72 explained (%)

6.1.3.3 Conclusions The GMM performed sucessfully 101 days of continuous monitoring in a thermogenic methane seepage site (Etiope et al., 2013). During the monitoring period, strong dissolved oxygen variations were recorded, even up to 200 μΜ. Thus, hypoxia and quasi-anoxia was observed at the deployment site, which was shallow (-7 m water depth) and a strong open sea influence. The hypoxic events seem to be linked to a combination of increased methane seepage and low current activity (Pattern B), whilst when current activity increases dissolved oxygen concentration is rapidly restored (Pattern A) and the seawater is again well oxygenated. 193

References Etiope G., Papatheodorou G., Christodoulou D., Favali P., Ferentinos G., E. Sokos (2006). Methane and hydrogen sulfide seepage in the NW Peloponnesus petroliferous basin (Greece): origin and geohazard. AAPG Bull, 90(5), pp. 701-713. Etiope, G., Christodoulou, D., Kordella, S., Marinaro, G., Papatheodorou, G. (2013). Offshore and onshore seepage of thermogenic gas at Katakolo Bay (Western Greece). Chem Geol, 339, 115–126. https://doi.org/10.1016/j.chemgeo.2012.08.011. Marinaro, G., Etiope, G., Lo Bue, N., Favali, P., Papatheodorou, G., Christodoulou, D., Furlan, F., Gasparoni, F., Ferentinos, G., and Masson, M. (2006). Monitoring of a methane-seeping pockmark by cabled benthic observatory (Patras Gulf, Greece) Geo-Mar Lett, 26 (5), pp. 297-302. Papatheodorou, G., Demopoulou, G. and Lambrakis, N. (2006). A long- term study of temporal hydrochemical data in a shallow lake using multivariate statistical techniques. Ecological Modelling, 193, 759–776.

194

6.2 Amvrakikos Gulf

6.2.1 Gas seepage-induced features in the hypoxic/anoxic shallow marine environment of Amfilochia Bay, Amvrakikos Gulf (Western Greece)

Data and figures are published in: Kordella, S.; Christodoulou, D.; Fakiris, E.; Geraga, M.; Kokkalas, S.; Marinaro, G.; Iatrou, M.; Ferentinos, G.; Papatheodorou, G. Gas Seepage-Induced Features in the Hypoxic/Anoxic, Shallow, Marine Environment of Amfilochia Bay, Amvrakikos Gulf (Western Greece). Geosciences 2021, 11, 27. https://doi.org/10.3390/ geosciences11010027

6.2.1.1 Introduction Amfilochia Bay is located in the eastern-southernmost margin of Amvrakikos Gulf, at the north-western coast of Greece. Amvrakikos Gulf has been widely studied because of its specific features, i.e., a high ecological value combined with a particular oceanographic regime and intense environmental pressures, that constitute it both valuable and fragile (e.g., Diamantopoulou et al., 2018; Papaefthymiou et al., 2013; Tsabaris et al., 2012; Ferentinos et al., 2010; Tsangaris et al., 2010; Kormas et al., 2001, Piper et al., 1982; Friligos et al., 1977). Amvrakikos is a shallow (<65 m), semi-enclosed, narrow (~35 km long and ~6 to 15 km wide) gulf, with a strongly stratified water column, and a seasonally anoxic bottom layer (Ferentinos et al., 2010). On the west, Amvrakikos Gulf is connected to the Ionian Sea through a narrow, elongated (~8 m deep, 6 km long, ~0.8-2 km wide) channel. Amvrakikos Gulf has been characterized as the only Mediterranean fjord-type system (Ferentinos et al., 2010; Kountoura and Zacharias, 2011) due to 195

its fjord-like water circulation and morphology. The Gulf’s water column is stratified in two layers, a surface layer with brackish water that is out- flowing and a dense, cold, saline water bottom layer that is inflowing from the narrrow channel that connects the Gulf to the open sea. The channel oxygenates seasonaly the western part of the Gulf. This density driven stratification of the water column strongly affects the vertical distribution of the dissolved oxygen, leading to hypoxia and seasonal anoxia at the bottom layer, particularly in the eastern, innermost part of the Gulf, where the Amfilochia Bay is located (Ferentinos et al., 2010). Based on studies that took place in the early 2010’s, anoxic conditions in Amvrakikos Gulf appeared during the 1980’s decade (Tsabaris et al., 2009; Ferentinos et. al., 2010, Tsabaris et al., 2012). The hypoxic/anoxic environment of Amvrakikos Gulf has caused several fish mortality events, with the most recent and massive having occurred in February 2008, when a mortality of ~900 t of fish was reported north of the Amfilochia Bay (Ferentinos et al., 2010; Friedrich et al., 2013). Moreover, release of enormous amounts of sulfur from a submarine “volcano” in the Bay, triggered by earthquakes and causing massive fish mortality events, have been reported on November 1847 and February 1866 by Miaoulis (1876). Although studies, have reported the presence of gas-charged marine sediments (Papatheodorou et al., 1993; Kapsimalis et al., 2005) and anoxic/hypoxic conditions (Ferentinos et al., 2010, Kountoura and Zacharias, 2011) in Amvrakikos Gulf, no attention has been given so far regarding the existense of gas-bearing geomorphological features and gas seepage in the water column, in relation to the anoxia/hypoxia in Amfilochia Bay. The purpose of this work was to study Amfilochia Bay by using a combination of geophysical and geochemical means combined with a multi-parameter platform. To this end, the seabed was explored for seepage-related geomorphological features. Moreover, gas seepage

(CH4 and H2S) from the seafloor was investigated, so as to evaluate 196

possible geological hypoxia/anoxia drivers, and the origin of the seeping gas was assessed. Finally, the possible link between oxygen deficiency (concentration decrease, oxycline variations, hypoxia/anoxia) and gas seepages was examined.

6.2.1.2 Geological and physiographic setting of survey area Western Greece is dominated by a fairly complex geological history and structure characterized by along strike changes in the subduction zone composition, geometry, and stress regime (Kokkalas et al., 2006 and references therein). Several studies suggest that the area of Western Greece, and specifically the Aitolo-Akarnania region forms a unique and rigid crustal block with different strain rates, GPS estimated velocities, crustal rotation and level of seismicity compared to the surrounding crustal blocks (Le Pichon et al., 1995; Hollenstein et al. 2008; Vassilakis et al., 2011; Perouse et al., 2017). This crustal block is bounded to the west by the Kephalonia transform fault (KTF), to the north by the Amvrakikos half-graben and to the east by the Katouna (KFZ) and Amfilochia (AFZ) fault zones (Fig. 5.2.1.1-2). The Katouna-Amfilochia F.Z. forms a composite NW-trending left- lateral fault zone, with a length of approximately 50 km, and is considered as one of the major tectonic features in the broader area (Fig. 5.2.1.1, 2; Clews, 1989; Underhill, 1989, Haddad et al., 2020). The KFZ is highly segmented and along the left-stepping fault segments the relay zones host several elongated depressions along the Amfilochia valley, which are filled with sediments or morphotectonic controlled lakes. Although the geomorphological expression of the Katouna and Amfilochia F.Z is impressive, its left-lateral kinematics comes mainly from seismological data (Hatzfeld et al., 1995) and GPS estimated sense of motion (Hollenstein et al. 2008; Ganas et al., 2009). Focal mechanisms from the region around the fault system are consistent with 197

N-S extension and left-slip on steep NW-trending faults (Fig. 5.2.1.1). Rock types exposed near the fault zone are primarily Oligocene-Early Miocene Ionian flysch to the east, while to the west and beneath the Amfilochia valley the Triassic evaporites have been remobilized in Plio- Quaternary time. This salt tectonics deformation produced small depressions filled with sediments and gypsum (Underhill, 1988; Vott et al., 2009) which can look similar to pull-apart structures formed by oblique strike-slip motion along the fault zone (Fig. 5.2.1.1). Towards the north the Amfilochia fault zone coincides with the eastern edge of the Amvrakikos Basin, which is an active morphologic depression bounded by roughly E-W trending normal faults that control the Upper Pliocene to recent sedimentation (Fig. 5.2.1.1). In the isotopic stages of MIS3 and MIS2 (ca. 50 to 11 ka BP), when the sea level was 55m below the current level, the western part of the gulf was shallow, while its eastern part was occupied by a lake (Kapsimalis et al., 2005). The invocation of the sea took place at 11 ka BP and the bay took its present form in 4 ka BP (Kapsimalis et al., 2005). From an environmental aspect, Amvrakikos is one of the most important Gulfs of Greece, due its ecological richness and productivity. In Amvrakikos Gulf the Pleistocene/Holocene boundary seems to be a gas accumulation horizon. The Holocene sequence is characterized by anomalous acoustic characters indicative of gas charged sediments and upward gas migration features (ATZ, gas pockets, gas plumes, doming). Moreover, the central basin of the Gulf is occupied by buried pockmarks developed on the Pleistocene/Holocene gas accumulation interface, (Papatheodorou et al., 1993; Kapsimalis et al., 2005). The study area, Amfilochia Bay, is located at the east-southeast end of Amvrakikos Gulf. It is a narrow and elongated bay with tectonically controlled steep slopes near the coast, with depths ranging berween ~35-40 m (Fig. 5.2.1.1, 2). In February 1866, a major sulfur release event was reported in Amfilochia Bay. Sulfur, as identified by its “rotten egg” smell, was released at such large quantities that it reached 198

the coasts of Preveza, a small port town on the other end (western) of Amvrakikos Gulf. Miaoulis (1876) reported that the sulfur was seeping from an underwater “volcano” whose “crater” was located ~270 m off the Amphilochia coast. According to Miaoulis, the “volcano” was distinct due to its shallow water depth (~5 m) in relation to the surrounding waters. Furthermore, Miaoulis (1876) reported testimonies from local fishermen who claimed that every time they recovered their gear from the “volcano” location, it was covered in sulfur.

Fig. 5.2.1.1 Geological map of Amfilochia Gulf and surrounding area based on Pérouse et al., 2017. Figure from Kordella et al., 2021.

Fig. 5.2.1.2 Field photos and stereographic projection of the steeply dipping composite normal fault surfaces controlling the surface morphology and bathymetry of the southeastern end of Amvrakikos Gulf adjacent to the coastal zone of Amfilochia town. Photos: Sotiris Kokkalas. 199

6.2.1.3 Methodology The field work in Amfilochia Bay took place between 2010 and 2015, and was organized into three distinct phases following a downscale approach. During the first phase (Phase 1), a marine geophysical survey was carried out throughout the Bay (Fig. 5.2.1.3), in order to investigate the gas-related morphological features, as well as possible anomalous acoustic characters within the sediments that may serve as proxies for the presence of gas seepage. The marine geophysical survey (Phase 1A) was combined with a small-scale hydrographic survey (Phase 1B) for datacollection of physicochemical parameters of the water column, for the study of the anoxic/hypoxic zone, in Amfilochia Bay. Phase 2 of the survey consisted of specific sites detailed investigation on spatial concentration of dissolved gasses, planned on the results of the first phase, using the MEDUSA system (Marinaro et al., 2011).

Fig. 5.2.1.3 Geophysical survey tracklines, MEDUSA tracklines and downcasting stations carried out in Amfilochia Bay. Figure from Kordella et al., 2021.

Phase 3 of the survey included seabottom water and sediment sampling with divers through one of the specific sites to determine the

13 isotopic composition of methane (δ CCH4). Phase 1A : Marine Geophysical survey The marine geophysical survey used: (i) an ELAC Nautic Seabeam 1185 multibeam echosounder system (MBES), (ii) an E.G. & G. Model 200

272-TD side scan sonar tow-fish in association with an EG & G Model 260 processor (Fig. 5.2.1.4) and (iii) a 3.5 kHz subbottom profiling system with Geopulse transmitter and a 4 array transducer (Fig. 5.2.1.4). Positional and navigation data was provided by a Differential Global Positioning System (DGPS) with an accuracy of 1-2 m. The side scan sonar survey was conducted at an operating frequency of 100 kHz and a range of 100 m per channel (swath width of 200 m). Slant-range correction was applied to the side scan sonar data during survey. Mosaicing of the side scan sonar data took place in Triton Map software (Triton Imaging Inc.) while the final resolution of the mosaic was set to 30 cm x 30 cm pixel size. Light tones in the side scan sonar mosaic represent high reflectivity (hard substrate or coarse grained sediments) and the dark tones represent low reflectivity (soft substrate).

201

Fig. 5.2.1.4 The 3.5 KHz subbottom profiler (top) and the EG&G side scan sonar (middle) systems aboard the research vessel Eirnini (bottom) used for the geophysical survey in Amvrakikos Gulf. Photos: Margarita Iatrou, Stavroula Kordella.

The subbottom profiler data were acquired using 3.5 kHz frequency, a 1 ms pulse duration with a pulse rate of 10 s-1. The vertical resolution of the system is about 0.5 m which is the minimum distance between the distinguishable reflectors. The geophysical survey tracklines that have taken place have a length of 11.733 km, covering a total area of 4.3 km2 (Fig. 5.2.1.3). The operational frequency of the multibeam echosounding system (MBES) was 180 kHz with a beamwidth of 131°, providing a vertical resolution of about 0.1 m. MBES acquisition supported by a Real Time Kinematics (RTK) GPS obtaining 10 cm position accuracy, vessel motion was acquired using a SMC IMU-108 MRU, sound velocity (SV) in the surface water was recorded using Valeport MiniSVS and in the water column using a Valeport MIDAS SVP. MBES tracklines having a total length of 37.5 km covered a total area of 10 km2. Acquisition and processing of MBES data performed using Hypack and Hysweep software. The processing of MBES data included the following stages: (i) Data Review: The first stage included the review of the navigation tracklines, heave-pitch-roll, tide-and-draft and sound velocity profile information. (ii) Swath-by-swath editing: Precise examination of bottom detail, clear spikes and errors. (iii) Cube: CUBE provides a near-automated editing of multibeam data, allowing for rapid turn-around data. Phase 1Β : Small-scale hydrographic survey During the small-scale hydrographic survey in Amfilochia Bay, measurements of physicochemical parameters of the seawater (downcasts), i.e., temperature, salinity, dissolved oxygen (DO), pH, and

H2S were carried out on two hydrographic stations (Station 2 and Station 30; Fig. 5.2.1.3) during summer period. For the hydrographic survey, an underwater multi-parameter CTD (In-situ TROLL 9500), and 202

a hydrogen sulfide sensor (Sea & Sun Technology – ATM) were used. Density was calculated based on the equation of state formula. One of the stations (Station 30; Fig. 5.2.1.3) was located over one of the specific sites that emerged through the geophysical survey, and the second one (Station 2; Fig. 5.2.1.3) at the centre of the Bay at water depth of 43 m and 35 m, respectively. Phase 2: MEDUSA underwater tow multi-parameter platform Focused measurements over the gas seeping morphological features were performed with the MEDUSA underwater tow multi-parameter platform (Fig. 5.2.1.5; Etiope et al., 2013), which includes sensors for the measurement of dissolved methane (METS), hydrogen sulfide and oxygen, CTD, altimeter and camera. All data was transmitted through a cable to the surface acquisition unit and recorded at a rate of 1 Hz at the same time as data from a DGPS differential positioning system for georeferencing data (Marinaro et al., 2011). With the MEDUSA platform, the same trackline of 250 m was repeatedly carried out, with the system towing at different water depths (4, 8, 16 m) and at fixed elevation of one meter above the seabed (towing water depth 29-42 m).

Fig. 5.2.1.5 The MEDUSA system (a) aboard the research vessel in Amfilochia Bay and (b) whileit's being towed. Figure from Kordella et al., 2021.

Phase 3 :Sediment and water sampling Eight water and six sediment samples were collected by divers from outside and inside of one of the gas seeping morphological features (Station 30 in Fig. 5.2.1.3) in Amfilochia Bay. Water samples were 203

collected closely over the seabed, directly into a 250 mL glass bottle and were sealed underwater with PTFE (polytetrafluoroethylene) butyl rubber septa and aluminum caps. Sediment samples were collected by divers in 500 mL aluminum cans. All samples were sealed at the sampling point to prevent contamination with atmospheric air. A bactericide (mercuric chloride) was added to both water and sediment samples, immediately after the samples were aqcuired, aboard the research vessel. Equal number of water and sediment samples were analyzed at Isotech Labs Inc. (Illinois, USA) and at the Stable Isotope Facility of the University of California, Davis (California, USA) in order to confirm the validity of the results. Analyses included complete compositional and

13 isotopic composition analysis of δ CCH4 of the gas in the water and sediment samples (at Isotech Labs Inc. by Finnigan Delta Plus XL mass spectrometer, precision ±0.3 ‰ (1σ) for 13C; at UC Davis by ThermoScientific PreCon concentration system interfaced to a ThermoScientific Delta V Plus isotope ratio mass spectrometer; ThermoScientific, Bremen, DE; http://stableisotopefacility.ucdavis.edu).

2 Methane and hydrogen sulfide concentrations were too low for δ HCH4

2 and δ HCH4 analyses.

6.2.1.4 Data Presentation - Results The examination of high resolution seismic profiling, side scan sonar and multibeam bathymetric data, collected in Amfilochia Bay, revealed the existence of anomalous acoustic characters in the seismic profiles and negative and possibly positive morphological features on the seafloor, indicative of gas in the interstatials of the sediments and gas escape features on the seabed. 204

6.2.1.4.1 Multibeam bathymetric data and seafloor morphology The multibeam bathymetric data showed that the seafloor of Amfilochia Bay is almost flat with a maximum depth of 47 m (Fig. 5.2.1.6). The deepest part of the Bay forms a very shallow elongated basin in a N.NW-S.SE direction, similar to the orientation of the NNW-trending oblique-normal faults that control the topographic relief on the eastern side of the bay (Fig. 5.2.1.2b, c). The seafloor is featureless with the exception of two almost circular, dish-shaped shallow depressions that were recorded at the Southern-Eastern part of the Bay (Fig. 5.2.1.6a). These depressions were interpeted as gas-induced pockmarks based on their general appearance that is identically described in many other areas worldwide (e.g., Judd and Hovland, 2007 and references therein). The two pockmarks are located in proximity to each other and to the eastern coastline of the Bay. The first pockmark (PM 1; Fig. 5.2.1.6a) is the largest among the two, and has an aerial extend of 1,500 m2, its diameter is about 40 m, the maximum depth is 38 m, which is 8 m deeper than the surrounding seafloor. PM1 can be considered as a composite pockmark, since it has been formed by a main and a secondary crater as recorded by the multibeam data. The second pockmark (PM2) has an aerial extend of 2,600 m2, a diameter of about 53 m, a maximum water depth of 35 m and it is 2 m deeper than the surrounding seafloor (Fig. 5.2.1.6a). Pockmark PM1 was the specific area that was chosen to be further investigated through the small-scale hydrographic survey and with the MEDUSA platform (Station 30; Fig. 5.2.1.3). Multibeam data revealed the presence of a small protruding seafloor mound that has ~100 m radius and ~16-6 m height compared to the surrounding seafloor, which is ~20-10 m deep (Fig. 5.2.1.6b). The mound is located very close to the coast and at the southern end of Amfilochia Bay, in the same location indicated by Miaoulis, (1876). 205

Fig. 5.2.1.6 Multibeam-bathymetric map of Amfilochia Gulf. Inset bathymetric maps showing (a) the two pockmarks; PM1 composite and PM2 single pockmark and (b) the small protruding seafloor mound and its intersection (AB). Figure from Kordella et al., 2021.

6.2.1.4.2 Side Scan Sonar Data The 100 kHz side scan sonar mosaic showed that the seafloor of Amfilochia Bay is flat and featureless. The side scan sonar data revealed that the seafloor of Amfilochia Bay is characterized, in general, 206

by low reflectivity which is indicative of the fine texture of the sediments (fine-grained sediments). Areas of high reflectivity on the southern most end on the Bay, represent the protrunding mound (Fig. 5.2.1.7b) and the rocky coastal slopes on the south-east. Very low reflectivity areas were recorded at the nothern part of Ampilochia Bay (Fig. 5.2.1.7). The two pockmarks were apparent also in the side scan sonar records, off the south-eastern coast of the Bay (Fig. 5.2.1.7a), exhibiting acoustic signal similar to that occurring in other worldwide sites (Judd and Hovland, 2007).

Fig. 5.2.1.7 100 kHz side scan sonar mosaic of Amfilochia Bay (a) the two pockmarks (PM 1 and PM2) at the south-eastern part of Amfilochia Bay, (b) the small protruding seafloor mound. Figure from Kordella et al., 2021. PM PM 1 2 HR P 207

6.2.1.4.3 Seismic stratigraphy and gas- bearing acoustic characteristics A variety of acoustic anomalies associated with shallow gas charged sediments, such as acoustic turbid zones (ATZ), enchanced reflectors (ER) and intra sedimentary gas plumes (ΙGP),which have been recorded on the high resolution seismic profiles, confirmed that the two shallow depressions in the southern-eastern part of Amfilochia Bay represent gas-induced pockmarks. The presence of gas in the seafloor seems that is not limited only at the area of the pockmarks but it has affected the total bay. The seismic profiles show a semi-transparent upper sequence with weak and strong internal parallel to subparallel reflectors overlying an acoustically-opaque, high reflective surface (Fig. 5.2.1.8). The acoustically-opaque horizon, which further blocks the sound penetration, can be attributed either to sharp interface in sediment texture or to gas in the interstitials of the sediments. However, there are various anomalous acoustic characters which are indicative of gas charged sediments (enhanced reflectors; ER), gas migration (gas plumes; GP), and seepage on the seabed (pockmarcks; pm). These suggest that the acoustically-opaque zone can be considered as a gas accumulation horizon and consequently it can be defined as acoustic turbid zone (ATZ) (Fig. 5.2.1.8).

Fig. 5.2.1.8 High resolution seismic profile showing typical stratigraphic pattern of Amfilochia Bay. A semi-transparent upper sequence (US) with weak and strong 208

internal parallel to subparallel reflectors overlying an acoustically-opaque, highly reflective surface (IGP: intra-sedimentary gas plume, ATZ: acoustic turbid zone). Figure from Kordella et al., 2021. based on the thickness of the upper sequence, the acoustic characteristics of the seabed and the acoustically-opaque zone, four distinct acoustic types have been defined. the spatial distribution of the acoustic types is shown in Fig. 5.2.1.9.

Fig. 5.2.1.9 Map of amfilochia bay showing the spatial distribution of the acoustic types defined by 3.5 kHz subbottom profiler. (MV: mud volcano?; PM: pockmark; B.PM: buried pockmark). Lines a, b, c correspond to the location of representative subbottom profiles of AT-A,B,C presented in Figure 5.2.1.10 and d of a profile indicative of AT-D shown in Figure 5.2.1.11. Figure from Kordella et al., 2021.

Acoustic type A (AT-A) consists of an upper semitransparent sequence that overlies an acoustically-opaque zone (Fig. 5.2.1.10). The upper sequence is further separated in three discrit units based on the 209

amplitude of the internal reflectors. It consists of two almost transparent units with very weak internal reflectors separated by a package of strong reflectors (PSR; Fig. 5.2.1.10). The seabed inside the AT-A area is characterized by very weak return. The AT-A province covers the southern part of the Bay (Fig. 5.2.1.9). Acoustic type B (AT-B) consists, similarly to AT-A, of an upper semitransparent sequence that overlies an acoustically-opaque zone (Fig. 5.2.1.10). In contrast to AT-A, the upper sequence of AT-B consists of two units; an almost transparent unit with very weak internal reflectors overlying the package of the strong reflectors (Fig. 5.2.1.10). The lower almost transparent unit of the upper sequence is absent in the AT-B. The acoustically-opaque zone of the AT-B has almost the same appearance as the opaque zone of AT-A. This type is characterized by a highly reflective seabed. Acoustic type B occurs at the northern part of the Bay (Fig. 5.2.1.9). Acoustic type C (AT-C) is characterized, similarly to AT-A, by an upper semitransparent sequence, divided in to three units, overlying a sequence consisting of (a) a weak acoustic turbid zone that allows the further sound penetration or (b) enchanced reflectors and intrasedimentary gas plumes (Fig. 5.2.1.10). This type is located southern of AT-A, where the pockmarks have been observed (Fig. 5.2.1.9).

Fig. 5.2.1.10 Representative high resolution seismic profile showing the acoustic types AT-A, AT-B and AT-C and the stratigraphic relationship between them (US: upper 210

sequence, TU: transparent unit, PSR: package of strong reflectors, IGP: intrasedimentary gas plumes). Figure from Kordella et al., 2021.

The acoustic type D (AT-D) is characterized by a thin semi-transparent upper unit with weak internal reflectors ovelying an acoustically-opaque surface (Fig. 5.2.1.11). AT-D is the less common type and is restricted at the southern end of the bay where the small protruding seafloor mound has been observed (Fig. 5.2.1.9).

Fig. 5.2.1.11. Representative high resolution seismic profiles showing the acoustic type AT-D, (US: upper sequence, AOZ: Acoustically opaque zone). Figure from Kordella et al., 2021.

Acoustic turbid zone in AT-A and AT-B appears with diffuse and chaotic character masking almost all other seimic reflections. ATZ, in most cases, is sharply cutting across the stratification indicating that it is probably not lithology-related (Fig. 5.2.1.12). Locally, acoustic turbid 211

zone appears weak, as indicated by the fact that the seismic reflectors underlying at the top of the ATZ are not entirely hidden, allowing their identification (Fig. 5.2.1.12). The top of the acoustic turbid zone is located ~10 m, ~9 m and ~11-12 m below the sea-bed surface, in AT-A, AT-B and AT-C, respectively.

Fig. 5.2.1.12 High resolution profiles showing (a,b) ATZ sharply cutting across the stratification (indicated by red arrows) and (c) ATZ appears weak as indicated by the fact that the seismic reflectors (red arrows) underlying the top of the ATZ are not entirely hidden (ATZ: acoustic turbid zone). Figure from Kordella et al., 2021.

Enhanced reflectors appear as anomalous high-amplitude reflections with abrupt terminations (Fig. 5.2.1.13). They have been observed in certain stratigraphic levels usually above the top of the acoustic turbid zone (Fig. 5.2.1.13). In one case, at the southeastern part of the Bay, within the AT-C provinence, enhanced reflector seems to be the result of the accumulation of gas due to migration through the 212

underlain upward curving sedimentary strata (Fig. 5.2.1.13). Similar migration pattern has been observed by Papatheodorou et al. (1993) in the Ionian Sea.

Fig. 4.2.1.13 High resolution profiles showing (a, b, c) enhanced reflectors and (d) enhanced reflectors due to gas migration through the underlain upward curving sedimentary strata (ER: enhanced reflectors). Figure from Kordella et al., 2021.

Intra-sedimentary gas plumes (IGP) have been observed almost exclusively in AT-A and AT-C (Fig. 5.2.1.13). These features show different acoustic signatures in the seismic profiles, more specifically appear as: (a) local high reflectivity patches (LHRP) accompanied by acoustic blanking (Fig. 5.2.1.14a) and (b) chaotic character with hyperbolic configuration (Fig. 5.2.1.14b). ΙGP’s have been detected just above the ATZ and within the upper semitransparent sequence. In some cases gas plumes reach the seabed surface (Fig. 5.2.1.14c). The relative position of ΙGP’s clearly suggets that these features were coming out of acoustic turbid zones. 213

Fig. 5.2.1.14 High resolution profiles showing: (a) local high reflectivity patches (LHRP) accompanied by acoustic blanking (b) IGP’s exhibit chaotic character with hyperbolic configuration, (c) ΙGP’s reaching the seabed surface . Figure from Kordella et al., 2021.

The area with the lowest reflectivity in side scan sonar mosaic is located at the northern part of Amfilochia Bay (Fig. 5.2.1.7) and coincides well with the spatial appearance of AT-B (Fig. 5.2.1.10) in which, the seabed reflector is very strong in seismic profiles. On the contrary, higher reflectivity on mosaic (Fig. 5.2.1.7) has been recorded at the south part of Bay where AT-A and AT-C are dominant (Fig. 5.2.1.10) Unexpectly, the seabed reflector in that area is very weak in the seismic profiles. That contradiction is probably due to the existence of large amount of gas in the surface sediments because of the upward gas migration from the underlying ATZ. The gas-charged surface sediments increase the backscatter on side scan sonar data, but the seabed return on seismic profiles remains very weak probably due to continuous disturbance of the sediment texture by the gas seepages. Two types of pockmarks were detected on seismic profiles: present-day pockmarks on the seafloor and paleo-pockmarks or v‐shaped amplitude anomalies in the sub-strata. Seismic profiles across the pockmarks which are located at the southeastern part of the Bay, inside the AT-C area, show the pockmarks as V-shaped and/or dish shaped incisions truncating all thethree units of the upper acoustically semitransparent sequence (Fig. 5.2.1.15a). The floor of the pockmark PM1 reaches the interface 214

between the upper semi-transparent sequence and the underlying acoustically opaque ATZbut without penetrating through it. PM1 is characterized by the existence of an acoustically turbid zone which is located under the floor of the pockmark (Fig. 5.2.1.15b, c). This suggests that: (i) there is a continuous upward migration of gas towards the floor of the pockmarks and (ii) the sediments below the floor and within the gas migration path are gas-charged. The sidewalls of the PM1 are characterized by a transparent acoustic character suggesting that the sediment texture has been disturbed by the gas venting (Fig. 5.2.1.15a). This, consequently, means that gas venting is not restricted only at the central part of the pockmark floor, but it also takes place through sidewalls. Sidewall gas venting has been also suggested by Hasiotis et al. (1996) in the active pockmark field of Patras Gulf, Greece. Seismic profile across PM1 shows a rotational slide affected the sidewall of the pockmark (Fig. 5.2.1.15d). This evidence may suggest that gas venting can be temporally blocked. Furthermore, this probably suggests that the development of a pockmark is a combined process of sediment removal by gas venting and sediment displacement by sliding, producing a more complex shape. Evidences of pockmark infilling by sidewall slumping have been documented in many cases worldwide (Hasiotis et al., 1996). Buried pockmarks have been observed at the western part of the Bay, within the AT-C area, suggesting past seepage manifestations (Fig. 5.21.16). The stratigraphic correlation of present day and buried pockmarks has clearly shown that the buried have truncated older strata than the presentday pockmarks (Fig. 5.2.1.16). This interpretation implies that gas releases and pockmark development occur at least since the Holocene in the southern part of Amfilochia Βay. 215

Fig. 5.2.1.15. High resolution profiles showing (a) PM1 truncating all the three units of the upper acoustically semitransparent sequence, (b, c) acoustically turbid zone which is located under the floor of the pockmark PM1 (P.ATZ), (d) a rotational slide (RS) affecting the sidewall of the pockmark PM1 (US: upper sequence, TU: transparent unit, PSR: package of strong reflectors, ER: enhanced reflector, ATZ: acoustic turbid zone). Figure from Kordella et al., 2021.

The two groups of pockmars (present-day and buried) seem to be linked with the same ATZ horizon, suggesting that both groups recharge the same gas accumulation horizon (Fig. 5.2.1.16). The spatial distribution of two groups of pockmarks shows an eastward migration of pockmark formation in the Bay. Although vertical stacking of buried pockmarks caused by reactivation of a specific fluid leakage through time is well studied, pockmark migration has not been addressed in detail yet. The occurrence of this pockmark migration through time is possibly caused by seismic fault activity driven changes. 216

Fig. 5.2.1.16 High resolution profiles showing active (PM1) and buried (B.PM) pockmarks (PFL: pockmark’s floor elevation). Figure from Kordella et al., 2021.

The stratigraphic relationships between the three acoustic types (AT-A, AT-B and AT-C) suggest that either the ATZ is migrating upward inside the AT-B province or a substantial amount of gas has been migrated upward and has escaped in the water column lowering the level of ATZ wihin the AT-A and AT-C (Fig. 5.2.1.10). The latter hypothesis is considered more likely, as it is further supported by the existence of important morphological features (PM) and acoustic characters (IGP, ER) indicative of gas upward migration at the AT-A and AT-C areas.

6.2.1.4.4 The “volcano” of Amfilochia Bay At the approximate location that Miaoulis, (1876) erroneously reported an underwater “volcano” the acoustic signal obtained from profiling and side scan imagery (Fig. 5.2.1.7, 9) prompts to a conical shaped feature suggesting that most probably the site is associated with sediment build-up processes such are mud domes and carbonate mounds (Loncke et al., 2004, Coleman and Ballard, 2001). Based on all the acquired acoustic data sets a protrusion mound has been recorded having a diameter of ~100 m and a height of 16 m compared to the surrounding seabed level which is 20 m deep. The side scan sonar mosaic showed that the mound is characterized by high back-scattering center reflecting hard substrates and low back- scattering rim reflecting accumulation of soft, fine grained material (Fig. 217

5.2.1.7b). This locally intense backscatter contrast could correspond to the development of: (i) seepage-related carbonate-cemented sediments (Vogt et al., 1997; Hovland and Judd, 1988), (ii) loose and/or consolidated mud deposits (breccia; Vogt et al., 1999), (iii) bioconstructions, formed by corals, sponges, or coralline algae on cemented sediments (Hovland and Judd, 1988; Hovland and Thomsen, 1997; Vogt et al., 1997; Hovland and Mortensen, 1999; Bouriak et al., 2000) and (iv) local gas bubbling (Hovland, 1990, Lyons et al., 1996; Vogt et al., 1999), at the center of the mound. The possibility of bioconstructions and/or gas bubbling as responsible factors for the high reflectivity at the center of the mound appears rather weak at the study area, because of the homogenous acoustic pattern and the absence of any evidences for gas seepages at least during the survey period. The profiles collected across the mound are characterized by a prolonged acoustic surface echo suggesting hard seafloor (Fig. 5.2.1.17) in agreement with the sonographs. The prolonged surface reflector could represent rocky outcrops on the seafloor or gas-induced sedimentary buildup causing acoustic blanking and/or preventing of the acoustic penetration. Downslope from its top, the mound is covered by a thin veneer of sediments which gradually exhibits stratification as indicated by parallel to sub-parallel, weak and discontinuous internal seismic reflectors (Fig. 5.2.1.17). At the flanks of the mound, sedimentary deposits, free of reflections, having lens shapes and convex tops, and presenting sharp transitions to the stratified sediments have been observed (Fig. 5.2.1.17). The acoustic characteristics suggest that these deposits are most possible sedimentary debris flows, originated from the center of the mound. This latter observation together with the circular shaped of the high-low reflectivity alternation on the sonographs (described previously) imply that the site could be the result of a mud volcano and the debris flows could be attributed to mud flows (mud breccia). 218

Fig. 5.2.1.17 High resolution seismic profiles showing the protruding mound which possibly represents a mud volcano (PAE: prolonged acoustic echoes, MF: mud flow deposits, GP: gas pocket). Figure from Kordella et al., 2021.

6.2.1.4.5 Physicochemical parameters of the Amfilochia Bay water column The physiography of Amvrakikos Gulf with the limited connection to the open Ionian Sea and the freshwater supply by the two main rivers is reflected also in its innermost part, Amfilochia Bay. The downcast measurements in the vertical profiles of stations St 2 and St 30 (Fig. 5.2.1.3), during summer period, present the general temperature, salinity, and density conditions of the water column in the central part of

Amfilochia Bay, accompanied by DO, pH, dissolved CH4 and H2S concentrations (Fig. 5.1.2.18). The vertical profiles showed that the water column of the Bay is divided in two well defined layers; a brackish surface layer (Fig. 5.2.1.18) and a saline bottom layer (Fig. 5.2.1.18), which are separated by a strong pycnocline (Fig. 5.2.1.18). The surface brackish layer is homogenous with a temperature of between 29 and 30 oC and salinity between 28 and 29 PSU. The bottom layer is also homogenous with a temperature of between 16 and 17 oC and salinity between 37 and 38 PSU. The thermocline develops between 5 and 18 m, whilst the halocline is very sharp and develops between 5 and 10 m in Station 2, and 8 and 10 m in Station 30. The intense stratification prevents the vertical diffusion of oxygen, from the oxygenated, surface layer to the bottom layer. Thus, dissolved oxygen (DO) vertical profile is also controlled by the two-layer structure 219

of the water column. The brackish surface layer is well oxygenated with a concentration of about 6.5-7 mg/L. Within the thermocline/halocline, the DO content continuously decreases reaching a concentration of ~0 mg/L below ~18 m water depth in Station 2 and below 20 m in Station 30. The pH distribution in the water column seems to be affected by the two layer structure only in Station 30, which is located above the pockmark PM1. In the surface layer the pH is just above 8 reflecting the entrainment and consequently the mixing of the inflowing sea water with the surface layer. Below the halocline the pH is about 7.6 indicating that the water is more acidic in the bottom layer of the water column over the largests pockmark of Amfilochia Bay.

Fig. 5.2.1.18 Downcasting data collected from two stations (Station 2 and Station 30) during the summer period. The Station 30 is located just above PM1 pockmark. Figure from Kordella et al., 2021.

Measurements showed dissolved methane concentrations of ~0.5-1 μM in the brackish surface layer of Amfilochia Bay. Hydrogen sulfide concentrations were low at two sampling sites ranging between 0 and 0.02 mg/L at the brackish surface layer and 0.03 mg/L near the sea bottom. 220

6.2.1.4.6 MEDUSA underwater tow multi- parametric platform The MEDUSA platform on adjacent floats at different depths above the PM1 pockmark, showed increasing concentrations of dissolved methane in correspondence with depth and the pockmark (Fig. 5.2.1.19). The shallowest float, at ~4 m, showed an almost constant dissolved methane concentration of about 0.12 μΜ. The floats at ~8 m and~15 m water depth, showed elevated dissolved methane concentrations that range between 0.26–0.49 μΜ. The deepest float (>0.5 m distance from seabed) that crossed pockmark PM1, showed even more CH4 concentration values, from 0.32 μΜ, reaching up to 0.58 μΜ right after the MEDUSA exited the pockmark. Thus, an increase in the methane concentration was recorded by the MEDUSA at the bottom layer of the stratified water column of the bay close to the detected halocline (Fig. 5.2.1.18), while at the surface layer the CH4 concentration was low and constant. Higher CH4 concentrations were detected after the crossing of PM1 and always under the halocline, but the highest detected values occurred near the seabed, inside and right after exiting the pockmark. The methane concentration increase was attributed to weak gas seepage (Fig. 5.2.1.19a, c). The fact that the highest values were measured during the exit of the MEDUSA platform from the pockmark is due to slow response of the methane sensor (Marinaro et al., 2006). Dissolved oxygen concentration declines with depth reaching very low values (>5 μΜ) in the bottom layer (Fig. 5.2.1.19b, c). But whilst approaching PM1, the increase in the methane concentration was accompanied by an even greater decline in dissolved oxygen in and around the pockmark (Fig. 5.2.1.19b). The additional oxygen decline is propably due to the methane seepage from the pockmark. Although this process has been reported bibliographically it is difficult to be recorded in the field (Schmale et al., 2010; Kessler et al., 2011). In the case of the Amfilochia Bay pockmarks, the MEDUSA platform testified the 221

reduction of oxygen in the water due to the escape of methane from a pockmark. The escape of methane appears to enhance the hypoxic/anoxic environment inside and around the pockmark. Hydrogen sulfide concentrations measured with the MEDUSA platform were very weak. A slight increase (0.4-0.5 mg/L) was measured near the bottom of the pockmark PM1, whilst at the upper water column H2S concentration ranged between 0.0 and 0.3 mg/L.

Fig. 5.2.1.19 a, b. The MEDUSA system tracklines at depths 4, 8, 16 and 30-43 m, plotted in combination with the corresponding (a) methane and (b) dissolved oxygen measured concentrations; (c) 3-D representation of the Amfilochia Bay pockmarks showing the position of the closest to the seabed MEDUSA system trackline (30-43 m) and the corresponding dissolved oxygen concentration and dissolved methane concentration. Vertical exaggeration: 3. 222

6.2.1.4.7 Isotopic data from sediment and water samples

13 Isotopic analyses of the CH4 carbon atoms (δ CCH4) in samples of water and sediments collected from the interior of and around PM1 pockmark showed values from -41 ‰ to -86 ‰ (Table 5.2.1.1), which can be attributed (a) to a different degree of oxidation of CH4 of microbial origin from methanotrophic bacteria (Schubert et al., 2006) or (b) mixed gases of mostly microbial and secondarily thermogenic origin.

It was not possible to analyze the isotopic composition of H2S due to the low concentrations in the samples.

Table 4.2.1.1. Isotopic analyses results from different samples that were collected from near and inside PM1 pockmark. 13 δ CCH4 UC Davis Isotech Outside pockmark -83.5‰ - sediment Inside pockmark -84.7 ‰ - 87 ‰ sediment Outside pockmark -41.5 ‰ - water Inside pockmark water -52.3 ‰ -

6.2.1.5 Discussion A comprehensive oceanographic survey using remote sensing, hydrographic and geochemical means together with a multi-parameter platform was carried out at Amfilochia Bay in the southeastern end of Amvrakikos Gulf. The synthesis of the collected data derived by that multidisciplinary approach suggests that Amfilochia Bay is characterized by gas charged sediments and morphological features (pockmarks and one possible mud volcano) which can be considered as recorders and indicators of past and present hydraulic seabed activity. The geophysical and bathymetric data showed piercement 223

structures at the southern part of Amfilochia Bay. There, two active pockmarks, one single and one composite, and several buried pockmarks were observed. The pockmarks seem to be related with the Amfilochia (AFZ) fault zone which is considered as a left-lateral strike slip fault, based on seismological and GPS data (Ganas et al., 2009). The buried pockmarks are located close to the north fault tip of the southern segment of the AFZ, which is running along the western coast of the Bay and trends south along the Amfilochia valley. The active pockmarks are located in the relay zone between the southern and northern segments of the AFZ, while the northern segment controls the eastern coast of Amfilochia Bay. The stratigraphic correlation and the spatial distribution of present day and buried pockmarks showed an eastward migration of pockmarks formation and activation in the Bay. The occurrence of this pockmark migration through time was possibly caused by the seismic fault activity driven changes of the AFZ. Although a plethora of gas seepage-related features, such as acoustic turbid zones (ATZ), enhanced reflectors (ER), intra- sedimentary gas plumes (IGP) were detected in the subbottom profiles of Amfilochia Bay, the measurements made by the hydrographic survey (Fig. 17) and the MEDUSA platform (Fig. 18) showed weak gas seepage from the largest, composite pockmark (PM1) at the south- eastern part of the Bay. In addition, the H2S concentrations detected inside the PM1 pockmark were too low to result in seepage into the atmosphere. However, in case of intense seismic events, the possibility of larger amounts of CH4 and H2S escape from the two pockmarks should not be ruled out. Similar seismic stratigraphic pattern has been observed in other locations with confirmed gas-charged sediments, e.g., in Patraikos Gulf, Greece (Hasiotis et al., 1996), Eckernforde Bay, Germany (Wever et al., 1998). In Patraikos Gulf, seismic activity triggered gas escape pockmark activations (Hasiotis et al., 1996; Christodoulou et al., 2009), and similar phenomena could occur and should be explored in Amfilochia Bay. 224

Αprotrusion mound has been found, approximately at the same area in which Miaoulis (1876) discovered a submarine “volcano” related to vast sulfur emissions. Although there is no direct evidence (sediment cores, visual inspection) that the protrusion mound is a mud volcano, the collected geophysical data strongly suggest this hypothesis. Yin et al., (2003) suggested that submarine mud volcano can be considered any positive relief on the seafloor related to the expulsion of fluid and/or gas. Milkov (2000) suggests two additional criteria for a more precise identification of submarine mud volcanoes: (i) the presence of specific submarine mud volcanic sediments in cores (mud breccia) and (ii) the presence of strong backscatter defined from side scan sonar images. Etiope and Martinelli (2009) and Etiope (2015) challenged the misuse of the “mud volcano” term and proposed, four rigorous criteria in the definition of mud volcanoes, emphasized to (i) the discharge of at least a three-phase system (gas, water, sediment) and (ii) the geotectonic setting (involvement of sedimentary rocks with a gravitative instability leading to the formation of mobile shales, diapirs or diatremes). The possible mud volcano of Amfilochia Βay shows a clear positive relief with very high backscattering acoustic facies at the position of the mound. It can be interpreted either as (i) very consolidated mud deposits consisting of mud breccia transported from underlying deposits (Vogt et al., 1999) or (ii) carbonate-cemented sediments, due to the presence of hydrocarbon within the escaping fluids (Vogt et al., 1997; Hovland and Judd, 1988). No gas seepage or fluid expulsion were recorded during the survey above the mound-like feature. On the other hand, the active and buried pockmarks that have been observed adjacent to the possible mud volcano mound, suggest recent or even presently active gas seepage reaching the seafloor of the southern part of the Bay. The absence of gas plumes in the water column over the mound from subbottom profiling and most importantly from side scan sonar data does not necessarily imply that this mound is not a mud volcano. It 225

should be noted that the detection of gas bubbles depends on the resonance frequency of the bubbles, which increases with water depth. The 100 kHz side scan sonar that has been used in the Amfilochia Bay survey detects bubbles greater than 0.006 cm diameter in 10 to 20 m water depth, while subbottom profiler only detects bubbles in excess of 0.10 cm diameter (Guinasso and Schink, 1973). The nondetection of gas seepage over the mound, as indicated by side scan sonar data, suggests that gas expulsion might be intermittent. Yin et al. (2003) also mentioned the absence of gas plumes over mud volcanoes in East China Sea and they suggested that those mud volcanoes were not active, because, among other reasons, the gas emission was not continuous but could be intermittent. The debris flow deposits that have been observed at the base of the mound and are interpreted as mud flow deposits (mud breccias), based on acoustic criteria, further support the hypothesis that the mound is a mud volcano. Similar debris flows have been found in numerous mud volcanos worldwide, indicating the mud volcano activity (Van Rensbergen et al., 2005, Somoza et al., 2012). The existence of the Amfilochia strike slip fault zone and the Triassic evaporites beneath the Amfilochia valley that have been remobilized in Plio-Quaternary time, may point towards fossil (either thermogenic or microbial) gas seepage which may have caused the erosional and piercement morphological structures such like pockmarks and mud volcanoes. Mazzini and Etiope (2017) have mentioned that mud volcanoes are distributed, among others, in compressional zones of accretionary complexes and thrust and overthrust belts and are specifically located in strike-slip and normal faults. Methane isotopic analyses from the composite pockmark PM1 1.4 km northeast of the possible mud volcano have been interpreted as

(i) a different degree of oxidation of CH4 of microbial origin from methanotrophic bacteria (Schubert et al., 2006) or (ii) mixed gases of mostly microbial and secondarily thermogenic origin. Methane isotopic composition shows variable carbon isotopic ratios, from -41 ‰ to -86 226

‰, which could be related to differential CH4 oxidation or mixing between microbial and thermogenic gas. Nevertheless, the degassing is clearly determined by a geologic structure: a fault-controlled pockmark.

Radiocarbon dating of CH4 could clarify if the gas is modern or fossil (e.g., Botner et al., 2018). This study also confirmed the intense stratification of the water column during the summer in Amfilochia Bay. The bottom layer showed hypoxic/anoxic DO levels (4-60 μM) and an anoxic seabed (DO<4.5 μΜ). The above is consistent with the observations made by other researchers (Ferentinos et al., 2010, Kountoura and Zacharias, 2011) who suggested that the main mechanisms for the formation of hypoxia in Amfilochia Bay and in the wider Amvrakikos Gulf are oceanographic and anthropogenic (Ferentinos et al., 2010). This can be considered as an effect of many years of continuous input of agricultural (Papaefthymiou et al., 2013), aquaculture and municipal waste in Gulf of Amvrakikos, coupled with small input of well oxygenated sea water from the Ionian Sea. The MEDUSA underwater tow multi-parameter system, in sequential floats, recorded weak methane seepage and even lower seepage of hydrogen sulfide, over the PM 1 gas escape pockmark.

Increase in the CH4 concentration was accompanied by an additional decline in dissolved oxygen in and around the PM1 pockmark in relation to the already low oxygen concentration in the bottom layer. This additional oxygen depletion may be attributed to the escape of CH4 (Kessler et al., 2011). Although this process has been reported bibliographically it is difficult to be recorded in the field (Schmale et al., 2010; Kessler et al., 2011). In the case of the Amfilochia Bay pockmarks, the MEDUSA system testified the reduction of oxygen due to the escape of CH4 from the PM1 pockmark. Nevertheless, the reported CH4 seepage is too local and too weak to be considered as a driver of hypoxia in Amfilochia Bay.

227

4.2.1.5 Conclusions The integrated oceanographic survey that took place in Amfilochia Bay showed that the Bay is characterized by gas-charged sediments, while the southern part is affected by gas seepage and related erosional and piercement morphological features. This study resulted in the discovery of a pockmark group and a possible mud volcano, off the southern coast of Amfilochia Bay. These are the only pockmarks found in Amvrakikos Gulf; a marine area extensively studied and eminent for its shallow gas accumulations (Papatheodorou et al., 1993; Kapsimalis et al., 2005; Poulos et al., 1995). A small protruding mound which is indicated as submarine “volcano” by Miaoulis (1876) is suggested as a possible mud volcano based on geophysical characteristics. The development of erosional and piercement morphological features in the southern part of Amfilochia Bay seems to be controlled by the ATZ fault zone and therefore could be considered as fault-controlled features. The main drivers of hypoxia-anoxia in Amfilochia Bay were confirmed to be oceanographic and anthropogenic (Ferentinos et al., 2010) since the detected methane seepage was weak and local, at least during the surveying period. Based on the results of this survey, methane seepage can further contribute to oxygen reduction, but only locally; in the interior and/or near PM1 pockmark (Fig. 5.2.1.19c). Methane seepage was too weak to cause DO perturbations in shallower waters or affect the depth of the oxycline over the pockmark (Station 30; Fig. 5.2.1.18), which is similar to the other non-seepage area in Amfilochia Bay (Station 2; Fig. 5.2.1.18). Methane shows variable carbon isotopic ratios, from -41 ‰ to -86

‰, which could be related to differential CH4 oxidation or mixing between microbial and thermogenic gas. However, the degassing seems to be determined by a geologic structure: a fault-controlled pockmark. Radiocarbon dating of CH4 would help deciphering its either modern or fossil nature (e.g., Botner et al., 2018). 228

Nevertheless, a well-organized, integrated study, that combines a geophysical, a geochemical and a two-dimension (vertical and horizontal) hydrographic survey, is considered the most effective method to study marine areas with shallow gas accumulations, since they complement each other, shedding light in all aspects of a case study.

Acknowledgements This study was funded by the European Union FP7 project “HYPOX – In situ monitoring of oxygen depletion in hypoxic ecosystems of coastal and open seas and landlocked water bodies” (EC grant 226213). Additional funding was provided by the project “IDENTIFICATION, CONSEQUENCES AND MANAGEMENT OF THE ANOXIC ZONE OF AMVRAKIKOS GULF (NW GREECE)”, EEA Grants, GR02-0010 and by the Laboratory of Marine Geology & Physical Oceanography, University of Patras own funds. The authors gratefully acknowledge the valuable contribution of Dr Giuseppe Etiope who oversaw the geochemical research and the MEDUSA platform measurements during the HYPOX project. Thanks go to the captains and crew of “EIRINI R/S”, “BARBANTONIS II, NP470” for their help during the research cruises.

References Bouriak, S., Vanneste, M., Saoutkine, A. (2000). Inferred gas hydrates and clay diapirs near the Storegga Slide on the southern edge of the Voring Plateau, Offshore Norway. Mar Geol, 163, pp. 125–148. https://doi.org/10.1016/S0025-3227(99)00115-2 Christodoulou D., Papatheodorou G., Fakiris E., EtiopeG., Ferentinos G. (2009). The activation of a fluid escape pockmark field after the Μw=6.4R earthquake on 8 June 2008, NW Peloponnesus. Proceedings 229

of 9th Symposium of Oceanography & Fisheries, Vol. Ι, pp. 3-8 (In Greek). Botner, E.C., Townsend-Small, A., Nash, D.B., Xu, X., Schimmelmann, A., Miller, J.H. (2018). Monitoring concentration and isotopic composition of methane in groundwater in the Utica Shale hydraulic fracturing region of Ohio. Environ Monit Assess, 190, 322. https://doi.org/10.1007/s10661-018-6696-1 Coleman, D.F. and Ballard, R D. (2001). A highly concentrated region of cold hydrocarbon seeps in the southeastern Mediterranean Sea. Geo- Mar Lett, 21, pp. 162–167. https://doi.org/10.1007/s003670100079 Clews, J.E. (1989). Structural controls on basin evolution: Neogene to Quaternary of the Ionian Zone, Western Greece. Journal of the Geological Society, 146(3), pp. 447 – 457.https://doi.org/10.1144/gsjgs.146.3.0447 Diamantopoulou, A.T., Kalavrouziotis, Ι. K., Varnavas, S. P. (2018).Geochemical investigations regarding the variability of metal pollution in the Amvrakikos Bay, Greece.Global NEST Journal, 21(1), pp. 7-13.https://doi.org/10.30955/gnj.002733 Depreiter, D., Poort, J., Van Rensbergen, P., Henriet, J.P. (2005). Geophysical evidence of gas hydrates in shallow submarine mud volcanoes on the Morrocan Margin. J Geophys Res, 110, B10103, https://doi.org/10.1029/2005JB003622 Etiope, G., Christodoulou D., Kordella, S., Marinaro, G., Papatheodorou, G. (2013). Offshore and onshore seepage of thermogenic gas at Katakolo Bay (Western Greece).Chem Geol, 339, pp. 115–126, http://doi.org/10.1016/j.chemgeo.2012.08.011 Etiope, G. (2015). Natural Gas Seepage, The Earth’s hydrocarbon degassing. Springer, Switzerland, 199 pp. Etiope, G. and Martinelli, G. (2009). “Pieve Santo Stefano” is not a mud volcano: comment on Structural controls on a carbon dioxide-driven mud volcano field in the Northern Apennines (by Bonini, 2009). J Struct Geol, 31, pp.1270– 1271.https://doi.org/10.1016/j.jsg.2009.06.009 230

Ferentinos, G., Papatheodorou, G., Geraga, M., Iatrou, M., Fakiris, E., Christodoulou, D., Dimitriou, E., Koutsikopoulos, C. (2010). Fjord water circulation patterns and dysoxic/anoxic conditions in a Mediterranean semi-enclosed embayment in the Amvrakikos Gulf, Greece.Estuarine, Coastal and Shelf Science. 88(4), pp. 473-481. https://doi.org/10.1016/j.ecss.2010.05.006 Friedrich, J., Janssen, F., Aleynik, D., Bange, H., Boltacheva, N., Çagatay, M., et al. (2013). Investigating hypoxia in aquatic environments: Diverse approaches to addressing a complex phenomenon.Biogeosciences vol. 11(4), pp. 1215- 1259.https://doi.org/10.5194/bg-11-1215-2014 Friligos, N., and Koussouris, T. (1977). Preliminary chemical, physical and biological observations in the Gulf of Amvrakikos (April 1974). Thalas Jugosl, 13. pp. 53-58. Guinasso, N. L. and Schink, D. R. (1973). A simple physicochemical acoustic model of methane bubbles rising in the sea. Report Reference 73-15-T. College of Geosciences, Texas A&M University. Ganas, A., Serpelloni, E., Drakatos, G., Kolligri, M., Adamis, I., Tsimi, Ch. and Batsi, E. (2009). The Mw 6.4 SW Achaia (Western Greece) Earthquake of 8 June 2008: Seismological, Field, GPS Observations, and Stress Modeling. J Earthq Eng, 13(8), pp. 1101-1124. https://doi.org/10.1080/13632460902933899 Haddad, Α., Ganas, A., Kassaras, I., M. Lupi, M. (2020). Seismicity and geodynamics of western Peloponnese and central Ionian Islands: insights from a local seismic deployment, Tectonophysics, 778, 228353. https://doi.org/10.1016/j.tecto.2020.228353 Hasiotis, T., Papatheodorou, G., Kastanos, N., Ferentinos, G. (1996). A pockmark field in the Patras Gulf (Greece) and its activation during the 14/7/1993 seismic event. Mar Geol, 130, pp. 333- 344.https://doi.org/10.1016/0025-3227(95)00131-X Hatzfeld, D., Nord, J., Paul, A., Guiguet, R., Briole, P., Ruegg, J. C., Cattin, R., Meyer, B., Hubert, A., Bernard, P., Makropoulos, K., Karakostas, V., Papaioannou, C., Papanastassiou, D., Veis, G. (1995). 231

The Kozani-Grevena (Greece) earthquake of May 13, 1995, Ms = 6.6. Preliminary results of a field multidiscipliuary survey. Seism Res Lett, 66, pp. 61-70. Hollenstein C., Mόller M.D., Geiger A. Kahle H.G., (2008). Crustal motion and deformation in Greece from a decade of GPS measurements 1993-2003.Tectonophysics, 449, 17 – 40.https://doi.org/10.1016/j.tecto.2007.12.006 Hovland, M. (1990). Do carbonate reefs form due to fluid seepage? Terra Nova, 2: 8-18. https://doi.org/10.1111/j.1365- 3121.1990.tb00031.x Hovland, M. and Judd, A.G. (1988). Seabed Pockmarks and Seepages: Impact on Geology, Biology and Marine Environment, Graham and Trotman, London.293 pp. Hovland.M. and Thomsen.E. (1997). Cold-water corals — are they hydrocarbon seep related? Mar Geol, 137, pp.159– 164.https://doi.org/10.1016/S0025-3227(96)00086-2 Hovland, M. and Mortensen, P.B. (1999). Norwegian coral reefs and processes in the seafloor. John Grieg Forlag, Bergen, Norway, (In Norwegean with English summary). Judd, A.G. and Hovland, M. (2007). Submarine fluid flow. The impact on geology, biology, and the marine environment. Cambridge University Press, Cambridge, pp. 475. https://doi.org/10.1017/CBO9780511535918 Kapsimalis, V., Pavlakis, P., Poulos, S.E., Alexandri, S., Tziavos, C., Sioulas, A., Filippas, D., Lykousis, V. (2005). Internal structure and evolution of the Late Quaternary sequence in a shallow embayment: The Amvrakikos Gulf, NW Greece. Mar Geol, 222-223 (1-4), pp. 399- 418.https://doi.org/10.1016/j.margeo.2005.06.008 Kessler, J.D, Valentine, D.L, Redmond, M.C., Du, M., Chan, E.W., Mendes, S.D., Quiroz, E.W., Villanueva, C.J., Shusta, S.S., Werra, L.M., Yvon-Lewis, S.A., Weber, T.C. (2011). A Persistent Oxygen Anomaly Reveals the Fate of Spilled Methane in the Deep Gulf of 232

Mexico. Science. 331(6015), pp. 312-315. https://doi.org/10.1126/science.1199697 Kokkalas.S., Xypolias. P., Koukouvelas. I., Doutsos, T. (2006). Post- collisional contractional and extensional deformation in the Aegean region, In: Dilek, Y., Pavlides, S., (eds) Post-collisional tectonics and magmatism in the Mediterranean region and Asia. Geological Society of America, Special Paper, 409, pp. 97–123. https://doi.org/10.1130/2006.2409(06) Kormas, K. A., Nicolaidou, A., Reizopoulou, S. (2001). Temporal variations of nutrients and chlorophyll-a and particulate matter in three coastal lagoons of Amvrakikos Gulf (Ionian Sea, Greece). Mar Ecol, 22, pp. 201-213. Kountoura, K., Zacharias, I. (2011). Temporal and spatial distribution of hypoxic/seasonal anoxic zone in Amvrakikos Gulf, Western Greece.Estuarine, Coastal and Shelf Science. 94 (2), pp. 123-128. Loncke, L., Mascle, J., Fanil Scientific Parties (2004). Mud volcanoes, gas chimneys, pockmarks and mounds in the Nile deep-sea fan (Eastern Mediterranean): geophysical evidences. Mar Pet Geol, 21(6), pp. 669-689. https://doi.org/10.1016/j.marpetgeo.2004.02.004 Lyons, A.P., Duncan, M.E., Anderson, A.L. (1996). Predictions of the acoustic scattering response of free‐methane bubbles in muddy sediments. The Journal of the Acoustical Society of America, 99(1), pp. 163-172. https://doi.org/10.1121/1.414500 Mazzini, A. and Etiope, G., (2017). Mud volcanism: An updated review, Earth-Science Reviews. V16, pp. 81-112. https://doi.org/10.1016/j.earscirev.2017.03.001 Le Pichon, X., Chamot-Rooke, N., Lallemand, S., Noomen, R., Veis, G., (1995). Geodetic determination of the kinematics of central Greece with respect to Europe: implications for eastern Mediterranean tectonics. Journal of Geophysical Research Atmospheres. 100, pp. 12,675- 12,690. 233

Mazzini, A., Etiope, G. (2017). Mud volcanism: An updated review. Earth-Science Reviews, 16, pp. 81-112. https://doi.org/10.1016/j.earscirev.2017.03.001 Marinaro, G., Etiope, G., Lo Bue, N., N., Favali, P., Papatheodorou, G., Christodoulou, D., Furlan F., Gasparoni, F., Ferentinos, G., Masson, M., Rolin, J.F. (2006).Monitoring of a methane-seeping pockmark by cabled benthic observatory (Patras Gulf, Greece). Geo-Mar Lett, 26, pp. 297– 302. https://doi.org/10.1007/s00367-006-0040-4 Marinaro, G., Etiope, G., Gasparoni, F. Furlan F., Bruni, F. (2011). Gas seepage detection and monitoring at seafloor.10th Offshore Mediterranean Conference and Exhibition, Ravenna, Italy, March 23-25, 2011, “Frontier exploration” session. Miaoulis, A.A. (1876). Report to the Naval Ministry. Regarding the Karvasaras (Amfilochia) volcanic phenomenon. Protocol number 17. In: Telonas, N. H. (2017). A trip to forgotten years, Amfilochia (Karvasaras) 1829 Vol. A. pp. 703-704.Amfilochia Public Library. Amfilochia, Greece. Milkov, A.V. (2000). Worldwide distribution of submarine mud volcanoes and associated gas hydrates. Mar Geol, 167(1-2), pp. 29-42. https://doi.org/10.1016/S0025-3227(00)00022-0 Papatheodorou, G., Hasiotis, T., Ferentinos, G. (1993). Gas charged sediments in the Aegean and Ionian Seas, Greece. Mar Geol, 112, pp. 171-184.https://doi.org/10.1016/0025-3227(93)90167-T Papaefthymiou, H., Athanasopoulos, D., Papatheodorou, G., Iatrou, M., Geraga, M., Christodoulou, D., Kordella, S., Fakiris, E., Tsikouras, B., (2013).Uranium and other natural radionuclides in the sediments of a Mediterranean fjord-like embayment, Amvrakikos Gulf (Ionian Sea), Greece.Journal of Environmental Radioactivity, 122, pp. 43-54. https://doi.org/10.1016/j.jenvrad.2013.02.020 Pérouse, E., Sébrier, M., Braucher, R. et al. (2017). Transition from collision to subduction in Western Greece: the Katouna–Stamna active fault system and regional kinematics. Int J Earth Sci (Geol Rundsch) 106, pp. 967–989. https://doi.org/10.1007/s00531-016-1345-9 234

Piper, D.J.W., Panagos, A. G., Kontopoulos, N. (1982). Some observations on surficial sediments and physical oceanography of the Gulf of Amvrakia. Thalassographica, 5, pp. 63–80. Poulos, S. E., Lykousis, V., Collins, M.B. (1995). Late Quaternary Evolution of Amvrakikos Gulf, Western Greece.Geo-Marine Letters. 15:9-16. https://doi.org/10.1007/BF01204492. Rogers, J., Kelley, J., Belknap, D.F., Gontz, A., Barnhardt, W.A. (2006). Shallow-water pockmark formation in temperate estuaries: a consideration of origins in the western Gulf of Maine with special focus on Belfast Bay. Mar Geol, 225, pp. 45– 62.https://doi.org/10.1016/j.margeo.2005.07.011 Schmale, O., Beaubien, S.E., Rehder, G., Greinert, J., Lombardi, S. (2010). Gas seepage in the Dnepr paleo-delta area (NW-Black Sea) and its regional impact on the water column methane cycle. Journal of Marine Systems, 80 (1-2), pp. 90- 100.https://doi.org/10.1016/j.jmarsys.2009.10.003 Schubert, C., Durisch-Kaiser, E., Klauser, L., Vazquez, F., Wehrli, B., Holzner, C., Kipfer, R., Schmale, O., Greinert, J., Kuypers, M. (2006). Recent studies on sources and sinks of methane in the Black Sea. In: Neretin, L.N. (Ed.) Past and Present Water Column Anoxia, Springer, Netherlands, pp. 419–441. Somoza, L., Vázquez, J.T., González-Aller, D., Dañobeitia, J.J., Medialdea, T., Moya, A.J., Fernández-Salas, L.M., León, R., Jurado, M.J., Paredes, M., Palomino, D., Vargas, I., González, F.J., López- Rodríguez, M., Corbalán, A., González, P., Hernández-Puentes, M.P., Martín-Jiménez, D. (2012). Geophysical and hydroacoustic study of the active submarine volcanism in the El Hierro Island (Canarian Archipelago). VIII Congreso Geológico de España, Oviedo (Spain) on 17–19 July 2012. Tsabaris, C., Kaberi, H., Alexakis, S., Georgopoulos, D., Kalfas, C. (2009). Environmental radioactivity applications in the Hellenic Centre for Marine Research.In Proceedings of the 9th Hellenic Symposium on Oceanography and Fisheries, 2009, Patras Greece. 1, pp. 105-108. 235

Tsabaris, C., Evangeliou, N., Fillis-Tsirakis, E., Sotiropoulou, M, Patiris, D.L., Florou, H (2012). Distribution of natural radioactivity in sediment cores from Amvrakikos Gulf (Western Greece) as a part of IAEA's campaign in the Adriatic and Ionian Seas, Radiation Protection Dosimetry, Volume 150, Issue 4, pp. 474–487, https://doi.org/10.1093/rpd/ncr436 Tsangaris C., Cotou E., Papathanassiou E., Nicolaidou A. (2010). Assessment of contaminant impacts in a semi-enclosed estuary (Amvrakikos Gulf, NW Greece): bioenergetics and biochemical biomarkers in mussels. Environ Monit Assess,161, pp. 259-269. https://doi.org/10.1007/s10661-008-0743-2 Underhill, J. R. (1989). Late Cenozoic deformation of the Hellenide foreland, western Greece. Geol Soc Am Bull, 101(5), 613-634. https://doi.org/10.1130/0016-7606(1989)101 Van Rensbergen, P., Depreiter, D., Pannemans, B., Moerkerke, G., Van Rooij, D., Marsset, B., Akhmanov, G., Blinova, V., Ivanov, M., Rachidi, M., Magalhaes, V., Pinheiro, L., Cunha, M., Henriet, J.P. (2005). The El Arraiche mud volcano field at the Moroccan Atlantic slope, Gulf of Cadiz. Mar Geol, Volume 219, Issue 1, pp. 1-17, https://doi.org/10.1016/j.margeo.2005.04.007 Vassilakis, E., Royden, L., Papanikolaou, D. (2011). Kinematic links between subduction along the Hellenic trench and extension in the Gulf of Corinth, Greece: A multidisciplinary analysis, Earth and Planetary Science Letters. Volume 303, Issues 1–2, pp 108-120, https://doi.org/10.1016/j.epsl.2010.12.054. Vogt, P., Gardner, J., Crane, K. (1999). The Norwegian–Barents– Svalbard (NBS) continental margin: Introducing a natural laboratory of mass wasting, hydrates, and ascent of sediment, pore water, and methane. Geo-Mar Lett, 19, pp. 2–21. https://doi.org/10.1007/s003670050088 Vogt, P.R., Cherkashev, G.A., Ginsburg, G.D., Ivanov, G.I., Milkov, A., Crane, K., Lein, A., Sundvor, E., Pimenov, N.V., Egorov, A.V. (1997). The Haakon Mosby mud volcano provides unusual example of venting. 236

EOS, Transaction, American Geophysical Union. 78, 48: 519, pp. 556- 557. Vött, A., Brückner, H., Brockmüller, S., Handl, M., Maya, S.M., Gaki- Papanastassiou, K., Herd, R., Lang, F., Maroukian, H., Nelle O., Papanastassiou, D. (2009). Traces of Holocene tsunamis across the Sound of Lefkada, NW Greece. Global Planet Change, 66, pp. 112-128. Wever, T,F, Abegg, F., Fiedler, H.M., Fechner, G., Stender, I.H. (1998). Shallow gas in the muddy sediments of Eckernförde Bay, Germany. Cont Shelf Res 18:1715–1739. https://doi.org/10.1016/S0278- 4343(98)00055-7 Yin, P., Berne, S., Vagner, P., Loubrieu, B., Liu, Z. (2003). Mud volcanoes at the shelf margin of the East China Sea. Mar Geol, 194(3- 4), pp. 135-149.https://doi.org/10.1016/S0025-3227(02)00678-3

237

6.2.2 Sogono The identical methodology and instrumentation as in Amphilochia Bay (see Subchaper 5.2.3) was followed in Sogono area, located on the western part of Amvrakikos Gulf (Fig.5.2.2.1). The geophysical and areal (MEDUSA platform) survey was conducted concecutively after the Amphilochia Bay survey in 2010 (see Subchapter 5.2.1).

Fig. 5.2.2.1 Location of Sogono in Amvrakikos Bay, Western Greece.

Based on the side scan sonar records, the Sogono area can be divided in two sub-areas according the level of backscatter intensity (reflectivity; Fig. 5.2.2.2). The southern area (water depth < 36 m) ischaracterized by high reflectivity and the northern area (water depth>36 m) is characterized by low reflectivity. The boundary of these sub-areas is an active fault as shown in the SBP profiles (Fig. 5.2.2.3). The southern high reflectivity sub-area is characterized by the intrusion of a salt dome in the upper sediments (Fig. 5.2.2.2).

238

Fig. 5.2.2.2 Side scan sonar mosaic from Sogono area.

Fig. 5.2.2.3 SBP profile showing the presence of the active fault which is theboundary of the sub-areas.

239

Fig. 5.2.2.4 SBP profiles showing the presence of a saltdome. Top to bottom profiles correspond to North to South positions shown in the side scan sonar mosaic (right;). Depth in TWTT. 240

Fig. 5.2.2.5 Map of sediment thickness derived from the seismic data.

The extraction of sediment thickness from the seismic data outlined the dome (blue area in Fig. 5.2.2.5). The sediment thickness over the piercement morphological structure is minimal, whilst on the nothern margin of the studied area the 3.5 kHz penetrated the seabed up to 30 m, due to the very thick fine sediments. An areal survey were performed in Sogono area Amvrakikos with the MEDUSA platform (more in Chapter 5.1.3) in correspondence with the salt dome and fault, that were previously detected in the seismic and side scan sonar records (Fig. 5.2.2.6).Dissolved methane values were very low (Fig. 5.2.2.6). Ogygen measurements confirmed that the seabed of Amvrakikos Gulf is hypoxic to anoxic, even on the western part and near the entrance Gulf (Fig. 5.2.2.6).

241

Fig. 5.2.2.6 The MEDUSA tracklines in corespondence to the CH4 (left) and dissolved O2 (right) measurements at the marine area of Sogono.

242

7. Synthesis The conclusion of this thesis can be summarised as follows: • Katakolo Bay results to be one of the most prolific thermogenic gas seepage zones in Europe (surely the biggest methane seep ever reported in Greece) and a natural laboratory to study gas bubbling in the sea and onshore seepage processes. The same type of thermogenic gas is seeping offshore and on shore in correspondence with two main normal faults converging to the harbour zone, which shows a spectacular field of gas bubble plumes. • Molecular and isotopic data suggest that the gas is related to a deep Petroleum System characterized by Jurassic carbonate reservoirs, Triassic source rocks, and Triassic evaporites. The contact between carbonates with hydrocarbons and evaporites

determine the production of H2S by Thermochemical Sulphate Reaction. Secondary alteration of gas (biodegradation) seems to occur at shallow depths, mainly in correspondence with the offshore seepage.

• Total offshore CH4 emission into the atmosphere (considering only ebullition flux) may range from 33 to 120 t y-1, over about

2 94,200 m . Onshore CH4 output was estimated to be around 89 t y-1 for only 15,800 m2 in 2010. These values, combined with the isotopic data, suggest the existence of deep pressurized reservoirs and a powerful petroleum seepage system, which may deserve further exploration. Assuming that the neotectonic regime has been established the last 3.000.000 yrs, this means

that a total amount of 99 to 360 million tones CH4 has been released from the Katakolo seepage zone. • Katakolo is an example of short-term variation of methane seepage into the atmosphere. Gas flux measurements in 2018 revealed considerable increase compared to the previous 2010 study. Methane emission increased from 57 to 225 kg d−1 (~4 243

times), with emission factor changing from ~4,000 to 15,000 t km−2 y−1. If similar sub-decadal, four-fold variations of seepage emission factor occur on large seepage areas on Earth, decadal changes of atmospheric methane budget can be affected. Thus, geological emissions contribution in the atmospheric methane should not be considered constant, as it is often assumed in atmospheric methane budget models. This study points towards the need for over time investigation of the emission factors of large seeps reported in the global inventories. • The area survey by MEDUSA and temporal monitoring by GMM benthic observatory in Katakolo, evidenced inverse correlations between oxygen and methane related to bubble pluems. GMM successfully performed a long-term monitoring of oxygen in a methane seepage site, where strong dissolved oxygen

variations (ΔO2 up to 200 μM): hypoxia and quasi-anoxia were observed at -7m water depth, in an open-sea-influenced site. Periods of dissolved oxygen decrease (hours) seemed to be

associated to enhanced CH4 seepage and low currents. Thus, hypoxic and quasi-anoxic events appear to be mainly controlled by current speed in more than 90 % of cases. • The gas flux measurements contributed to improve the global data-set of marine and onshore geologic emissions of methane, ethane and propane to the atmosphere; the bubble plume output data, in particular, can be used to refine the global statistics and emission factors of marine seeps, still today poorly assessed.

• In Amvrakikos Gulf there is seasonal hypoxia & anoxia mapped throughout the gulf. There is no gas seepage detected in the Western central zone (Sogono), but there is weak seepage detected over a pockmark in South-Eastern sector (Amphilochia 244

Gulf); where the CH4 increase may induce slight O2 reduction. The seeping gas is probably of mixed (microbial and thermogenic) origin. There are no significant variations in shallower oxycline. The main hypoxia drivers are confirmed to be oceanographic and man-made; the weak seepage can

further contribute to O2 reduction only locally, episodically and close to seabed.

The present complex and multidisciplinary study bears three main implications:

(a) It was demonstrated that seepage can be locally an important driver of hypoxia, especially in closed basins. Similar studies shall be performed in areas similar to Katakolo (for example offshore Spain, Rias Baixas or in the Caspian Sea).

(b) Katakolo is a major natural source of greenhouse for Greece. The emission data shall be examined and compared with other natural and anthropogenic methane sources in Greece to develop refined national greenhouse-gas emission inventories.

(c) It was demonstrated that short-term variation of methane seepage into the atmosphere is possible. Hence, it is neccessary to investigate over time the emission factors of large seeps and atmospheric methane budget models should stop considering geological emissions contribution in the atmospheric methane a constant.

245

Supplementary Material for Chapter 5.1.2 “Increased Methane Emission from Natural Gas Seepage at Katakolo Harbour (Western Greece)”

Photos S1-11

Photo S1.The main pier of Katakolo Harbour. Cracks are visible throughout the pier. The location of the two ~200 m fissures, the two manholes and the K137 crack are marked. Photo: S. Kordella, November 2018.

246

Photo S2. Cracks outside the Duty Free building located on the main pier of Katakolo Harbour. Photo: S. Kordella, November 2018.

Photo S3. Cracks outside the Duty Free building located on the main pier of Katakolo Harbour. Photo: S. Kordella, November 2018. 247

Photo S4. Cracks on the pavement outside the Port Authority building.Photo: S. Kordella, November 2018.

Photo S5. Crack on the asphalt outside the Port Authority building. Photo: S. Kordella, November 2018.

248

Photo S6. Cracks on the asphalt outside the Port Authority building.Photo: S. Kordella, November 2018.

Photo S7. Cracks on the pavement outside the Port Authority building.Photo: S. Kordella, November 2018.

249

Photo S8. Cracks on the asphalt at the parking area of Katakolo Harbour.Blue arrows represent methane exhalation. Photo: N. Georgiou, November 2018.

Photo S9. Cracks on the swelling asphalt at the parking area of Katakolo Harbour. Photo: S. Kordella, November 2018.

250

Left: Photo S10. Large crack developed during a 6.6 M earthquake on 26 October 2018, on a second pier (not affected by seepage) of Katakolo Harbour. Photo: G. Ciotoli, November 2018. Right: Photo S11.Large crack developed during a 6.6 M earthquake on 26 October 2018, on a second pier (not affected by seepage) of Katakolo Harbour. Photo: G. Ciotoli, November 2018.

a b c

Photo S12. The two manholes, (a) and (b-c), included in the methane flux measurements. From 2004 (photo b) to 2018 (photo c) the biggest manhole did not show significant changes in the structure and opening of the fissures 251

bordering the cap. The gas flux increased however considerably from 2010 to 2018 (see text). Photos a, c: G. Ciotoli, November 2018. Photo b: Photo: G. Etiope, 2004.

-1 Table S4. Basic statistics of CH4 calculated output (kg d ) from cracks, fissures, and manholes.

N Mean Median GM σ Min Max Including K137 102 1.23 0.007 0.006 10.1 0.0001 110.9 Excluding K137101 0.14 0.007 0.005 0.6 0.0001 4.6 N: Number of values, GM: Geometric Mean, σ: Standard Deviation Tables S5 and S6 Tables S5 and S6 report the number and size of the 177 non-measured

-1 cracks and fissures. In the third column, the estimated CH4 output (kg d ) is calculated based on the non-measured cracks and fissures size (length or area) and the mean chamber-size emission factor (excluding the large emission from K137 fissure): 0.05 kg d-1. The total estimated methane output from linear and non-linear non-measured cracks and fissures results in ~100 kg d-1 (~36.5 t y-1).

Table S5. Estimation of CH4output from linear non-measured fissures, based on the mean chamber-size emission factor (0.05 kg d-1, excluding the large output of the K137 fissure): and number of 0.3 m segments (derivable by the length divided by 0.3 m).

-1 Count Length (m) Estimated CH4 output (kg d ) 1 26 4.3 65 2 21.5 2 198 65.3 1 22 3.6 1 24 4.0 Total methane output 98.7

252

Table S6.Estimation of methane emission from, non-measured, non-linear cracks, based on the mean chamber-size emission factor (0.05 kg d-1, excluding the large output of the K137 fissure).

2 -1 Count Area (m ) Estimated CH4 output (kg d ) 100 >0.07 5.0 7 1 5.0 Total methane output 10.0 Figure S1

-1 Fig. S1. Frequency histogram of the CH4 calculated output (kg d ) from 101 cracks, fissures, and manholes, excluding the large emission from K137 fissure.

Figure S2

Fig. S2. Annual sequence of 146 seismic events ≥4.5 M that occurred starting from the reconstruction of the modern Katakolo Harbour (1995) until the gas flux survey presented in this work (November 2018). The seismic event of M 6.6 occurred in October 2018 (2 weeks before our gas survey) produced the large crack shown in Photos S10-S1.