Deep and surface factors of local gas-saturated zones formation with anomalously high gas pressure and gas emission craters in frozen soils

Khimenkov Aleksandr Nikolaevich PhD in Geology and Mineralogy Leading Scientific Associate, the Institute of Geoecology of the Russian Academy of Sciences 101000, Russia, Moskva oblast', g. Moscow, ul. Ulanskii Proezd, 13, stroenie 2 a_khimenkov@mail.ru Other publications by this author

Stanilovskaya Yuliya Viktorovna Permafrost Infrastructure Interaction Specialist, Total 101000, Russia, Moscow, Ulanskii Pereulok Street 13 e-mailyulia.stanilovskaya@total.com Other publications by this author

Introduction

Zones of abnormally high pressure in frozen rocks, in recent years, have attracted increased attention of researchers. This is due to numerous gas emissions and accidents during drilling of frozen rocks, as well as gas discharge funnels discovered in the north of Western Siberia. Currently, a significant amount of data has accumulated on the manifestation of natural explosive processes in permafrost rocks, leading to the formation of gas emission funnels. The number of examined craters is currently approaching 20. If we analyze the proposed hypotheses of their genesis, we can notice several general provisions. Firstly, they all proceed from the fact that the cause of natural explosions is an in-ground gas under considerable pressure ^{[1, 2, 3, 4, 5]}. Secondly, all hypotheses are based on the idea of the stages of the processes of preparation of natural explosions, ^{[4, 5, 6, 7]}. Thirdly, the role of the thermal factor in creating conditions for the occurrence of a gas-saturated zone in frozen rocks is noted everywhere. Either cooling occurs when the taliks freeze ^[5], or the temperature of frozen rocks increases when the gas hydrates contained in them decompose ^[4], or the frozen rocks are warmed up by warm deep gases coming from below ^{[6, 7]}. In the case of an increase in the temperature of frozen rocks, a complex of internal processes is activated in them: the amount of unfrozen water increases, mechanical strength decreases, volumetric thermal expansion of the gas component begins, filtration capacity increases, gas hydrates begin to decompose, the number of cryopags increases in saline soils, etc. With a decrease in temperature and directed freezing of thawed rocks, a complex of cryogenic processes occurs: cryogenic concentration and localization of groundwater, intra-soil gases, salts. In frozen rocks, filtration decreases with increasing strength and deformation properties. Conditions are being created for the formation of gas hydrates.  All this leads to the emergence of local zones of concentration of free gas with high pressure or hydrate-containing frozen rocks. It was stated ^[4], according to which, in these zones, when heated, local gas-dynamic systems begin to form, the stage-by-stage self-development of which can eventually lead to an explosion and the formation of a gas emission funnel. The heterogeneity of surface, geological and tectonic conditions determine various scenarios for the preparation of explosions ^[8].

Hypotheses of the formation of gas-saturated zones with increased pressure and associated funnels of gas emission

            The study of gas discharge funnels and associated gas-saturated zones with abnormally high pressure began quite recently, starting in 2014. Let us briefly consider the main hypotheses of their formation formulated to date. The article will use the names of gas discharge funnels that have already been established in the scientific literature on this topic.

Freezing of taliks

Some elements of the structure of gas discharge funnels allow us to consider their genesis from the positions formulated at the very beginning of the emergence of geocryology as a scientific direction. This, first of all, concerns the formation of long-term heave mounds preceding gas emissions. It is obvious that the traditional ideas about the formation of mounds due to the comprehensive freezing of the talik, the formation of gas-saturated zones and subsequent pneumatic explosions should have been realized in the corresponding hypothesis. This hypothesis was presented most convincingly by Yu. B. Badu and K. A. Nikitin ^[9]. They proposed a mechanism for the formation of gas-saturated zones and the development of gas explosions forming funnels, as the final result of the freezing of the talik and the formation of a bulge of heaving according to the classical scheme in an open or closed system. A typical section of the heaving hill has the following typical structure: in the upper part there is a frozen peat layer with a thickness of up to 1 m, under it there is a mineral soil, usually consisting of layered lake sediments. The thickness of this reservoir varies greatly from one to several meters. Below is the horizon of taberal deposits (thawed and then frozen again). The mineral soil is underlain by ice, forming a dome-shaped vault with its surface, the thickness of the ice can reach tens of meters. The growth rate varies from a few millimeters per year, to 0.5 m or more per year. The apical parts of the ice are enriched with large air inclusions, due to the release of air during ice formation, as a result of which extensive air cavities are formed here ^[10]. The sizes of ice bodies vary within large limits: the diameter of the base is from 20 to 250 m, and the height is from 2 to 70 m. Growing mounds are often underlain by water lenses up to 2 m thick (Fig. 1) and having hydraulic pressure ^[11].

Fig. 1 Diagram of the structure of a growing pingo in Northern Alaska ^[11].

A kind of gas trap is formed under the growing bulge of heaving, in which the gas saturates the water lens, and after its freezing is incorporated into the ice core. Gas in sedimentary rocks accumulated even during their primary formation, and was preserved during the formation of talik. When the layer of frozen soil forming the upper part of the heave hillock is destroyed by heave forces, the core ice will be exposed and gas (methane) begins to evaporate intensively from it. When the concentration of methane in the air is 9 -16%, the hill explodes. In this case, the gas is ejected and ignited instantly from both the ice core and the aquifer. The authors believe that potentially explosive heaving mounds are formed in a specific geocryological situation – when taliks freeze in gas–saturated rocks only within the area of gas-bearing systems, in anticline traps from a lithological pair of clay-sand layers.

S. N. Buldovich and co-authors ^[5] proposed a similar scheme for the specific conditions of the formation of the Yamal crater, also based on traditional ideas about the development of long-term heave mounds (Fig. 2). In their opinion, the heave mound at the site of the Yamal crater collapsed under cryogenic hydrostatic pressure accumulated in a closed system of freezing talik. This happens before the freezing is complete, when the core of the unfrozen wet soil included a large amount of carbon dioxide that reached the maximum possible gas saturation, while the excess pressure exceeded the strength of the overlapping frozen layer. When the long-term bulge of heaving explodes, a cylindrical crater is formed, corresponding to the remaining talik core.

Fig. 2 Freezing of the atmospheric talik and the formation of a bulge of heaving: (A) talik under the lake; (B) lateral freezing of the talik during shallowing of the lake; (C) the formation of a freezing closed talik; (D) the growth of a bulge of heaving ^[5].

Intake of deep warm gas

V. I. Bogoyavlensky proposed a model for the formation of gas discharge funnels due to the flow of deep warm gas through faults into the surface layers of frozen rocks ^[6]. According to this model (Fig. 3), the formation of gas-saturated cavities with high pressure in the strata of icy frozen rocks occurs under the influence of endogenous processes (abnormal heat flows, increased pressure, physico-chemical reactions, gas-hydrodynamic processes) confined to faults. At the initial stage, under the influence of an abnormal heat flow and the supply of warm gas, a channel of thawed rock is formed in the fault zone, suitable from below to the ice-saturated upper horizon of frozen rocks.  At the first three stages, the underground ice is intensively pulled out from below and the thawed supply channel is expanded, through which the gas under pressure is pumped into the formed cavity. Subsequently, the frozen roof cannot withstand the pressure coming from below and a pneumatic explosion occurs, forming a crater. It is stated that the presence of thawed channels of gas inflow into the cavity is a prerequisite for its accumulation and subsequent release (explosion) ^[6].

Fig. 3 A complex schematic model of the formation of cavities in underground ice arrays, heave mounds and gas emissions with the formation of giant craters in the Earth's cryosphere. Designations: 1 — ice, 2 — water, 3 — thawed rock in the fault zone (talik), 4 — thawed re—deposited rock, 5 — frozen rocks (including active/seasonally thawed layer), 6 — gas, 7 - ascending and descending flows of water (a) and gas (b) along the fault, 8 — ascending gas flows in the water column and atmosphere, 9 — fault (subvertical cracks) ^[6].

Decomposition of gas hydrates due to surface heat sources

A. N. Khimenkov and co-authors proposed a mechanism for the formation of gas-saturated zones with subsequent gas emission and the formation of a funnel, due to the decomposition of gas hydrates, during the thawing of frozen rocks under surface reservoirs (Fig. 4) ^[4]. Gas discharge funnels are the result of self-development of local gas-dynamic geosystems formed in the thickness of permafrost rocks. The preparation of the explosion takes place in several stages ^[8].

1. The starting point that caused all subsequent events is the local warming of the permafrost strata (MMP). A talik and a zone of thawing of frozen rocks with higher negative temperatures than in the surrounding permafrost rocks are formed under the surface reservoir ^[12]. Under the action of local heating under a surface reservoir in frozen rocks, the temperature increases within negative values. In the area of local warming of the MMP under the lakes, the temperatures of permafrost rocks are in the range -1 - -3 ° C.

2. After the temperatures in the gas hydrate layer exceed the values ensuring their stable state, the dissociation process begins with the release of methane, with an initial pressure of 2.2 - 2.6 MPa ^[13].

3. The gas under pressure begins to filter into the least durable frozen rocks of the high-temperature zone, simultaneously deforming them. A paragenetic relationship is established between gas filtration and plastic deformations, which determines the self-development of the geosystem. Gas under pressure penetrates into the frozen rock, which significantly weakens its strength and causes plastic deformation of ice. The cracks and dislocations that have appeared accelerate gas filtration.  The removal of gas (due to filtration) stimulates the process of dissociation of gas hydrates and equalizes the gas pressure to the previous level. This process maintains high pressure in the filtered gas bubbles. Gradually, in the area of a single filtration space, an ice-ground gas-saturated rod permeating an array of permafrost rocks is formed. Along the entire height of the rod, the pressure in the gas bubbles will correspond to the values observed in the dissociation zone. High pressure in the permafrost massif with simultaneous pressure from below will lead to the movement of the ice-ground rod. At the same time, both in the stem and in the contacting layers of the host rocks, a zone of annular cracks oriented parallel to the flow is formed. This zone is an additional channel for gas supply to the upper horizons.

4. The upward movement of the gas-saturated ice-ground mass leads to plastic deformations of the upper layer of thawed rocks (talik), its buckling and freezing. As a result, a low–temperature gas-tight, durable 6 - 8-meter layer of permafrost is formed, which is a screen for the gas flow moving from below. During its deformation of the screening horizon, under the influence of pressure from below, a bulge of heaving develops.

5. After the plastic deformations of the frozen roof reach the limit values, its rupture occurs and the release of ice-ground material saturated with gas under increased pressure ^{[4, 8]}.

            Various initial conditions, as well as the variety and intensity of the processes occurring during the preparation of the explosion, cause various scenarios for the development of gas discharge funnels: from the gas outlet into the talik (in the case of a deep reservoir existing for a long time) to an explosion without the formation of a hillock (in the case of gas hydrates at a shallow depth, and their rapid decomposition).

Fig. 4 Stages of development of the Yamal crater (I, II, III, IV).

Designations: 1 — cover horizon; 2 — layer of ice between thawed and frozen rocks; 3 — frozen gas—saturated ice crater with traces of plastic deformations; 4 -infiltration—segregation ice; 5 — gas—tight roof of the MMP; 6 - temperature rise zone in the MMP under the lake; 7 - MMP outside the warming effect of the lake; 8 — hydrate—containing MMP layer; 9 — decompression zone in the hydrate—containing MMP layer adjacent to the crater; 10 — direction of fluid movement; 11 — gas fluids; 12 — grottoes and caverns in the lower part of the crater; 13 - lake; 14 - crater formed after the release of gas-saturated ice ground; 15 — talik; 16 – subvertical layering in frozen rock and ice ^[8].

In the North of Western Siberia, the lowest temperatures (-6 ? -8 ° C) are characterized by frozen rocks on elevated and vegetation-free watershed areas of the sea plains. Higher (from -2.5 to -4.5 ° C) is the temperature in more drained floodplain areas, often occupied by willows. In the lower parts of the gentle leeward slopes, as well as in the bottoms of dens, runoff hollows, ravines overgrown with dense willow, the temperature of the rocks rises to 0 - -4 ° C. The most typical capacities of the seasonally shallow layer are from 0.3 - 0.8 m on peat bogs and poorly drained torn-off surfaces of watersheds and floodplains with sedge-moss vegetation to 0.8- 1.5 on drained sections of watersheds and gentle slopes of terraces with shrub-moss-lichen vegetation. The formation of reservoirs has a warming effect on permafrost rocks. The results of forecast thermal engineering calculations show that at the base of a reservoir with a depth of 0.5 m, the average annual temperatures of permafrost rocks during a ten-year period increase by 2.0 - 4.0 ° C. At a depth of 1.5 - 1.8 m, the bottom soils do not freeze in winter and the formation of talik begins ^[12]. Analysis of the materials of the study of gas discharge funnels found in the north of Western Siberia shows that surface conditions can have a significant impact on the formation of gas discharge funnels. All the discovered craters are confined to the so-called "warm" permafrost landscapes: near a reservoir (Deryabinskaya funnel), in Khasyreya (Yamal crater), on the banks of a watercourse overgrown with bushes (Yerkutinskaya funnel), sandy swells (Antipayutinskaya funnel), in a moistened hollow (funnel discovered in 2020), under a surface reservoir (Seyakhinskaya funnel). The continuity of the MMP from the surface is disrupted by non-penetrating under-conditions and suspicious taliks. Water-thermal taliks are the most widespread, radiation-thermal taliks are much less common.  The former are formed and exist under riverbeds with constant and seasonal runoff and under lakes, the latter are confined to local sections of river floodplains, bottoms of logs and flow hollows.

A synthetic hypothesis involving the interaction of talik freezing and the flow of gas from the underlying horizons.

            The hypotheses discussed above are related to certain factors affecting the development of natural explosive processes in frozen rocks. These are: 1) heat exchange processes in the upper horizons of frozen rocks (freezing/thawing), leading to cryogenic concentration of underground gases; 2) the entry of deep heated gas into the surface horizons of ice-saturated rocks; 3) decomposition of gas hydrates contained in frozen rocks under the influence of surface conditions, primarily surface reservoirs. There are a number of synthetic hypotheses that combine surface and deep factors. Separate mechanisms are selected to explain the development of the gas-dynamic geosystem from the beginning of the localization of the gas phase to the pneumatic explosion forming a funnel of gas emission. Since the selection is based on hypothetical constructions that reflect only the author's preference, their combination can be very different. As an example, we give the hypothesis proposed by E. M. Chuvilin and co-authors (Fig. 5). It includes surface factors (formation and freezing of the atmospheric talik) and the intake of deep gas ^[14]. The preparation of the explosion goes through certain stages.

Stage I: migration of deep gas through frozen rock and atmospheric with subsequent release into the atmosphere.

Stage II: saturation of the all-round freezing zone of the talik with deep gas.

Stage III: deformation of the overlying frozen layer of rocks due to gas pressure, formation of a bulge of heaving ^[14].

Fig. 5 Accumulation of gas and increase of its pressure in the freezing talik saturated with migrating deep gas: (I) talik under the surface reservoir before freezing; (ii) saturation of the freezing talik with deep gas; (iii) deformation of the roof of frozen rock under the influence of gas pressure and the beginning of the growth of the hillock ^[14].

Discussion

            The hypotheses discussed above reflect the views of their authors on the processes of formation of gas-saturated local zones and funnels of gas emission in permafrost rocks. It should be considered whether these representations correspond to the available data on natural conditions, geological and cryogenic structure, as well as the characteristics of the environment in which the processes (temperatures, pressures) forming these formations were realized.

The position of proponents of the development of explosive processes that form gas discharge funnels due to the freezing of taliks is quite understandable. It consists in the fact that gas emissions are preceded by the formation of long-term heaving mounds, the mechanism of which is well studied. A formal conclusion is made, it always precedes, so it is always formed according to the traditional scheme. Consequently, this traditional mechanism is the only possible one to explain the reasons for the formation of gas-saturated zones with abnormally high pressure and the subsequent stage - the formation of gas discharge funnels. According to generally accepted concepts, the freezing of taliks leads to the localization of groundwater, the creation of zones of increased pressure and the formation of a bulge above them. In Yamal, V.I. Bogoyavlensky and his colleagues identified more than 7 thousand (7185) ^[15], and concluded that there is a huge risk of gas emissions associated with this. Yu.B. Badu and K. A. Nikitin ^[9] limit the formation of gas-saturated zones and associated gas explosions under mounds, territories of hydrocarbon deposits. Here, through faults and cracks, gas flows to the surface of the Earth. Under the growing heaving mounds, the freezing front forms domed irregularities in which gas and water accumulate, the freezing of which forms gas-saturated ice. The subsequent destruction of the hillock leads to the release of gas, its ignition and explosion ^[9]. With all the external evidence of the evidence cited, there are data that allow us to doubt these representations.

The ignition of gas and the formation of a funnel of gas emission due to a chemical reaction was not observed in all cases. Where there was no ignition, there was a pneumatic release of frozen material with the formation of cylindrical channels. In this case, the pressure required to release the rock is determined by the strength of the frozen roof. We present the results of calculations of the necessary pressures for the formation of the most studied funnel of gas emission "Yamal crater".  According to V. I. Bogoyavlensky and I. A. Garagash, a pressure of 1.25 MPa is sufficient to destroy the frozen tire of a funnel with a capacity of 8 m. A further decrease in pressure leads to the fact that the stresses in the tire do not reach the ultimate strength, deformations remain elastic, the fracture surface is not formed and rock ejection does not occur ^[16].  Close values (1.74 MPa) (Table. 1), were obtained by V. P. Merzlyakov ^[4]. The pressures observed in the lenses of water at the base of perennial heave mounds are significantly lower. D. R. McKay, who studied perennial heave mounds (pingo) on the Arctic coast of Canada, measured the pressures in water lenses lying at the base of the mounds. The obtained values of the hydrostatic pressure did not exceed 0.35 MPa, with the thickness of the frozen thickness of about 25 m. This pressure was sufficient for the deformation of the frozen roof and the formation of pingo, but not enough for the development of the explosive process ^[17]. It should be taken into account that during the formation of heave mounds, numerous discontinuous deformations of the frozen roof constantly occur, which should lead to an outpouring of water on the surface, and a release of gas pressure. Of course, these ruptures heal when the water freezes, and the pressure is restored, but it will not reach the large values necessary for the development of the explosive process. 

Some features of the structure of the funnels of gas emission are not noted in the structure of the "classic" perennial heave mounds. For example, ring structures found in the walls of some craters (Fig. 6). The study by E.I. Galeeva and co-authors showed that the layering is due to the viscoplastic flow of ice. Shear deformations lead to the formation of folds, secondary layering (cleavage) oriented at an angle of up to 60° to the horizontally lying primary layering ^[18].

Fig.6 The subvertical ice layering of the ring structure bordering

Yamal crater July 2014 ^[19].

These formations indicate the movement of a gas-saturated ice-ground massif relative to the stationary enclosing thickness of icy frozen rocks. Some formations recorded in gas emission funnels do not fit into the classical ideas about the processes that form heaving mounds. These include: zones of grottoes and caverns located in the lower part of the funnels (Fig. 7); ice-ground blocks pressed into the thickness of frozen rocks, found in the walls (Fig. 8), etc. ^[8]. Geophysical studies have not found talic zones under the funnels ^[20], but they should have been located under the heave mounds, in the case of traditional ideas about their formation (Fig. 10).

Fig. 7 Large grottoes formed due to the merger

a series of small grottoes and caverns. November 2014 Photo by V.A. Pushkarev.

Fig. 8 Deformed gas-saturated ice "pushed" into

layered ice-ground array. July 2015 Photo by A.V. Lupachev.

Separately, it is necessary to consider the mechanism of formation of funnels of gas emission, proposed by S. N. Buldovich and co-authors ^{[5, 21]}. According to their ideas, the formation of the Yamal crater is associated with the explosive destruction of the atmospheric talik. The thickness of the frozen soils above the talik at the time of its destruction ranged from 7 to 9 m. The Talik was a water-soil gas-saturated mixture with dissolved gas (mainly carbon dioxide) of bacterial origin. The authors' hypothesis is based on the analysis of samples taken from depths of 10-17 m, and corresponds only to the upper part of the crater (total depth of 60 m). Deposits in which an increased gas content is traced form a lenticular body with a capacity of about 10 m. The largest total gas content is confined to the central part of the lens. This layer corresponds to the bilateral freezing of the atmospheric talik. A similar structure is observed with respect to the content of carbon dioxide. In the central part, its highest content is observed, with a decrease in the lower and upper parts of the lens. With respect to the methane content, an inverse relationship is observed in the central part of the lens, its content is the lowest, and in the underlying and host sediments the greatest (Fig. 9). This distribution of gas composition indicates that in the upper 10-meter part of the frozen sediments saturated with methane, a talik saturated with carbon dioxide was formed under the influence of the lake. The subsequent two-way freezing of the lenticular talik, due to cryogenic concentration, the carbon dioxide content led to its observed distribution.

Fig. 9 Gas content in underground ice ^[21].

The data given indicate that the capacity of the talik under the lake did not exceed 10 - 15 m. In the section, this talik corresponds to a lens of icy layered rocks, with a high content of carbon dioxide. It is embedded in a frozen massif with a high methane content (Fig. 9). Studies conducted in the summer and autumn of 2014 showed a high methane content in the air inside the funnel and in air bubbles in the ice in the lower part of the crater ^[22]. The same results were obtained by M.O. Leibman and co-authors, according to which the methane content in the water of the lake formed at the bottom of the Yamal crater significantly exceeds the values in the water of ordinary lakes (approximately 500 - 900 ppm compared with 15 ppm on average for surrounding lakes) ^[23]. The mechanism of formation of gas discharge funnels proposed by S. N. Buldovich and co-authors is based on materials obtained for the upper layer of frozen sediments (10-17 m). But the depth of the Yamal crater was about 60 m . Below 10-15 meters, frozen rocks are saturated with methane, but not carbon dioxide. Conclusions about the leading role of talik freezing can be confidently extended to the upper part of the section, but for the entire thickness in which the Yamal crater is formed, this assumption cannot be considered convincing, it requires additional justification.

The materials considered show that the mechanism of freezing of taliks does not explain many features of the formation of funnels of gas emission. Probably, conditions will be found in the future when this will be possible, but even in this case the process will not be "classical". Undoubtedly, it will be necessary to make adjustments to the traditional ideas about cryogenic processes in freezing precipitation with an increased gas content in them

The "Complex schematic model of the formation of cavities in underground ice arrays, heaving mounds and gas emissions with the formation of giant craters in the Earth's cryosphere" proposed by V. I. Bogoyavlensky is implemented in the conditions of the near-surface part of the MMP section, which includes reservoir ice deposits. The realization of this hypothesis is due to the interaction of an abnormal heat flow and a gas-hydrodynamic process in the talik zone, confined to the fault ^[6]. The following main provisions of the proposed model can be distinguished:

- it is necessary to have formation ice in the section of frozen rocks;

- formation of a cavity filled with gas in the lower part of the ice mass, the formation of which occurs due to abnormal heat flow;

- a prerequisite is the presence of gas supply channels into the cavity;

- formation of a melt zone at the base of the cavity exceeding the transverse dimensions of the gas supply channel;

- the flow channel of the gas supplied from the bottom up is at the same time a channel for diverting the flow of water directed from top to bottom;

- saturation of the cavity with gas in the free state and an increase in pressure in it above the hydrostatic (AVPD) due to the subvertical migration of gas from high-pressure zones lying below the permafrost distribution area.

Each item is mandatory for the implementation of the mechanism for the formation of the AVPD of gas and gas discharge funnels, and the absence of at least one makes the preparation of the gas release impossible.

Let's consider how well these conditions are met. First of all, this concerns the mandatory requirement of the existence of funnels at the base, channels with gas moving through them with a positive temperature and abnormally high pressure. Comparison of V. I. Bogoyavlensky's model ^[6] with the results of geophysical studies in the Yamal Crater area showed their inconsistency with each other. In the Yamal crater zone, based on the study of which the main provisions of the hypothesis under consideration were developed, there is no melt channel (Fig. 10). The Talik zone, which should have formed at the base of the Yamal crater, did not manifest itself in any way. According to V. V. Olenchenko and co-authors ^[20], the large electrical resistances and polarizability of the first layer are explained by the increased iciness of the upper part of the section and the presence of formation ice. In the depth range of 60 – 80 m, there is a low-power layer with an abnormally high UES (400 - 880 ohms? m). The abnormally high value can be explained by high iciness or a large number of inclusions of relic gas hydrates (methane clathrates). As a rule, gas hydrates are deposited in reservoirs representing horizons with layers of sand or sandy loam with a reduced salt content ^[24]. The position of the layer with an abnormally high UES correlates with the depth of the maximum occurrence of gas occurrences in the section of the Bovanenkovsky gas condensate field. Below, the underlying layer of rocks with a WES of 7.5 - 11.0 Ohms is interpreted as frozen Lower Middle Pleistocene marine sediments of the Yamal series. On the geoelectric section at a depth of 135 - 190 m, the roof of the conductive layer is distinguished, which is interpreted as the water–ice phase boundary in the soil, i.e. the sole of the permafrost layer ^[20]. According to drilling data, in the alluvial floodplain of the Mordyakha River (sq. 610-P-3) loams have a massive cryotexture, and at a depth of 165 m the sole of the MMP was opened ^[25].

Fig. 10 Geological section in the cryogenic crater area ^[3].

The Yamal crater is located in the intersection zone of supposed faults that are not distinguished by geophysical methods. It can be assumed that the fault really exists, and the gas comes from the underlying horizons. In this case, the absence of a talik under the funnel should be taken into account, that is, the gas temperature should be negative. Therefore, it is necessary to consider the conditions of gas filtration in frozen rocks. At a positive gas temperature, the formation of a melt channel would occur, which is not observed in real conditions. And, of course, an intra-soil cavity measuring tens of meters across would not have formed due to the pulling out of the formation ice. The mechanism of movement of two multidirectional flows in one channel is not fully understood: the filtration of gas (under pressure) upwards and the movement of water downwards. In this case, the water must overcome the higher pressure directed from the underlying horizons.  The study of the cryogenic structure of the walls of the Yamal crater and other craters showed that during the preparation of the gas release, structures are formed that cannot be explained from the standpoint of the mechanism under consideration, such as ring structures in the walls of the crater, deformations of the primary stratification, the movement of individual blocks of rocks in the frozen massif (see the previous section). The same can be said about the genesis of gas. Obviously, with this mechanism, the gas must be thermogenic, that is, it must come from the underlying horizons. The analysis of the carbon isotope composition of methane (?13 C) from the Yamal crater showed its "bacterial origin" (from -58% to -75%), while in one sample, ?13 C is close to -45% "thermogenic methane" ^[26]. This sample may indicate a possible limited inflow of gas into the Yamal crater from Upper Cretaceous deposits lying below the sole of the MMP. In the vault of the Bovanenkov rise, the Cenomanian roof lies at a depth of about 500 m – only 300 - 330 m below the permafrost sole. Methane from the Seyakhinskaya funnel (formed on June 28, 2017), according to the results of the analysis conducted by F.M. Rivkin (Yamal LNG OJSC), unequivocally indicate its biogenic genesis (?13 C = -80.6 %) ^[27]. These and other questions are not answered in the "Complex schematic model of the formation of cavities in underground ice arrays ..." and they undoubtedly need to be solved in order for this model to be used in practice.

When justifying the possibility of the formation of gas discharge funnels due to the inflow of deep gas, the identification of not hypothetical, but real supply gas channels could be the most convincing argument. Geophysical methods allow us to do this successfully. On the offshore shelves, gas channels that bring gas from great depths to the surface, gas accumulation zones (gas pockets), and places of explosive gas outlet to the surface (pokmark) are successfully identified. Complexes combining pokmark, gas pockets and gas pipes are fluid dynamic gas-saturated geosystems. Each element of this geosystem corresponds to a certain stage of its development. The central part (Fig. 11) shows an example of the complete development of this geosystem.  The paragenetic connections of the pokmark, the gas pocket and the gas channel are clearly visible. Directly under the Pokmark crater lies a gas pocket, which was formed due to the supply of gas through a gas channel from deeper horizons. The accumulation of gas in the gas pocket and an increase in pore pressure in the upper part of the sedimentary cover led to a pneumatic explosion, the release of gas into the aqueous medium and the formation of a pockmark. In the gas pocket, after pressure relief, the accumulation of gas flowing through the gas channel continues. The right part of the figure shows an example of the incomplete development of a gas-dynamic geosystem. In this case, two of the main stages of development are expressed, the formation of a gas channel and the formation of a gas pocket. The accumulation of gas in the gas channel has not yet led to an emission.

Fig. 11 Seismic profile of deep gas migration in marine sediments

^[28] (with additions).

If the sedimentary layer consists of permeable sediments and there are no screens from unfiltering horizons, a fluid dynamic gas-saturated geosystem may consist only of a gas channel (gas pipes). During geophysical studies, gas pipes, in the form of vertical zones of correlation loss, appear throughout the visible sedimentary column, up to the bottom surface. However, more often the "pipes" do not reach the bottom surface, in addition, they may look not only as clear "pillars" of correlation loss to transparency, but also simply as elongated signal attenuation zones ^[29].

It should be noted that the author of the hypothesis under consideration has other ideas about the movement of deep gas to the surface. In accordance with them, permafrost rocks existing in most of the northern (Arctic) land areas and on the shallow shelf (up to the isobath of about -120 m) are a regional fluid barrier (tire). The gas moving in the subvertical direction, encountering an obstacle in the form of an MMP, begins to spread in the subhorizontal direction, thus forming large extended deposits. Getting into weakened zones (faults, talics), the gas breaks through to the surface ^[27]. Here we are no longer talking about the mandatory formation of taliks during the development of gas-saturated zones in permafrost rocks. For the Bovanenkovsky gas condensate field (GCM), significant gas flows from frozen rocks are noted, reaching hundreds and thousands of ^m3/day. with a high (up to 90%) degree of filling of ground pores with ice and unfrozen water. For example, in Yamal, well 64-P-2 uncovered a gas deposit at a depth of 72-80 m, from which a gas inflow of ³ thousand m3 was obtained, which later decreased to 500 ^m3/day. The diameter of the deposit is estimated at 320 m (an area of 80 thousand ^m2), and gas reserves are about 0.5 million. m ^{3 [30]}. The deposit is located in the thickness of frozen rocks. Almost all gas manifestations at the Bovanenkovskoye gas field are confined to detached dusty sands occurring in the section of the frozen strata to depths of about 130 m. In turn, more than 85% of gas occurrences in the deposits of the Yamal series are recorded at depths of about 60-80 m (Fig. 12), where it is possible to distinguish a horizon sustained over an area of about 120 ^km2. Gas manifestations are not recorded at great depths composed mainly of clay rocks. That is, the confinement of gas-saturated zones is rather determined by areal geological causes, rather than local tectonic disturbances.

In our opinion, the mechanism of formation of gas-saturated zones without the obligatory formation of a thawed cavity in frozen rocks is more realistic, although in this case a deep theoretical study of the conditions for its implementation is required. Nevertheless, the idea of the significant role of gas supply from the underlying horizons in the saturation of frozen strata with gas is extremely interesting and important, both in theoretical constructions and for practical application. Undoubtedly, this problem needs to be solved within the framework of geocryology, since all processes are implemented in frozen rocks. At the same time, it should be recognized that the theoretical and practical aspects of this topic are still poorly developed.

A number of indirect signs indicate the widespread use of the gas hydrate form of finding gas in frozen rocks. The prevalence of gas occurrences has an areal regional character, and is not confined to local zones of tectonic disturbances. This indicates the regional features of cryolithogenesis associated with epigenetic freezing of marine gas-saturated facially inhomogeneous sediments. V.S. Yakushev ^[33] believes that the formation of gas hydrates occurs during epigenetic freezing of sediments, when, with a decrease in their temperature, the front of hydrate formation outstrips the front of phase transitions of pore moisture into ice. At the same time, the gas previously concentrated in front of the freezing front in the sand lenses turns into a clathrate form. After passing the freezing front, the gas hydrate partially decomposes into water and gas, while the temperature drops sharply. The resulting water, freezing, forms an impenetrable ice film that preserves hydrate, preventing it from further decomposition. This allows gas hydrates to remain in a metastable state under nonequilibrium conditions at shallow depths.  For the entire thickness, the occurrence of gas emissions is traced to rocks with reduced salinity, as well as an increase in the total salinity of rocks below the horizons of recorded gas occurrences, which may indicate cryohydrate extraction of salts. The stability of the "preserved" gases depends on the macrostructure of the rock, temperature, the possibility of sublimation of moisture from the hydrate surface, salinity, and the presence of mechanical action ^[31]. It is important to note that some gas occurrences are recorded in the formation ice, since a strong smell of gas is recorded during the penetration of some ice bodies ^[31].

Fig. 12 Typical section of Bovanenkovsky deposits  GCM ^[32].

The role of gas hydrates in the cryogenic structure of rocks in the north of Western Siberia is clearly illustrated by the cryohydrate profile of the Yamal region in frozen rocks of various gas and gas fields (Fig. 13) ^[34]. It shows that the layer of metastable gas hydrates is widespread in Yamal, for example, at the Bovanenkovsky NGCM, it can be traced to the entire thickness of frozen rocks.

Fig. 13 Regional cryohydrate profile of the Yamal region, for example, S-S –U-V (Rusanovskoye, Kharasaveyskoye, Bovanenkovskoye, Sredneyamalskoye, Arctic, Yamburgskoye deposits) ^[34].

The occurrence of accumulations of gas hydrates in the gas fields of the north of Western Siberia is explicable to the wave. The more gas is contained in rocks before freezing, the greater the probability of accumulation of gas hydrates during the epigenetic formation of frozen strata. 

Considering the problems of dissociation of gas hydrates, it is necessary to assess the role of self-preservation in the heating of metastable gas hydrates in real conditions. The very existence of metastable gas hydrates is due to self-preservation. The creation of an impenetrable ice layer stops the decomposition of gas hydrates and allows them to be preserved in non-equilibrium conditions. Thus, the process of self-preservation "prohibits" the dissociation of gas hydrates to play a significant role in the saturation of frozen rocks with gas. Nevertheless, data are accumulating that the intensity of the self-preservation process during the decomposition of gas hydrates is largely determined by the granulometric and mineralogical composition, as well as the salinity of the host frozen rocks. V. S. Yakushev and co-authors ^[35] during laboratory experiments, the behavior of metastable relic methane hydrates in clay saline sediments extracted from the Yamal crater was studied. Heating of the samples in the temperature range -6.75 - -6.57 °C led to the complete decomposition of the gas hydrates contained in them. Studies by D. A. Davletshina have shown that the intensity of decomposition of gas hydrates largely depends on the content of clay particles. With an increase in their content, the intensity of dissociation increased. In a sample of sand with montmorillonite particles, dissociation proceeds more intensively than in sand containing kaolinite particles. And in pure sand, the intensity is the lowest ^[36]. These data indicate that the role of self-preservation of metastable gas hydrates in natural conditions is insufficiently studied and, possibly, exaggerated.

The synthetic hypothesis combining surface and deep factors of formation of gas-saturated zones with abnormally high gas pressure includes the mechanisms discussed above.  It makes it easy to explain certain observed phenomena, but gives little to study the processes that form them. When identifying the causes and sequence of the development of local gas-saturated zones and the subsequent preparation of conditions for an explosion, it is necessary to identify the change of in-ground processes and associated cryogenic structures. These include: primary gas transit channels, gas localization and accumulation zone, areas of deformations of the enclosing frozen rocks and roofs, etc. These studies can only be carried out based on the mechanisms outlined in the previous hypotheses. The conducted research should be based on the study of paragenetic relationships in the gas dynamic geosystem. The mechanical connection of various theoretical concepts will not allow us to identify the sequence of interrelated processes that form gas emission funnels. To do this, it is necessary to develop criteria based on objective quantitative indicators that eliminate subjective quality assessments.

Decomposition of permafrost gas hydrates due to surface heat sources

In the upper part of the section (the first tens of meters), hydrate-containing frozen rocks are in a metastable state, any, even insignificant heating can lead to irreversible dissociation of gas hydrates. Zonal and regional changes in the temperature of frozen rocks are associated with climatic changes. But these changes occur at a slow rate.  The temperature increase in the horizon of gas hydrates to the temperatures at which their dissociation occurs can also be influenced by surface conditions. Almost all gas discharge funnels are located in the so-called "warm landscapes" (rivers, lakes, areas of increased snow accumulation, etc.) ^[8]. The most effective source of local temperature increase of frozen rocks is surface reservoirs. There are several examples of the existence of lake basins in the locations of gas discharge funnels.

The gas discharge funnel discovered in 2020 is located on the flat surface of the 3rd marine terrace within a swampy watered hollow formed on the site of a previously existing surface reservoir (Fig. 14) ^[37]. The presence of the lake is indicated by a zone of secondarily frozen taberal deposits of lenticular shape (Fig. 12). The thickness of this horizon varies from 3 to 10-12 m . The contact with the underlying icy clays is clear and even (Fig. 15).  Taberal deposits are underlain by frozen rocks with a cryogenic structure typical of marine sediments of the III marine terrace (mesh cryotextures) (Fig. 16).

Fig. 14 A gas discharge funnel located within

hollows on the surface of the third sea terrace ^[37].

Fig. 15 The structure of the upper part of the funnel discovered in 2020 ^[37].

Fig.16 A gas discharge funnel discovered in 2020 in Yamal.

1 - the lower part of the funnel; 2 – the zone of deformed cryotextures; 3 – the zone of marine loams with mesh cryotextures; 4 – the zone of sea loams that were pulled out under the lake and again frozen. Screenshot of the video ^[38]

The presence of thawed and then frozen precipitation (taberal) is observed in the locations of some other funnels of gas emission: the Yamal crater (Fig. 17) and in the Deryabinskaya funnel found on the Gydan Peninsula (Fig. 18).

Fig. 17 The upper part of the Yamal crater July 2014. Dotted line

the lower boundary of taberal precipitation is indicated. Photo by M. O. Leibman.

                                     Fig. 18 Deryabinskaya voroka. Taberal precipitation of lenticular shape

they lie over a layer of ground ice. Photo. V.I. Solomatin.

In all three craters considered, taberal sediments are underlain by icy frozen rocks with a primary, cryogenic structure. The examples given point to an obvious source of heat entering the frozen strata - surface reservoirs.

The considered materials, testifying to the presence of taberal deposits in the area of the gas discharge funnels, do not deny other sources of heat. They only illustrate the possibility of developing an intense heat flow due to local warming of frozen rocks under lakes. Decomposition of gas hydrates in the thickness of frozen rocks is also possible due to the local heat flow from below. The source of the increased heat flow can be a large oil and gas-bearing structure, the heat flow from which can increase the temperature of gas hydrates and thereby bring them to a stable thermodynamic state with the release of a large amount of gas. At the same time, the temperatures of frozen rocks may remain negative.

It should be noted that the occurrence of abnormally high pressure in the thickness of perennial rocks due to the decomposition of gas hydrates does not always lead to the release of gas and the formation of a corresponding funnel. Let's take this as an example of the Yamal crater. The gas hydrate horizon of gas hydrates identified according to geophysics lies at a depth of 60-80 m ^[3]. According to the calculations of V.P. Merzlyakov ^[39], at this depth, a pressure of about 10 MPa is required for the discharge of the soil mass (Table. 1), the hydrostatic pressure under these conditions is about 0.7 MPa. When decomposing gas hydrates at negative temperatures, the pressure created does not exceed 2.6 MPa ^[13].

Table 1. Dependence of critical pressure on depth ^[39]

From the above data, it can be seen that at this depth, in the decomposition zone of gas hydrates, pressures are created three times higher than hydrostatic. These pressures do not provide the possibility of ejection, but lead to the development of local deformations of frozen rock. The gas under pressure begins to penetrate through cracks into the overlying thickness of permafrost rocks. The cracks and dislocations that have appeared accelerate gas filtration. A filtration flow is formed in which the gas fluid is filtered from an area with high pressure to an area with lower pressure (usually towards the surface). The totality of these processes, defined by the authors as a filtration - deformation mechanism, allows gas fluids to penetrate deep into the frozen massif, to higher levels, thereby preparing the possibility of explosive processes. When approaching the surface, the gas pressure decreases, but it can reach the values necessary for the rupture of the frozen roof and the release of gas-saturated icy frozen rocks ^[8].

The ratio of natural factors causing the formation of gas-saturated zones with abnormally high gas pressures and gas discharge funnels in the cryolithozone

The materials considered showed that attempts to explain the formation of gas-saturated zones with abnormally high gas pressure in frozen rocks, based on hypothetical representations, are fraught with considerable difficulties. The evidence base of such constructions remains unconvincing, primarily due to the lack of reliable classification features. In order to create a reliable and evidence-based system of these features, it is necessary first of all to consider the entire complex of theoretically possible mechanisms for the formation of local gas-saturated zones in frozen rocks. Figure 19 shows the ratio of various natural factors and mechanisms that cause the occurrence of gas-saturated zones with increased pressure.

Fig. 19 Causes of the occurrence of gas-saturated zones with increased pressure at the initial stage of the development of explosive processes in permafrost rocks.

1. The flow of deep gas through tectonic deformations.

2. Cryogenic gas concentration during epigenetic freezing of thawed sediments (formation of gas pockets) or freezing of taliks during the formation of perennial heave mounds.

3. Gas release during decomposition of gas hydrates contained in permafrost rocks.

1.2. The flow of deep gas into the freezing taliki.

1.3. The warming effect of deep gas on hydrate-containing frozen rocks leading to the decomposition of gas hydrates and gas release.

2.3. The flow of gas formed during the decomposition of gas hydrates into the freezing talik.

1.2.3. The joint flow of deep gas formed during the decomposition of gas hydrates and gas from gas pockets into the freezing talik.

Regardless of the origin of the gas, the processes forming gas-saturated zones with abnormally high pressure, as well as the processes of subsequent preparation of conditions for an explosion with the release of overlapping frozen rock, are combined within the framework of unified developing local gas dynamic geosystems. The explosive process is the final result of the self-development of these geosystems arising in the thickness of frozen rocks. The diversity of their structure is determined by the paragenetic relationships between the processes of phase transitions, gas filtration and deformation of gas-saturated ice-ground material (from viscoplastic movement to brittle fracture). The formation of these gas-dynamic geosystems is caused by the change of complexes of cryogenic processes and their corresponding cryogenic formations. Of course, not all local areas of high pressure are realized in the form of funnels, excess pressure in some of them can be removed due to gas leaks with numerous deformations of the roof.  But where the explosion occurred, the formed funnel, together with the rocks containing it, preserves information about the history of the development of the primary gas-dynamic geosystem. The most important areas of study of these parageneses can be identified: isolation of supply gas channels, horizons of hydrate-containing frozen rocks; areas of taberal sediments fixing the contours of existing taliks; study of the cryogenic structure of the walls of gas discharge funnels, etc. These works will allow us to move from hypothetical constructions of the formation of zones with abnormally high pressures and associated gas discharge funnels to systematic studies of the mechanisms of local gas accumulation in frozen rocks and the creation of conditions for the implementation of explosive processes.

During the preparation of the explosion, the primary structure of the frozen strata is deformed and rebuilt in accordance with the emerging pressures and volumes of incoming gas. It is necessary to distinguish between factors affecting the occurrence of local gas-saturated zones and factors leading to the development of gas-dynamic geosystems associated with them. The first cause the creation of conditions for the violation of thermodynamic equilibrium in the frozen thickness, the second ensure the implementation of the processes of formation and development of the local gas-dynamic geosystem preparing the explosion. Gas in a frozen massif can accumulate due to various mechanisms: filtration along a fault from a depth, decomposition of gas hydrates, cryogenic concentration of free gas or its accumulation in a lithologically conditioned gas pocket (a sandy lens in a clay column), etc. The totality of processes forming gas-saturated zones and gas discharge funnels should be classified as cryogenic, since they are due to the strength and deformation properties and phase transitions of water in freezing and frozen rocks, their structural and textural features and mass transfer processes. Successful study of these processes can be carried out only with the expansion of traditional concepts in geocryology with the inclusion of provisions developed within the framework of geology, glaciology, tectonics, volcanology. The variety of hypotheses of the occurrence of gas-saturated zones in which gas is under high pressure (deep gas intake, comprehensive freezing of taliks, decomposition of gas hydrates under the influence of thawing), and the mechanisms of preparation of explosive processes correspond to the difference in geological, tectonic and landscape conditions. To give any of these factors a dominant role would be a clear simplification. These hypotheses will be implemented in the form of various scenarios for the development of local gas-dynamic geosystems corresponding to specific landscape, geological and tectonic conditions. The current state of knowledge of gas-saturated zones and gas discharge funnels in frozen rocks requires a transition from hypothetical ideas to the identification of scenarios for the development of local gas-dynamic geosystems in specific geocryological conditions.

Conclusion

The analysis of the structure of the funnels of gas emission showed that the totality of cryogenic formations forming them form naturally constructed gas dynamic geosystems. The development of these geosystems, from the formation of gas-saturated zones with abnormally high pressure, to the subsequent preparation of the explosive process, is due to the change of complexes of cryogenic processes and the corresponding parageneses of cryogenic formations.

Analysis of the materials of the study of gas discharge funnels found in the north of Western Siberia allowed us to conclude that both surface conditions and deep tectonic processes can have a significant impact on the formation of gas discharge funnels. Both can contribute to the emergence of non-equilibrium conditions that trigger the mechanisms of formation of gas-saturated zones with abnormally high gas pressure and preparation of explosive processes in frozen rocks.

The main hypotheses of the occurrence of gas cavities in which gas is under high pressure are: the intake of deep gas, comprehensive freezing of taliks, decomposition of gas hydrates under the influence of thawing of frozen rocks.  The mechanisms of preparation of explosive processes correspond to a variety of geological, tectonic, landscape conditions. To give any of them a generalizing character would be a clear simplification. These hypotheses will be implemented in the form of various scenarios for the development of local gas-dynamic geosystems corresponding to specific territories.

The current state of knowledge of gas discharge funnels requires a transition from hypothetical ideas about the cause of the occurrence of local zones of increased gas pressure in frozen rocks to the identification of scenarios for the development of local gas dynamic geosystems in specific geocryological conditions. This requires further improvement of geocryological, geochemical and geophysical research methods and their wider application in the study of these formations.

Insufficient knowledge of the conditions for the development of gas-dynamic geosystems in frozen rocks does not currently allow us to accurately determine the cause and scenario of the formation of gas-saturated zones with abnormally high gas pressure in them. Therefore, there is no reason to give priority to one or another mechanism of their formation. One can only consider various combinations of mechanisms that can be implemented in natural conditions.

The processes forming these formations are cryogenic. They are based on phase transitions, mechanical characteristics, mass transfer processes, structural and textural features characteristic of frozen rocks and therefore should be included in the theoretical and applied apparatus of geocryology. The problems arising in the study of gas-saturated zones and gas discharge funnels can be solved only by expanding the classical concepts of theoretical and applied geocryology, by including provisions developed within the framework of geology, tectonics, volcanology, etc.