TITLE OF THE INVENTION [0001] SEA ICE HABITAT RESTORATION PLATFORM TECHNICAL FIELD [0002] The present invention is directed to a pack ice habitat restoration platform and, more particularly, but not exclusively, to a floating marine platform for extracting oncoming pack ice and depositing the extracted ice on passing pack ice to form through freeze-bonding rafted ice providing an augmented habitat for native ice-obligate and ice-associated species including semi-aquatic marine mammals, terrestrial mammals, seabirds, sea ducks, fishes, micro-, meio- and macrofauna and flora on, within and around the pack ice. The artificially induced rafting from the platform over the course of an Arctic winter produces a meandering disturbance patch and corridor potentially hundreds of kilometers long, and provides a structural framework which sets in motion a succession of direct and indirect habitat restoration processes from fall freeze-up through summer melt. BACKGROUND [0003] Due to climate change, the extent and thickness of Arctic sea ice is decreasing and during summer disappears from marine habitat over shelf waters in the Bering, Chukchi, Beaufort, East Siberian, Laptev, Kara and Barents Seas (Kwok et al., 2009; Stroeve et al. 2014, p.1216). Seasonal (first year or FY) ice has replaced multiyear (MY) ice as the dominant type since the 1990’s (Perovich et al., 2019, Fig.3). In the period between autumn freeze-up and spring break-up, which continues to shorten, the FY ice that does form is declining in thickness due to declining freezing degree days (FDD), and to warming of sea surface temperatures during open water season (Anderson, 1961; Stroeve et al., 2018, Fig. 2c). A related problem with thinner level ice is that it melts sooner than thicker level ice due in part to increased surface-to-volume ratio and is more prone to break-up (Bilello, 1980, pp. 29-34). Thus, the decline in Arctic pack ice habitat extent, durability, and carrying capacity as a platform, shelter and substrate threatens the survival of countless native species that the ice has supported for hundreds of thousands of years (Worsley and Herman, 1980). [0004] It should be noted at the outset that the sea ice restoration platform is not intended as a permanent strategy to augment pack ice, which functions in countless ways from micro to global scales using many processes still not well understood (Dieckmann and Hellmer, 2010, pp.1-22). As one aspect of its limitations, the sea ice restoration platform can offer no direct support to landfast ice, which is a similarly vital and threatened ecosystem. There are also other aspects of climate change, such as ocean acidification, that increasingly threaten to undermine any approach with platforms or even geoengineering such as cloud brightening, and these other aspects must be contended with in parallel to avoid ultimate failure. Given the persistent lack of global resolve to work towards a timely reduction in greenhouse gas emissions, the sea ice restoration platforms are a temporary means to buy time for these species while the systemic changes to reduce global emissions slowly materialize. Assuming the Earth’s carbon cycle is eventually brought back into balance, and the sea ice ecosystems gradually reestablish themselves with the species remaining, possibly requiring centuries (Solomon et al., 2009, Fig.1), the platforms would be removed. [0005] With the foregoing context, in order to maintain some remnants of pack ice habitat for the aforementioned ice-obligate and ice-associated species (Moore and Huntington, 2008, p. S158-S159; Tynan et al., 2010, p.395-396) including walrus, pagophilic phocid seals, polar bears, Arctic fox, Arctic cod (Boreogadus saida), spectacled eider, ivory gulls, thick- billed murres and black guillemots, a habitat restoration platform and method in accordance with some preferred embodiments may produce a pack ice disturbance patch and corridor with increased ice thickness, load capacity, durability, lair-making opportunities, enhanced sea ice derived primary production from ice algae, increased spatial distribution of rafted ice via drifting after ice break-up in spring, and enhanced organic carbon flux to the benthos (Pickett and White, 1985, pp.3-13; Forman and Godron, 1986, pp.83-155; Grebmeier and Barry, 2007, pp.363-365; Ray and McCormick-Ray, 2014, pp.129-132). Ecosystem And Landscape Ecology Context [0006] As it implies, the term ‘patch’, from the field of landscape ecology, indicates an area of habitat and its native community of species set within an overall ecosystem - the ‘matrix’ - that is typically less supportive and possibly hostile or fatal to some members of the community (Pickett and White, 1985, pp.4-5). ‘Corridor’ indicates a curvilinear patch that allows community member dispersal between patches, and a group of patches in the matrix may be linked together with a system of corridors. A corridor might be the result of natural biotic and abiotic processes over hundreds of thousands of years, as in the case of a river and riparian ecosystem formed by a confluence of rainfall, snowmelt, ground water, freeze-thaw, faunal and floral temporal succession, and other biogeochemical processes (Hilty et al., 2006, pp.50, 200-201). Examples of ‘disturbance’ patches and ‘disturbance’ corridors include an isolated area of forest severely burned due to a lightning strike and just starting to be recolonized by pioneer species, a cleared strip through forest for a powerline (Forman and Godron, 1986, p.143), or a wetland emerging over decades upstream of a beaver dam in a forest (Johnston and Naiman, 1987, pp.47-50). It is suggested that the ongoing anthropogenic pulse of greenhouse gas emissions beginning approximately with the Industrial Revolution is an example of a global disturbance with what is projected to be a duration of centuries to thousands of years depending on the effect under consideration (Pickett and White, 1985, pp.6-13; Solomon et al., 2009). It would probably be classified at the ‘catastrophic’ level of disturbance using the definition according to J.L. Harper (1977, p. 627; Pickett and White, 1985 p.8), since it tends to decrease fitness given the brief geologic time span and frequency permitted for adaptation, particularly for long-lived K-selected species like semi-aquatic marine mammals. Thus, disturbances come in a wide variety and do not necessarily correlate with a net loss of diversity, abundance or other long term ecosystem damage, and intermediate disturbance theory suggests that a maximum diversity and abundance tend to occur with some moderate level of disturbance with respect to its frequency, duration, magnitude and severity, so that all community members have opportunity to recover, redistribute, compete, and potentially accrue increased fitness due to the natural selection processes generated by the disturbance (Sousa, pp.120-121; Pickett and White, 378-379). [0007] In the case of the restoration platform, mechanically forcing ice to the sides creates a disturbance in the recurring pattern of the pack ice matrix. The resulting aquatic disturbance patch - a micro polynya flanked on each side by rafted ice - transitions to a disturbance corridor of rafted ice and open water that begins to freeze given sufficient FDD thereafter (Hilty et al., 2006, pp.89, 116). This aquatic disturbance patch in the pack ice matrix opens multiple, persistent pathways for species to move: 1) across the plane of the micro polynya’s surface, and 2) through the surface into the water column, to bottom ice habitat, to remote breathing holes and possibly lairs, to openings in the corridor further downdrift, to cracks and micro polynyas formed by neighboring platforms, and down to the benthos and back. [0008] At the same time multiple edge conditions, ‘ecotones’, are created by the micro polynya, transitional flood zone, rafted slabs and channel edges. These ecotones support ‘edge species’ that are well adapted to negotiate and exploit this sharp transition in habitats (e.g., walrus, pagophilic phocids, diving seabirds and sea ducks) and they will be naturally attracted to them (Forman, 1995, pp.61, 85, 97). The adjective ‘semi-aquatic’ applied to some members of this community expresses this two-fold adaptation and dependence on the water column and platforms, whether of ice, soil, rock or other material. Their distribution along the length of the disturbance corridor will change as the water freezes based in part on details provided further below. In addition to purely thermodynamic freezing, there will be dynamic deformation of the new ice by ever-shifting drift forces, which will alter channel width, form new cracks, new ridges, rubble and induce secondary rafting. [0009] Through the lens of landscape ecology, the matrix of pack ice emerges from, drifts upon, and melts back into a parent matrix of the water column, which overlays the ‘patch mosaic’ of benthic communities (Melnikov, 1997, pp.39-40; Pittman, 2017, pp.3-18, 36-37, 57-66). The water column may be viewed as a highly three-dimensional seawater matrix filled with dynamic, mobile patches such as phytoplankton blooms, floating ice crystals caught in Langmuir circulation, advected or upwelled nutrient rich water, plumes of pollution, and marine snow to name a few (Pittman, 2017, pp.57-66). Benthic patches on the seafloor are often coupled to the naturally recurring edge zones of the pack ice above via flaw leads, polynyas and the marginal ice zone (MIZ), i.e., the wind and current driven gradient of ice floes from about 80% pack ice areal coverage to the open ocean. This coupling is mediated by the dynamic patches and matrix processes of the seawater, and consequently is innately complex and variable on many scales. For brevity, this three-layered patch-matrix model suggested is necessarily a highly simplified treatment to convey only essential ecosystem mechanisms that relate to platform functions claimed. [0010] Benthic patch communities in the Arctic are composed predominately of detritivores, suspension feeders, scavengers and predators since it is generally too deep and hence too dark for photosynthesis, and thus are dependent on marine snow and other forms of carbon flux, or possibly the advection of nutrients (Grebmeier, 2006, p.331). The continental shelf’s benthic matrix supports patches whose location, size, community composition, diversity and health is strongly influenced by depth, current speed, current nutrient load, and the mosaic of the predominantly soft bottom sediment types they are situated in (muddy, sandy, gravelly and combinations thereof) with sandy silt and silty sand being common and generally favorable for diversity and abundance in much of the Chukchi Sea (Toimil, 1978, p.21A, Fig.6; Nelson, 1994, Fig.8 and 9). Generally, the shallower, the more marine snow and carbon flux that reaches the benthic community and the more biologically productive it tends to be (Grebmeier and Barry, 2007, p.380). Though one limit on the benefits of shallow depth is increased potential of ice scour from pressure ridge keels, which can severely plow and locally decimate benthic communities, though climate change is altering this risk (Toimil, 1978, pp.11A, 30A, 36; Conlan et al, 1998). [0011] The Arctic spring break-up and summer melt seasons are regional scale examples of the coupling between the pack ice and the benthos, in which a large portion of the micro-, meio- and macrofauna and flora are transported into the water column via branched brine channels and finally by complete melting. This includes sea ice algae, copepods, amphipods, and nematodes, along with particulate organic matter (POM) and dissolved organic matter (DOM) circulating within and just below the ice. These species are then either consumed by predators (grazed), or eventually die and drift towards the seafloor, typically over the course of days or weeks (Alonso‐González et al., 2010). This sea ice derived organic carbon flux is a substantial component of the water column’s marine snow in the Arctic, and that not grazed eventually reaches the benthos, a diverse community broadly categorized as epifauna and epiflora and infauna and inflora, and in the Chukchi and Beaufort Seas typically include many species of brittle stars, bivalve mollusks, burrowing polychaetes, oligochaete worms, sea cucumbers, benthic crabs, sieve kelp, benthic amphipods, benthic fishes such as sculpin, and the common sunstar. [0012] Community members such as bivalve mollusks and crustaceans that depend on the flux of carbon are key prey for adult walrus, diving seabirds and gray whales. Benthic amphipod species are common prey of seabirds and gray whales, and bearded seals prey on benthic fishes such as sculpin and crab species. Walrus, gray whales, and diving seabirds and sea ducks also contribute some amount of bioturbation of the benthos as they forage (Nelson, 1994, pp.1-24, Fig.4, 5 and 12). Thus, many species well known at the surface are dependent for their existence on the benthos, and their enhanced access and interactions with it comprise one aspect of the biological disturbance set in motion by the platform’s disturbance patch and corridor. [0013] One example of a group of highly productive benthic patches is in the vicinity of Hanna Shoal, which rises to only 20 m from the surface at its center surrounded by an average depth of around 43 m (Grebmeier et al., 2015, pp.93-97). Some semi-aquatic marine mammals such as walrus and bearded seals, as well as diving seabirds and sea ducks, spend a significant portion of their time and energy diving between the pack ice edges and benthic patches (Jay et al., 2006, p.617; Grebmeier et al., 2006, p.356), thus shallower foraging areas if abundant enough in prey can increase energy intake, and the attraction of walrus to benthic hotspots like Hanna Shoal each year becomes understandable (Udevitz et al., 2017, p. 12). [0014] It is suggested that the temporal succession of the disturbance patch and corridor during freeze-up and their transition to the remnant rafted and freeze-bonded ice floes that are dispersed during spring break-up and summer melt may interact to create supportive concentrations of habitat for many of the aforementioned species, in some cases accompanied by increased seasonal snow drift habitat, all of which may locally increase and maintain rates of species interaction, energy conservation, social cohesion, migratory regularity, foraging and predation success, reproduction, and recruitment - locally reconnecting some of the trophic structure increasingly fractured by climate change. The fostering of networks of such lightly managed concentrations that may help keep the aforementioned species distributed over as much of their historic ranges as possible is suggested as preferable to letting the pack ice further degrade into a state which finally triggers their extirpation or effective extinction, which we will be likely unable to reverse, or eventually become resigned to as other crises of climate change around the globe continue to mount. [0015] This also involves the principle of keeping a species as spread out and embedded in as many variants of their native ecosystem as possible in order to maintain abilities of their subpopulations to deal with different predators, parasites, prey availability, seasonal changes and so forth (Den Boer, 1968). If one subpopulation declines there remains some chance others may be able to recolonize. Over the last decade especially, the opposite trend is occurring: ranges are often contracting or shifting poleward to the point they are no longer over shelf waters, and large, haphazard concentrations of some species at increasing frequency are taking refuge on land due to declining sea ice habitat, often with deadly results (Fischbach et al., 2009, p.1), in parallel with reports of declining body condition (Peacock et al., 2010, pp.105, 111). [0016] With multiple platforms properly spaced, species during freeze-up may be permitted to migrate from one disturbance patch and corridor to another in an array of them that make up a stepping stone corridor, which from above may resemble a plowed field of meandering rows widely spaced, crossed periodically by leads and pressure ridges at varying angles (Hilty et al., 2006, pp.90, 147) that may functionally resemble an internal MIZ, which the aforementioned ice-obligate and ice associated species seek out as favored habitat (Ray and McCormick-Ray, 2014, pp.179-186; Keighley et al., 2021, pp.42-43). As spring break- up proceeds, areal ice coverage declines and the rafted ice portions from each disturbance corridor break down further into individual rafted ice floes. These remnant rafted floes then become melting patches in an increasingly open water matrix. [0017] The term ‘pack ice’ or just ‘pack’ used throughout this disclosure indicates all types of sea ice that is not landfast. While some prefer to reserve the term ‘pack ice’ for drift sea ice that is above a certain percent areal coverage, e.g., 70%, that is not its meaning here. The preferred location for a habitat restoration platform is in shelf waters up to about 70 m depth since the approximate, most commonly observed diving limit of the spectacled eider is about 70 m and that of walrus about 80 m (Fay, 1982, p.163; USFWS, 1994, p.6), limits which are generally shallower than that of pagophilic phocid seals. Contrary to ice-obligate and ice associated species, some seasonally migrant cetaceans (e.g., fin whales, gray, minke and killer whales) may be positively affected by increased ranges made possible by ice loss (Moore and Huntington, 2008, p. S158). The pack ice restoration provided by the platform will not provide enough ice to even modestly reduce invasion by seasonally migrant species such as killer whales, which are known to prey on and at times compete with native species. [0018] As to some specifics of the rafting process, the artificial rafting induced by the habitat restoration platform - as distinct from naturally occurring rafting - simultaneously exposes an area of open water of comparable width between the two parallel rafted zones (Shirasawa et al., 2009, pp.1182, 1198; Weeks, 2010, p.336). By redirecting ice drift forces up platform ramps, a strip of pack ice is extracted by shearing and application of a hogging moment which cracks the ice under its own weight, generally in flexure, and deposits it laterally onto the passing pack ice to form through freeze-bonding a consolidated, rafted ice approximately twice the extracted ice thickness, and with nearly the same and possibly greater strength upon full freeze-bonding as comparably thick monolithic ice (Jizu et al., 1991, p.761; Bailey et al., 2010, p.2). Given sufficient oncoming pack ice thickness (some of which may already have been naturally rafted), rafted ice created with this process may support the loads of the aforementioned species, increase pack ice durability and drift range, extend sea-ice derived primary production, enhance the flux of nutrients into the water column and eventually to the benthos (Moore and Huntington, 2008, p. S158; Ray and McCormick-Ray, 2014, p.179). Translucent Aquifer – Comparison with Flood Basalt [0019] Sea ice is quite unlike manufactured ice, or even lake ice. To provide some insight into its structure, complex relationship to the species that have evolved with it, and better understand aspects of the platform’s effects described further below, it is suggested to imagine pack ice as a translucent, cemented soil or weathered, porous rock profile which shares qualities with more familiar terrestrial formations, which have accumulated a much larger and richer body of human experience, culture, and scientific and engineering literature. The analog suggested and described briefly below is flood basalt - a type of lava formation found around the world, notably in Siberia, Washington State, and west central India, as well as on other planets including the Moon and Mars (McMahon et al., 2013, p.1.20, Fig.3A and 3B). This particular flood basalt is assumed to be transformed over time into an unconfined aquifer (i.e., a water-bearing and transmitting porous material such as rock, gravel or sand open to atmospheric pressure), in this case with a groundwater table fluctuating just below the surface to align with the freeboard condition of pack ice in seawater. As far as hypothetically considering pack ice as a type of aquifer or at least in many respects being the functional equivalent, saline aquifers certainly exist around the world, and to demonstrate the suggestion is not too contrived, consider some properties and processes in parallel below. [0020] On a molecular scale, the microstructure of sea ice is composed of water molecules knit together with hydrogen bonds into 3D lattices of hexagonal cells (type 1h ice; Petrich and Eichen, 2010, p.26), while basalt is a mixture of silica and other minerals arranged into more complex lattices with covalent and ionic bonds on the order of ten times or more the strength (Weeks, 2010, p.54). This difference in bond energy is the basis of the relatively low melting point of sea ice, about -1.8°C at a typical 32-35‰ parent seawater salinity compared to the 1000-1200°C of many basalts. Since the upper surface temperature of Arctic pack ice only falls to around -20 to -40°C for extended periods in winter, with the bottom surface of level pack often remaining around -1.8°C, pack ice survival is confined to a seasonally shifting gradient of only 2 to 38°C. (Considering that the IPCC Sixth Assessment Report projects a global average surface temperature rise of 1.5°C with respect to a baseline average for 1850-1900 as soon as the early 2030’s with an amplification factor for polar regions roughly twice the global average, the increase of roughly 3°C narrows its range still further (Melnikov, 1997; pp.31-33; Arias et al., 2021)). [0021] While their bond energies and hence solid phase thermal ranges are quite different, the top-down crystallization of seawater generates a pack ice stratigraphy reminiscent of many flood basalts, that when thick enough and cooled slowly enough, undergo differentiation into a three layered macrostructure. The top layer of frazil ice with its jumbled, porous microstructure of crystalline needles and disks solidifies first, resembling the similarly nonuniform and porous top layer of cooling lava that is later classified as vesicular basalt (Petrich and Eichen, 2010, pp.30, 41; McMahon et al., 2013, Fig.3A and 3B). The frazil ice is subsequently underlain by the second, more orderly transition zone due to shielding from surface wind and a more stable substrate of frazil upon which to accrete. This transition layer resembles the entablature, the second layer of a three-part flood basalt. The transition zone is then underlain by the third layer of vertically grained columnar ice, terminating in a delicate skeletal layer – the often 1-3 cm thick leading edge of columnar crystal growth - bathed in a nutrient rich layer of water called the ice-water interface. This interface, whose thickness is influenced by the current speed, can be on the order of a few meters deep, and transitions into the water column which often ranges from 20 to over 100 m depth in shelf waters (Thomas, 2004, pp.94-96, 115; Petrich and Eicken, 2010, pp.27-28). The overall topography of the skeletal layer is typically undulating on a scale of roughly one to ten meters, and with sufficient current speed and surface amplitude, zones of relatively high flow on upstream faces and calmer, sheltered zones on downstream faces may establish themselves. Sympagic amphipods are known to favor the calmer zones, possibly in part due to the reduced energy expenditure to stay in place, and possibly for the increased availability of prey. (Melnikov, 1997, pp.105-108). On the topside, frazil ice may or may not have snow cover, which strongly affects heat exchange, and its presence reduces ice growth rate during freeze-up according to its thickness. Snow also attenuates PAR light (photosynthetically active radiation: 400-700nm wavelength; Lund-Hansen et al., 2020, pp.71-80), thus its removal my melting or other means can be part of the sequence leading to the onset of ice algae blooms. [0022] Returning to the flood basalt analogy, as gas inclusions and air bubbles come out of solution and become trapped as they rise to the surface in both solidifying seawater and flood basalt, they generate a gradient of pore space that typically varies with depth and rate of solidification. The gas bubbles coming out of solution from the cooling molten basalt are a result of the pressure drop experienced when the pressurized magma is brought to the earth’s surface, whereas the bubbles forming in the ice (not including those which have floated up from the water column) are a result of dissolved gasses in seawater being rejected by the crystallization process. The upper layers of both pack ice and molten flood basalt solidify faster than lower due to the steeper thermal gradient, and thus trap the most gas, so much at times in the case of basalt that the uppermost layer may be classified as scoria, a sponge-like vesicular basalt (Weeks, 2010, p.223). As with the more slowly crystalized columnar or congelation ice below the transition zone, the slower cooling, lower layer of flood basalt may form into a striking columnar pattern, though on a much larger scale. [0023] By the end of winter, the average salinity of FY Arctic sea ice falls to 4-6 ‰ from the parent seawater’s salinity of 32-35‰ due to the brine expulsion of the crystallization processes (Timco and Weeks, 2010, p.109). The ice inclusions filled with the rejected brine of a much higher salinity - up to over 200 ‰ (Mundy and Meiners, 2021, p.272) - gradually interconnect during spring break-up and summer melt seasons, forming 3D treelike networks that provide a system of meandering pathways analogous to terrestrial river systems (Schwarz and Weeks, 1977, p.502), that provide micro habitat and corridors for the aforementioned microorganisms to circulate. Each drains into the ice-water interface creating a hypersaline outflow region, the reverse of a typical terrestrial river that creates a less saline estuary than the receiving sea. (Petrich and Eicken, 2010, p.40, Fig.2.11). Most fully developed brine channels range between 1 to 10 mm in diameter at their terminus in the skeletal layer (Golden 2014, p.8; Lund-Hansen et al., 2020, pp.31-32), and channel density ranges between 50 to 300 channels per m ² making pack ice highly porous vertically and less so laterally (Lønne, 1999, p.210). [0024] Fluid flow rate through a connected pore space as defined by Darcy’s Law reveals that pack ice permeability may be similar to vesicular basalt (McMahon et al., 2013, p.1.19; Petrich and Eicken, 2010, pp.42-43). Vesicular basalt’s porosity is often in the range of 0- 25% (with scoria up to 45-75%; Saar and Manga, 1999, p.112: referencing Fig.2 ‘flow 2’ vesicular basalt properties), which closely aligns with pack ice that has a typical porosity of 2-25% - most of which is brine volume (Mellor, 1986, p.246). Similarly, the permeability of pack ice is on the order of 10 ^-14 to 10 ^-11 m ² with corresponding brine volume at around 2-10% respectively, which is close to the range of permeabilities of some vesicular basalts with 10 ^-14 to 10 ^-9 m ² given a porosity on the order of 5-25% respectively (Saar and Manga, 1999, p.112: referencing ‘flow 2’ vesicular basalt properties in Fig.2; Petrich and Eichen, 2010, pp.42- 43). Sea ice, though porous, is not necessarily permeable, and as the “Rule of Fives” summarizes: permeability starts at or around -5°C with a salinity and brine volume fraction of about 5‰ and 5%, respectively (Weeks, 2010, p.173). This implies that the internal connectivity that the sympagic fauna and flora rely on for brine channel nutrient fluxes and migration may be enhanced in zones of the ice with these or warmer conditions, e.g., some bottom ice. [0025] Regarding strength, the unconfined compressive strength of FY winter pack is in the range of 20-30 MPa (Weeks, 2010, p.257), significantly below 75-250 MPa for some vesicular basalts with porosity from 2-25%, and more in the range of limestone. Yet this is what one would expect based on the difference in their bond strengths – roughly one order of magnitude difference. As with vesicular basalt, pack ice strength inversely correlates with the amount of pore space (Al-Harthi et al., 1999). Consequently, as pack ice continues to melt in summer and the pore space and brine channel volume (the interior habitat) increases at the expense of internal structure, strength may eventually decline to only 10% of peak winter compressive strength, which can pose a challenge for habitat restoration (Weeks, 2010, p.257). [0026] The colonization of the flood basalt aquifer’s pore space from above the groundwater table by pioneer epiliths and endoliths followed by successor species, may take years to decades depending on porosity, permeability, chemical composition, crystalline structure, climate, biotic community make up, and subsequent disturbance, whereas pack ice begins to be colonized as it forms, generally from below the waterline, mostly by the scavenging or trapping of organisms available in the water column and decreasingly by contact with MY ice communities. The eventual loss of colonization from MY ice has been a cause of concern, as there may be species with life histories that cannot adapt to a sympagic- pelagic cycle (Thomas, 2004, pp.113-114; Thomas et al., 2008, p.196; Bluhm et al., 2010, p. 358; McMahon et al., 2013, pp.1.17-1.18). [0027] Stygofauna, the community of invertebrates that colonized and evolved in groundwater and aquifers long ago, and that likely would be found in the flood basalt aquifer of this comparison, consist of many of the taxa found in the pack ice community including amphipods, cyclopoids, oligocheates, turbellarians, nauplii, nematodes and others (Hahn and Matzke, 2005, pp.35-36; Eisendle‐Flöckner and Hilberg, 2015, pp.368-377). Stygofauna graze biofilms (typically excreted by subsurface bacteria), alter pore space, and transport material through the groundwater environment (Hancock et al., 2005, p.98), and their community composition and abundance can help indicate aquifer connectivity, pore space, water residence time, and other characteristics, while changes in populations or body condition can indicate declining or improving water quality (Hancock et al., 2005, p.98; Eisendle‐Flöckner and Hilberg, 2015, p.373). Based on the extreme abundance and global diversity of terrestrial nematodes, it is unsurprising that they often develop annually into one of the most abundant species in pack ice where they may also graze biofilms containing bacteria, and diatoms not found in terrestrial aquifers (Tchesunov and Riemann, 1995, p.35; Thomas, 2004, p.122-125). [0028] Diatoms - and to a small extent bacteria - in sea ice have been shown to excrete sticky, gelatinous, exopolymeric substances (EPS) that inhibit freezing of the brine around them that would otherwise immobilize and potentially kill them, increase their tolerance for osmotic stress from brine, and provide a barrier to viruses among other benefits (Thomas et al., 2010, pp.446-448). These substances simultaneously sculpt the generally disconnected pore space of the pack ice through localized melting via freeze point depression and thereby increase connectivity, creating an enlarged interior (beyond that which the Rule of Fives would allow) that puts diatoms, bacteria and other microbes further out of range of some predators. EPS also increases brine viscosity and may locally constrict pores and channels, both of which slow brine expulsion to the ice-water interface, resulting in higher brine concentrations in the ice which correlates with lower ice strength, which may pose another concern for strategies to maintain or enhance pack ice strength and durability (Weeks, 2010, pp.250-253; Krembs et al., 2011, p.3657). Similar effects of reduced permeability and clogging have previously been documented in terrestrial EPS-secreting bacteria in soil and aquifers (Baveye et al., 1998, p.123; Ross et al., 2001, p.2029). Internal sculpting of ice with EPS is also reminiscent of the erosion of rock surfaces, cracks and pore space by lichens and endolithic bacteria (Chen et al., 2000, p.132-133; Thomas et al., 2008, p.54-58). [0029] Both pack ice and the basalt aquifer eventually contain within them areas of highly altered pore space that support diverse communities, which based on the practical constraints of field sampling organisms which are generally either invisible, or indistinguishable from one another to the unaided eye, are commonly categorized by the smallest mesh size they cannot pass through. The sets of mesh size gradations often vary by field of specialization, but a common one and the one used in this disclosure delineates three mesh size categories: micro- (less than 0.062 mm), meio- (0.062 – 0.5 mm), and macrofauna and flora (> 0.5 mm); Horner, 1989, pp.139-140). None of these categories reveal anything about the organisms in particular, though size may hint at ecological niche and biogeochemical process they may play a part in or be sensitive to. As with rock and soil formations, biotic community compositions in ice vary with depth and temperature, with the macrofauna and flora in pack ice generally restricted to the bottom due to pore space size limitations. [0030] From the perspective of large scale dynamics, pack ice is driven around the Arctic by wind, currents and Coriolis effect (predominantly wind), which set up annually recurring patterns at different scales including the Beaufort Gyre, Transpolar Drift, polynyas (Uspenski’s ‘Arctic ring of life’; Stirling, 1980) which are loosely analogous to tectonic zones of the earth’s lithosphere driven by convection of the molten asthenosphere. These dynamic pack ice regimes may be broadly classified as: [0031] 1) Divergent zones – that create leads and polynyas, and notably the open water ‘ice factories’ created by offshore winds along the northern Siberian coast (Melnikov, 1997, pp.23-24); [0032] 2) Convergent zones – that create pressure ridges and hummocked ice fields, the most massive along the northeast coast of the Canadian Arctic Archipelago and Greenland. (This MY ice will likely be the last to melt as warming continues). Convergence also includes rafting, which is the pack ice analog of subduction, though the lower ice sheet does not melt to the extent that a subducted plate eventually would as it descends (Weeks, 2010; pp.330-374); [0033] 3.) Shear zones – that contribute to stamukhi (grounded rubble sea ice formations at the pack ice and landfast ice interface), and to ridge building within the pack. [0034] This dynamic landscape of ice can also be said to include a kind of hyper seismicity in which the wind, current and Coriolis effect deform the icescape on the order of fifty million times the drift rate of earth’s current tectonic plates. Accordingly, areas of deformation due to ridging, hummocking, rafting or polynyas may establish themselves within only a year, with full maturity of most heavily deformed MY ice patterns requiring up to only a decade or so. [0035] Within this dynamic ecosystem, keystone primary producers like ice algae (predominantly diatoms by spring) harvest what may seem to us meager levels of light shining through the ice – yet with efficiencies so well adapted that some can photosynthesize sugars with less than 10 ^mol of photons/m ² -sec, more than one order of magnitude lower than terrestrial plants and approaching the theoretical limit (Arrigo et al., 2010, p.290; Lund- Hansen et al., 2020, pp.50-55). Ice algae like Melosira arctica stream down here and there from the bottom of FY ice, eventually sinking with the rest of the ice algae, bacteria and virus into the water column, with those not grazed along the way continuing to the benthic communities that exist and often flourish because of them, in what is most often total darkness, though full of other sensory stimuli for those evolved there (Melnikov, 1997, p. 73). The aforementioned seabirds, sea ducks, walrus, pagophilic phocids and gray whales dive into and out of all of this, further knitting together the three matrices of pack ice, water column and benthos. [0036] Zoomed out to this regional scale of hundreds of kilometers, this meandering comparison with a colonized flood basalt aquifer does not lead to the commonly held notion of pack ice as a frozen wasteland, but rather a thermally fragile landscape of unexpected, often hidden fertility, inhabited by countless species many of whom take refuge ‘underground’, in the relatively warm, stable, nutrient rich zone just above the ice-water interface afforded by the floating ‘translucent aquifer’ of pack ice. For a species like ours, understandably invested in a terrestrial, often lower-latitude perspective, this sort of vast changeability, fragility and sensitivity that we did not evolve with can appear unattractive, and thus can be easily dismissed as flawed, useless or obstructive to the familiar environments we instinctively seek. Extending Sea Ice Derived Primary Production and Distribution [0037] All disturbance events exhibit effects at many scales, and based on the preceding micro to regional perspective, it follows that there are communities and processes within pack ice that can be significantly affected by the platform’s method of generating a disturbance corridor. As a prime example, it is estimated that Arctic sea ice algae primary production - from photosynthesis - is only a small part (about 4-6%) of the entire Arctic pelagic production, yet it is estimated that it provides most of the primary production (70-100%) that via intervening trophic levels supports a polar bear’s diet (Brown et al., 2018 ; “ice algae → zooplankton → fish → seals → polar bears”; Lund-Hansen et al., 2020, p.4). A similar though more variable dependence on sea ice primary production has been demonstrated for the diet of ringed seals (Brown et al., 2014), Atlantic walrus (Yurkowski et al., 2020), and Arctic cod (Kohlbach et al., 2017, pp.62, 71-72). [0038] To prolong this sea ice derived primary production, the rafted ice from the habitat restoration platform stores ice algae and other micro-, meio- and macrofauna and flora community members between layers of ice. The bottom ice community of organisms on the extracted pack ice slab and any top ice community of the level ice are transformed into internalized communities of the rafted composite (Arrigo et al., 2010; Arrigo, 2013; van Leeuwe et al., 2018, Fig.4). If freeze-bonding takes place as preferred, the newly internalized communities reside within and just above and below the freeze-bond zone. It is suspected that much of the micro-, meio-, and macrofauna and flora on and within the ice should survive the freeze-bonding process since they have cryoprotectant systems that facilitate their incorporation into the ice, such as the aforementioned EPS, which acts as a local antifreeze to prevent cellular damage from ice crystal growth, improve mobility within ice, and as a viscous sticky fluid to adhere to the ice, among other functions (Thomas, 2004, p.88). [0039] While time required for full freeze-bond formation has not been verified in field experiments, thermodynamic simulations of Arctic conditions combined with ice tank tests suggest a consolidation time on the order of 2 days (3000 min.) for simply rafted ice with slabs each of 50 cm thickness, and just over 4 days (6000 min.) for simply rafted slabs of 1 m each (Bailey et al., 2010, p.9, Fig.7). During spring break-up, rafted ice floes derived from the artificially rafted winter pack with the internalized communities still within them function as distribution devices with extended drift range. By spring break-up, these internalized communities will be transformed with respect to their composition, dissolved organic matter and spatial distribution within the rafted composite, due to differences in mobility, competitive advantage, and other effects that play out over the course of winter and spring(Lund-Hansen et al., 2020, p.44). BACKGROUND ART [0040] Current marine vessels with a cantilevering ramp having a leading edge at the bow which is beveled downward below the waterline have claimed to be effective in icebreaking and the creation of ice-free channels. However, such prior art vessels so configured do not deposit extracted pack ice on passing pack to form rafted ice providing an augmented habitat for the aforementioned ice-associated and ice-obligate species; rather, the level sea ice is broken into fragments and pieces and deposited as rubble ice in windrows on passing level sea ice at the edge of ice-free channels. [0041] For example, U.S. Patent No.4,436,046 (Braley) discloses an icebreaker that uses its hull at the stem to break up an ice field into pieces and fragments of rubble ice. Ramps are rigidly connected to and substantially surround the hull at the stem. The ramps have sloped surfaces to facilitate the movement of the broken ice fragments out of the water and are contoured to deposit the ice fragments in windrows on the surface of the unbroken ice beneath the outboard extending and downwardly angled sponsons on either side of the hull. Cutting edges with serrated teeth, molded to the ramps and along the hull at the bow, facilitate in breaking a portion of the ice mass into ice fragments. Movable sweep assemblies extending laterally outward from hull sweep the deposited ice fragments (rubble) away from the port and starboard sides of the hull. The ice-breaking boat disclosed in Braley does not raft ice into wide, uniform layers creating or restoring a habitat or ecosystem able to bear the weight of semi-aquatic marine mammals, increase pack ice durability and drift range, increase ice carrying capacity of the other aforementioned species, extend sea-ice derived primary production, enhance the flux of nutrients into the water column and eventually to the benthos. [0042] U.S. Patent No.857,766 (Stangebye), another example, discloses an ice-breaking boat used to make open lanes through an iced-over harbor for the entry and exit of ships to and from the harbor. The ice-breaking boat has a submerged bow portion. The immediate continuation of the bow above the water line is formed into a double inclined plane extending into wings beyond the sides of the boat. The forward portion of the double inclined plane extends downwardly from the superstructure of the ice-breaking boat to the submerged bow portion and is also inclined transversely on opposite sides. The ridge of the double inclined plane is raised at its middle and gradually diminishes toward its juncture with the bow portion at one end and its juncture with the superstructure of the boat at the other end. As the ice- breaking boat is forced ahead by its propulsion system, the ice is lifted from the surface of the water, broken and caused to slide onto the surrounding ice by the double inclined plane and wings, thus claiming to form a clear and unobstructed lane of water as the boat progresses. The ramps of Stangebye terminate with high transverse rafting angles (about 60°), so that the bottom edge of the highly rotated extracted ice often places a high linear load at or near the channel edge, which may result in the extracted ice breaking through and sliding into the channel, or under the ice rather than rafting onto the top. The ice-breaking boat disclosed in Stangebye does not raft ice into wide, uniform layers creating or restoring a habitat or ecosystem able to bear the weight of semi-aquatic marine mammals, or provide the other aforementioned ecosystem benefits; rather, the Stangebye icebreaker treats ice as obstructive, hazardous debris to be eliminated in creating ice free lanes bounded by rubble ice. SUMMARY OF THE INVENTION Technical Problem The Artic ice cap has been receding at about 13% in area per decade for the past forty years (Garcia-Soto et al., 2021). Loss of the ice cap was initially estimated by computer modeling to occur by 2100; however, presently, near total loss of the ice cap in the summer is estimated to occur as soon as the 2030’s (Wang and Overland, 2009; 2012). Arctic ecosystem populations, such as walrus, pagophilic phocid seals, polar bears, Arctic cod, seabirds, ice algae and others of the aforementioned species which use sea ice as a platform, shelter or substrate will lose increasingly large portions of their habitat for much of the year. Aside from geoengineering proposals such as cloud brightening, properly designed platforms and methods having the ability to increase sea ice thickness, bearing capacity, durability, surface area, distribution, and the other benefits noted above for the restoration of habitat presently do not exist. Solution to the Problem [0043] The loss of increasingly large portions of the Artic ice cap and the concomitant thinning of pack ice over shelf waters to levels unable to provide habitat for ice-obligate and ice-associated species may be mitigated by the construction and deployment of sea ice habitat restoration platforms in accordance with the present invention by mooring the sea ice habitat restoration platforms in a marine ecosphere having oncoming level pack ice, extracting a pack ice slab from the pack ice using the forward ice ramp of the marine platform, hogging the extracted sea ice slab by applying a bending moment M to the extracted sea ice slab cracking the extracted sea ice slab under its own weight with a vertex of the forward ice ramp portion of the marine platform as the extracted sea ice slab longitudinally traverses the forward ice-ramp portion at a ramp velocity about equal to the drift speed of the oncoming pack ice, forming a micro polynya channel in the marine ecosphere aft of the habitat restoration platform, depositing on passing pack ice having a free edge forming a bounding edge of the channel the extracted sea ice slab as rafted pack ice setback from the free edge to form a disturbance corridor of sea ice habitat having a terraced ice edge facilitating semi- aquatic marine mammal access to and from the water, or traversal across the disturbance corridor’s young ice. [0044] The marine platform may have a design waterline, a beam, a bow, a stern, a longitudinal centerline, a port side, a starboard side, a platform longitudinal axis, A _x , extending from the platform bow to the platform stern, a platform transverse axis, Ay, orthogonal to the platform longitudinal axis, A _x , and a platform vertical axis, A _z , orthogonal to the platform longitudinal axis, Ax, and the platform transverse axis, Ay. A symmetrical cantilevering double inclined-plane ice ramp may be supported by the marine platform. [0045] The double inclined-plane ice ramp may have a linear longitudinal vertex with a positive slope ^. The vertex may apply a hogging moment, M, to the extracted pack ice exceeding the flexural strength σf of the extracted sea ice as the extracted pack ice longitudinally traverses the forward ice-ramp portion at a ramp velocity about equal to the drift speed of the pack ice. The port and starboard ice-ramp portions may have a transverse slope, ^, progressively increasing negatively from the bow toward the stern. [0046] A forward ice-ramp portion may have a forward ice-ramp portion inlet with a forward ice-ramp portion inlet half-width, R, based on Equation 6 which is dependent on ice thickness h, ice flexural strength σf and ice density ρi of the oncoming pack ice. The forward ice-ramp portion may have a forward ice-ramp portion shearing zone aft of and contiguous with the forward ice-ramp portion inlet. In the shearing zone, the extracted ice may be lifted above the level pack ice and separated from the level pack ice due to shearing under its own weight. An aft ice-ramp portion may have port and starboard aft ice-ramp portion outlet zones aft of and contiguous with the forward ice-ramp portion shearing zone and configured to deposit the extracted sea ice slabs on passing pack ice spaced from the free edge of the pack ice. Ramp Surface Geometry [0047] In some embodiments, the symmetrical cantilevering double inclined-plane ice ramp may be a ruled surface bounded by a lower bounding curve and an upper bounding curve. The rulings of the ruled surface may be oriented transverse to the marine platform longitudinal axis, Ax. The lower bounding curve may be formed by the intersection of a horizontally oriented surface and a vertically oriented surface. The upper bounding curve may be formed by an intersection of a frustrum of a right circular cone with a sinusoidal surface extruded from a sinusoidal curve. [0048] A rafting wedge may be supported by the marine platform amidship. The rafting wedge may bisect the aft ice-ramp portion into an aft port ice-ramp portion and an aft starboard ice-ramp portion. The aft port ice-ramp portion may extend from a rafting-wedge prow aftwardly along the port side of the marine platform and be cantilevered outboard from the port side. The aft starboard ice-ramp portion may extend from the rafting-wedge prow aftwardly along the starboard side of the marine platform and be cantilevered outboard from the starboard side. [0049] The rafting wedge may have a port rafting wedge wall extending beyond the port side of the marine platform and a starboard rafting wedge wall extending beyond the starboard side of the marine platform. The port and starboard rafting wedge walls may have a rafting wedge wall width, W, which is based on Equation 5 below relating offset distance D for depositing the rafted ice, and ramp inlet width R. [0050] The bow of the marine platform may have a leading edge beveled downwardly below the design waterline and have a general shape corresponding to a truncated frustrum of an ellipsoid configured to break sea ice having a pressure ridge. [0051] The marine platform may have a mooring assembly amid the rafting wedge structure. The mooring assembly may be configured to receive and retain therein a pier with pier canopy securing the marine platform to the sea floor. The marine platform may weathervane about the mooring assembly and may heave with respect to the mooring pier due to changes in sea level or sea state up to and including a design wave, with any coincident weather and sea ice conditions. [0052] In some embodiments, the marine platform may have an ice rudder with lateral and aft surfaces, each canted to maintain pack ice loads on the rudder in-plane with the level pack ice to minimize rideup and rubbling. Advantageous Effects of Invention [0053] The sea ice habitat restoration platform and method for extracting and depositing pack ice from oncoming pack to form a pack ice disturbance patch and disturbance corridor, in accordance with the present invention, is based on the principles and corresponding advantageous effects set forth below. [0054] With the sea ice habitat restoration platform in accordance with the present invention, rafting extracted sea ice on level pack ice occurs from above, not from below. To maximize the freeze-bonding rate, the interface between the deposited ice pieces and the top of level ice should be just below sea level in order to maximize exposure to air freezing the water. This is the principle at work in the failed bond-making experiments of Marchenko and Chenot (2009, p.5; Bailey et al. (2010, p.13). In one experiment, a 3 × 0.5 × 0.5 m sea ice beam was placed below level ice in the Barents Sea in April 2006. Marchenko and Chenot note that during the experiment the mean air temperature was “-8°C, fluctuating between - 11°C at night to -6°C during the day”. The sea ice beam had still not bonded to the bottom of the level ice after a day. This experiment helps demonstrate a principle that may seem obvious but is easily overlooked: the thermal resistance of the overlying sea ice and seawater retard the extraction of heat by the cold air. This is why thicker level ice grows more slowly than thinner level ice on a strictly thermodynamic basis. Hence using a special hull or other means to raft ice from below rather than from above is not as efficient in freeze-bonding ice for habitat restoration. One example of the rafting-from-below method is the Waas bow for ice breaking (U.S. Patent No.3,984,091; Tatinclaux, 1988, pp.23-25). With the sea ice habitat restoration platform in accordance with the present invention, large, minimally damaged slabs of sea ice are deposited, not rubble ice. This minimal-damage approach favors ramps with (i) gradual slope changes (fair curves), (ii) low rafting angle (maximum transverse angle ^ ^preferably < 25°), (iii) no steps (with the exception of the ramp outlet termination), (iv) no serrated edges as in Braley, (v) ramp surfaces with low coefficients of friction, (vi) fair wedge wall curvature, and (vii) satisfying an extracted ice segment buckling criterion ^ (See Appendix A for derivation). To ensure low sloped ramps remain unobstructed, actuated surfaces may be incorporated into ramps and wedge walls in order to clear any rubble or pile-ups. [0055] With the sea ice habitat restoration platform in accordance with the present invention, a preferred vertical step height H minimizes fall height and thus impact load that can damage the ice deposited and the level ice supporting it. To this end, the distance H - from the lower surface of the outer tips of the ice-ramp outlet zones down to the average level pack ice surface - is preferred to be equal to the average pressure ridge sail height H _s in the area of deployment. In the first preferred embodiment, Hs is for illustrative purposes assumed to be 1.6m, a mean sail height documented for pressure ridges in some parts of the Chukchi Sea (Cammaert and Muggeridge, 1988, p.25). In this case, if the platform is not deballasted, the ramp cantilever portions will collide with and to a variable extent breakthrough or ride over and push down into the water that portion of pressure ridges that exceed the mean. This aspect of design must be tailored to the area of deployment and coordinated with the other design requirements. Ballasting and deballasting to avoid all pressure ridges is possible, but may also prove impractical as they can occur at a frequency of 3-10 per km. Since FY ice is now nearly the only ice remaining in the proposed areas of deployment (Arctic shelf waters), large ridges typically found in MY ice should no longer pose a substantial problem (Cammaert and Muggeridge, 1988, p.25; Shirasawa et al., 2009, Fig.5, Fig.6 (a-c)). [0056] With the sea ice habitat restoration platform in accordance with the present invention, extracted sea ice slabs are generally deposited flat on the level pack ice - though doubly rafted (three layers of ice) and canted-rafted (partially doubly rafted, with one end of the top slab supported on level ice and the other on a rafted slab) slabs do occur - with potential benefits detailed further below. To minimize horizontal gap size between rafted ice and level ice and thereby minimize bonding time and the chance of bonding failure, the extracted ice should be deposited flat. This is in contrast to prior art methods of extracting and depositing that needlessly stress the ice, often with resultant fractures and rubble, e.g., as taught by U.S. Patent No.4,436,046 (Braley) with the resulting rubble ice ‘windrows’. Depositing rubble on level ice in floodwater (flooding that Braley apparently overlooked) could still produce bonding (like cobbles haphazardly set in a bed of fresh mortar), but with reduced beam-like properties. The overall height of consolidated rubble on level ice will be quite uneven and lower in overall height than rafted slabs on average, assuming the same amount of extracted ice in each case. The moment of inertia and section modulus will also be significantly degraded. [0057] With the sea ice habitat restoration platform in accordance with the present invention, average floodwater immersion depths of approximately h/2 to h/7 may be achieved for freeze-bonding at or around a design ice thickness, h. As an example, for a 0.54 m thick continuous 4 m wide strip of rafted ice slabs offset from the channel edge about 10 m to the strip centerline, and assuming about 800 kPa flexural strength (hard winter ice), 5.5 GPa elastic modulus and 910 kg/m ³ ice density, the average immersion depth from the bottom of the rafted slabs is about 0.16 m, or roughly h/3. In contrast, 2.15 m thick level ice – near the upper limit still found in the Chukchi – rafted into a strip about 8 m wide and offset from the canal edge about 10 m to its centerline with the same properties, has an average immersion of about 1.1 m, or roughly h/2. The deeper immersion is due the wider rafted slabs, assuming breakage sizes according to Equation 6 (The thicker or stronger the ice, the wider the rafted slabs become; See Description of the Embodiments section of this disclosure). If the 2.15 m rafted strip width is reduced to 4 m for comparison with the 0.54 m ice, with the same 10 m offset, the immersion depth of the rafted strip falls to around 0.3 m, or roughly h/7. Hence, the greater the thickness and elastic modulus of the level ice, the more widely spread the rafted ice load is over the seawater elastic foundation. Of the basic properties affecting immersion depth of rafted on level ice (ice thickness, strength and elastic modulus), the last two vary with brine volume (brine and gas voids), which in turn depend on ice temperature and salinity. Temperature varies with season and region, and salinity with region, ice age, and amount of brine drainage that has occurred. [0058] Roughness of the top and bottom ice surfaces will also affect contact area of each surface with the floodwater and therefore influence freeze-bond strength. Immersion depth is also affected by the offset distance D as selected by the designer and is related to Equation 4 (See Description of the Embodiments section of this disclosure). Finite element analysis with a tensionless Winkler foundation, followed by scale model testing in an ice tank is suggested as one means of refining estimates of immersion depth. Some surface roughness may be desirable to increase freeze-bond contact area, and as a shear key to resist longitudinal shear in the horizontal plane of the ice slabs as generated by flexural loads imposed on the rafted and freeze-bonded slabs. [0059] The deeper the immersion depth of the rafted slab to level ice interface, the slower the freeze-bond formation since it will be further from the cold air (in winter at times down to -20 to -40 C depending on region) that conducts the heat out and drives freeze-bonding forward, hence creating a decreasing likelihood of bond completion (Marchenko and Chenot, 2009, p.5; Bailey et al., 2010, p.13). Freeze-bond formation rate also needs to be considered in conjunction with the range of pack ice drift velocities (speed and direction) typically observed at the proposed site and along the full length of the projected disturbance corridor, taking into account pack ice convergence and divergence (contraction and expansion of the velocity vector field), as these may induce dynamic deformation – ridge building, hummocking, naturally occurring rafting, lead formation, and other discontinuities, which may affect pack ice thickness and the quality of the platform’s rafted ice - its uniformity, flatness, freeze-bond strength and durability. Areas of convergence may need to be avoided or some mitigation measures undertaken if intense enough since sufficiently high compressive forces in the plane of the pack can close the micro polynya and corridor. Deployment Site Feasibility Analysis [0060] While eventually requiring analysis and design by a team of scientists, engineers, contractors, indigenous experts, and other stakeholders, a conceptual framework for evaluating a potential site for the platform is sketched below. This needs to include what are generally considered five basic functions or effects of any ecosystem corridor as described by Forman (1995, pp.145-153), and further expounded by Hess and Fischer (2001, p.201, Fig. 1) and others including Hilty et al. (2006, pp.89-115, limited to terrestrial applications), Ray and McCormick-Ray (2014, pp.183-192, with a focus on various pack ice densities in the MIZ), and Pittman (2017, limited to iceless littoral zones). The five functions or effects are conceptually adapted here to the suggested three-layered matrix concept of seasonally frozen icescape, water column and benthos. Effects generated vary across species, spatiotemporal scales, and of course not all effects can be beneficial for all aspects of the ecosystem at all times. As with most ecological interventions, sufficient data analysis, judgement, debate and consensus must be used to balance the tradeoffs, many of which will only be sufficiently understood with prototyping and then clarified with years of field experience. [0061] The platform does have the advantage of creating effects that are local, monitorable and controllable in real-time, and that are readily reversible - as compared to some forms of geoengineering, or compared to various forms of doing nothing. Scaled prototyping in a tank with model ice followed by in-situ full-scale prototyping will be needed to sufficiently confirm any proposals. The micro polynya disturbance patch is an integral

part of the disturbance corridor, and for brevity, references to the disturbance corridor generally include the micro polynya unless noted otherwise. The five basic functions or effects to be estimated for a proposed pack ice disturbance corridor are outlined in Table 1 below: Corridor Habitats Estimate extents of corridor habitat for all seasons for each species, projected down to the benthos and into the atmosphere to a height where the corridor microclimate ends. Estimate the temporal succession of habitats and their timing. Estimate projected population gradients and dynamics from micro to regional scales. Cross Corridor Estimate connectivity for species across the polynya and corridor. Connectivity and Estimate permeabilities for biotic and abiotic material and energy Permeabilities exchanges across the polynya and corridor. Estimate their gradients, dynamics, potential negative effects, potential synergies, and tradeoffs. Map out signs that make the corridor detectable and navigable for each species and how they change over the year: i.e., visual, auditory, olfactory, tactile, thermal, osmotic, and other sensory cues. Along Corridor Estimate connectivity for species along the polynya and corridor. Connectivity and Estimate permeabilities for biotic and abiotic material and energy Permeabilities exchanges along the polynya and corridor. Estimate their gradients, dynamics, potential negative effects, potential synergies, and tradeoffs. Corridor Sources Estimate the flux of species, materials and energy types from the corridor into the three matrices of pack ice, water column and benthos. Include net dispersal of species from the corridor into a matrix or into a neighboring ice disturbance corridor. Corridor Sinks Estimate the flux of species, materials and energy types from the three matrices of pack ice, water column and benthos into the corridor. Include net dispersal of species into the corridor, from a 25 neighboring disturbance corridor; Include predation, and other forms of mortality that the corridor may play a role in. Table 1: Five basic functions or effects to be defined for a pack ice disturbance corridor. [0062] To characterize some preliminary responses to the above corridor functions or effects, consider a hypothetical platform site in the Chukchi Sea, midway between Icy Cape, Alaska (a previous mass haulout site for walrus) and Hanna Shoal. The site is in about 43 m of water, about 80 km from shore (roughly an 8-hour swim for walrus in open water at a typical speed of around 10 km/hr; Fay, 1982, p.21) and directly south of Hanna Shoal. The time of year is assumed to be the start of freeze-up, which is now in late November to early December. As noted previously, Hanna Shoal and vicinity is a favored foraging area of many Pacific walrus (Jay et al., 2012, pp.3, 10) since it is a benthic community hotspot, especially regarding bivalves that are key prey for walrus. The rafted sea ice habitat generated along the prospective ice disturbance corridor begins with the micro polynya, which is flanked on each side by newly rafted ice on a depressed, slightly canted pack ice area sloping down towards the channel edges with resultant flooding. Ice slabs riding up the platform ramps during extraction and rafting onto level ice may undergo partial brine drainage if the pore space is sufficiently connected, given the reduction of head pressure, which will result in micro-, meio- and some macro fauna and flora being transferred onto the ramps and into the water column predominantly around the platform forebody. As the rafted slabs rest partially submerged in flood water on the level ice, they may continue to drain, which may transfer more micro-, meio- and macro fauna and flora into the floodwater along with EPS-laden brine, which may invite predation by seabirds, sea ducks and others. [0063] The disturbance corridor’s near field longitudinal axis weathervanes about the platform pier according to the pack ice drift direction, which is related to the local wind rose, with an offset that may be about 25-30° to the right of the downwind direction, what will be referred to as the ‘downdrift’ direction - due to the Coriolis effect (Weeks, 2010, p.435). For this site, the prevailing drift direction is likely to create a disturbance corridor oriented towards the west-northwest, which roughly parallels the southwest perimeter of Hanna Shoal. Depending on the accumulation of FDD and drift speed, the relatively warm open water microclimate of the micro polynya - at times made visible with frost smoke in winter - is succeeded downdrift by new ice (e.g., nilas or frazil, growing up to 10 cm), which given further FDD is succeeded by young ice (10-30 cm), on up through FY ice (Weeks, 2010, pp. 81). This gradient of thermodynamically grown ice will be accompanied by varying amounts of the aforementioned dynamic processes causing mechanical deformation. [0064] Pack ice drifting will occasionally reverse direction but typically for only short periods of up to a week. The corridor’s jagged overall curvilinear plan will be formed by all the impinging drift forces accumulated over the course of the ice season, and may finally span several hundred kilometers in unfolded length (Weeks, 2010, p.437, Fig.16.1). Depending on the platform’s location, some portions of the disturbance corridor may still be over shelf water habitat by the end of summer melt, though some may have drifted over the continental slope and deeper water, which will be of little direct use for many of the semi- aquatic species noted as it is beyond some of their favored foraging depths. Nonetheless, there may be benefit to the overall pack ice ecosystem, possibly to micro-, meio- and macrofauna and flora in and on the ice, and to species not as limited in diving depth or having no need of diving to the benthos. [0065] The ice disturbance corridor habitat has a pronounced vertical dimension approximated by projecting the outline of the platform’s rafted ice and floodwater down to the benthos, and projecting the same outline up to include the atmospheric microclimate, which includes increased ocean-atmosphere exchange of light, heat, vapors and gasses via the micro polynya and its succession of downdrift new ice. This volume approximately encompasses many causal chains of direct and indirect disturbance prior to spring break-up when the rafted ice framework of the corridor begins to disperse into potentially tens of thousands of rafted, freeze-bonded ice floes. Many of the species subpopulatons supported by the disturbance corridor will disperse with the rafted ice remnants. [0066] As an example of indirect disturbance set in motion due to the increased connectivity across and along the disturbance corridor, the resultant increased mobility of semi-aquatic marine mammals, diving seabirds and sea ducks to and from the benthos, enables their bioturbation of benthic sediments as they forage, which resuspends sediments and nutrients that may benefit of the benthic community and in shallower areas may be entrained as nutrients into higher portions of the water column (Nelson, 1994, pp.1-24, Fig. 4, 5 and 12; Ray and McCormick-Ray, 2014, pp.129-130). [0067] Examples of permeability of materials, nutrients and energy transmitted across the corridor include the increased flux of organic carbon (a component of marine snow) from enhanced primary production in the micro polynya due to increased solar insolation, and from the eventual melting and carbon flux of the rafted ice remnant floes months later. The disturbance corridor’s permeability also encompasses the increased brine flux into the water column from new ice formation, which increases vertical mixing due to its higher density and hence nutrient flow. Reduced wind speed and wind chill, and the resultant accumulation of drifted snow is an example of cross-corridor permeabilities that are less than that of the matrix. [0068] Leads, which are often precursors of pressure ridges, are typically oriented to ice drift vectors at oblique or nearly perpendicular angles (Wadhams, 2000, p.152). Hence, pack ice disturbance corridors from platforms sufficiently spaced will tend to cut across pressure ridges to produce quadrilateral zones of level ice bounded on two sides by depressed rows of rafted ice and on the other two by pressure ridges or leads that may yet form them. From below, the result will be a loose, organic grid of submarine berms from the cracked and canted level ice supporting the rafted slabs, intersecting with pressure ridge keels. It is suggested that this organic grid may act as a network analogous to hedgerows above and mostly below the waterline, that allows some of the aforementioned aquatic species such as Arctic cod and amphipods to disperse and colonize along them. Such a grid may also impede the spread of oil below pack ice in the event of a spill (Wadhams, 2000, pp.273-274). [0069] Along with analyzing a site according to the five basic attributes of a corridor, mapping the succession of pack ice disturbance habitats will help identify which portions of the corridor will be colonized by which species over a particular period. A few examples follow: [0070] Adult walrus can break through ice that is up to about 20 cm thick (young ice; Fay, 1982, p.21), but for haulout they need approximately 60 cm (Robards, 2008, p.19). Walrus in the Chukchi Sea may be attracted to haulout on a restoration platform during freeze-up - as they may be to a natural island - and swim and dive in the micro polynya in its wake. For the Artic marginal seas where they range, beluga (Delphinapterus leucas), bowhead whales (Balaena mysticetus) and narwhals (Monodon monoceros) may also benefit from a platform’s micro polynya and portions of a disturbance corridor where the ice is thin enough to break through. [0071] While it is not certain that walrus will choose to overwinter in a micro polynya formed by a platform, it may not be critical for their preservation that they do so. They may still benefit from thicker rafted ice during spring break-up and summer melt if they overwinter elsewhere, for example in their traditional haulouts in the Bering Sea, though the ice there is thinning and receding more rapidly than at higher latitudes. Nonetheless, there are accounts of walrus not migrating ahead of the advancing winter ice front and instead overwintering in recurring leads or polynyas, such as that which forms north of Wrangel Island, and within the pack of northeastern Svalbard (UFWS, 1994, p.8; Wiig et al., 1996, p. 769; Belikov et al., 1997, pp.267-269). Though it is suspected that the avoidance of long migrations may provide some energetic benefit, for example the annual migration of most of the Pacific walrus subpopulations between the Bering and the Chukchi and Beaufort Seas, there will be increased risk of overgrazing by overwintering on and around platforms if statically placed. Thus, as former migration routes contract due to receding ice habitat extents, platforms will probably require periodic rotation, for example over an array of piers, so that some areas may lie fallow to allow benthic recolonization. As an example, bivalves like Mya truncata that walrus may rely on as prey, require some years to mature and can live to over 40 (Fay, 1982, pp.145, 150; Leontowich, 2004, pp.5, 39). As the ice over shelf waters continues to decline, there will be increased risk of overgrazing by walrus and other benthivores in any case. [0072] In a platform-based overwintering scenario for walrus, as freeze-up continues they will be restricted from swimming downdrift of the micro polynya further than young ice that exceeds about 20 cm. For a number of weeks after freeze-up begins, they will also be limited to pack that can reach a thickness of around 30 cm which the platform could then raft over a number of days into a freeze-bonded composite of about 60 cm to begin supporting their haulout and resting loads. Thus with the FDD accumulation rate for a site, one can estimate the time at which the ice will reach a 30 cm thickness, so that via the rafted ice they can move off the platform and onto natural ice habitat. [0073] The distance downdrift from a platform that walrus and other species find sufficiently stable, safe, comfortable and advantageous for various activities is not certain, and there may be a period of acclimation. For example, the noise of rafting ice from the platform should be similar to that created by ridge building or of floes colliding with one another in the MIZ - their preferred home in the pack ice. Research to determine a ‘safe’ standoff distance for icebreakers to avoid disturbing walrus based on some unquantified combination of noise and visual cues has provided a range of recommendations: 300-500 m and 600-800 m (Keighley et al., 2021, p.276). At a drift speed of about 0.1 m/s (Table 2) this translates to a downdrift time of about 2 hours (nearly 9 km/day). Assuming only thermodynamic growth estimated with Anderson’s equation (Anderson, 1961) at a rate of about 25 FDD/day based on January data in the Chukchi Sea (Francis et al., 2016, p.23), two hours of exposure in the winter Chukchi pack ice may produce new ice thickness of about 1 cm in the channel, which is easily broken by walrus and the other aforementioned semiaquatic marine mammals, though possibly not by sea birds or sea ducks. [0074] In some contrast to walrus, ice-obligate bearded seals may be able to break through ice with their head up to about 10 cm thick (Cameron et al., 2010, p.9), but can dive to depths beyond 200 m (Cameron et al., 2010, p.10), and thus will have a habitat range overlapping with walrus, as they often do in the MIZ, but they may congregate closer to the platform in winter due to the 10 cm limit, though with significantly deeper foraging potential. They are benthic feeders like walrus, but epifaunal rather than infaunal in selecting prey (Burns et al., 1985, p.59). Similar estimates can be made for other pagophilic phocids. In general, where the ringed and bearded seals go, the polar bears quickly follow given sufficient ice (Stirling, 1998, pp.61, 178), and where the bears go, Arctic fox may be close by. [0075] To generalize, as an animal’s kinetic energy is converted via impact into strain energy of a deflected ice sheet to the point of cracking, the energy to break through is related to the square of the speed that the animal can generate prior to impact, its mass, impact angle, and ability to withstand impact. Consequently, more massive, faster animals will be more likely to pioneer thicker new ice, and lighter, slower possibly more fragile species may be limited to thinner ice when on their own, or need to be successors to the pioneers. Thus, the further downdrift of the micro polynya, the more the disturbance corridor may be restricted to animals of higher mass, speed, and resistance to impact injury. One exception in heavier ice will often be ringed seals, which maintain breathing holes by abrasion with their fore flippers. Further Disturbance Corridor Metrics [0076] A few metrics suggested for mapping out a proposed ice disturbance corridor to graphically clarify the five basic functions or effects include the items listed below: 1. Bathymetry of the benthos at and downdrift of the proposed platform site. Approximate drift track envelopes can be estimated using published drift buoy research for some areas of the Chukchi and Beaufort Seas (See for example Pritchard and Thomas,1985 pp.243-250). Also see Hibler (1989, pp.50 and 68) for drift buoy tracks and simulated average annual drift vectors. 2. Distance and elapsed time downdrift of platform for formation of new ice (e.g., nilas or frazil), young ice (10-30 cm), and FY ice (>30 cm formed over one winter). 3. Floodwater estimated width and depth, and rate of new ice production (m ³ /day). 4. Distance downdrift from the platform at which full freeze-bonding of rafted ice occurs. 5. Total length and area of the rafted ice portion of the disturbance corridor over the course of a typical year. In many areas in the Chukchi and Beaufort Seas where the aforementioned species forage, pack ice annual drift can range from 100 – 500 km. Thus, a well-placed platform will be capable of producing a corridor of rafted pack ice and new ice habitat of similar length. Details and Considerations of Rafted Ice Deposition [0077] With the sea ice habitat restoration platform in accordance with the present invention, slabs may be deposited consistently though with some beneficial random variation along the channel edge with gap widths between slabs that will vary with ice thickness as well as the particular design geometry of the platform ramps. For a one-meter ice thickness, gaps may be on the order of two meters or more, which allow the largest marine mammals to circulate to and from level ice to the ice edge in the near field and eventually across newly frozen ice in the far field. Gapping of slabs also induces more snow drifting, which can enhance ringed seal lair-making opportunities (Smith and Stirling, 1975, Fig.2, 3 and 4). Uniformly placed rafted slabs distribute their weight evenly for less stress on the level ice. While freeze-bonding of rafted and level ice will only proceed with sufficient FDD, there still remains a structural advantage to two or more layers of unbonded ice (i.e., resisting a load in tandem rather than compositely). While no freeze-bond formation with floodwater poses an extreme case, it is also likely that in winter some amount of regelation (localized freeze- bonding activated by contact pressure) will occur at contact points between slabs since the dead load pressure of the overlying slab will be significant – for example between two double rafted slabs lying flat against one another and not immersed in freezing water. The point at which the remaining FDD for a given year become insufficient for the amount of freeze- bonding required will vary by region and year. [0078] With the sea ice habitat restoration platform in accordance with the present invention, rafted sea ice is setback from the channel edge in order to create a tiered icescape. This creates steps for a more gradual transition into or out of the water, which may be more amenable to hauling out and climbing over. This is another aspect that must be tailored to the particular ice habitat restoration required. A larger setback distance will also more widely distribute rafted ice loads. Ice edge load capacity is approximately 2.5 times less than that setback enough to approximate an infinite floating plate (Kerr, 1975, pp.10-11), and thus poses more risk of rafted ice breaking through the level ice or of sliding back into the channel. BRIEF DESCRIPTION OF THE DRAWINGS [0079] The foregoing Summary of the Invention, as well as the following detailed Description of the Embodiments of the invention, will be better understood when read in conjunction with the appended drawings. For the purpose of illustrating the invention, there is shown in the drawings embodiments which are presently preferred. It should be understood, however, that the invention is not limited to the precise arrangements and instrumentalities shown. [0080] In the drawings: Fig.1 is a port side axonometric view of a preferred embodiment of a sea ice habitat restoration platform viewed from above in accordance with the present invention; Fig.2 is a front elevation view of the platform of Fig.1; Fig.3 is a rear elevation view of the platform of Fig.1; Fig.4 is a plan view of the platform of Fig.1; Fig.5 is a bottom view of the platform of Fig.1; Fig.6 is a side elevation view of the platform of Fig.1; Fig.7 is an axonometric view from above of the platform of Fig.1 in-situ, rafting pack ice and creating a micro polynya and ice disturbance corridor; Fig.8 is an axonometric view of two construction surfaces, whose intersection (heavy dashed) forms the lower bounding curve of the ruled surface of the platform of Fig.1; Fig.9 is a cross-sectional side elevation view of the two construction surfaces of Fig. 8 taken along the lower bounding curve (heavy dashed) of the ruled surface; Fig.10 is a plan view of the two construction surfaces of Fig.8 intersecting to form the lower bounding curve (heavy dashed) of the ruled surface; Fig.11 is an axonometric view of a right circular cone and a sinusoidal surface intersecting to form the upper bounding curve (heavy dashed) of the ruled surface of Fig.1 at the base of the rafting wedge and along the vertex. Also indicated is wedge wall surface 58 (heavy dashed) as formed by a portion of the frustrum; Fig.12 is a side view of the frustrum of the right circular cone and the sinusoidal surface of Fig.11 intersecting to form the upper bounding curve (heavy dashed) of the ruled surface of Fig 1; Below it in side view is the lower bounding curve (heavy dashed); Fig.13 is a plan view of the frustrum of a right circular cone intersecting with the ramp surface to form a portion of the upper boundary curve (66) of the wedge wall surface (heavy dashed) of the ruled surface of Fig.1; Fig.14 is a logic flow diagram of a preferred method for restoring a pack ice habitat for semi-aquatic marine mammals and the other aforementioned species in accordance with the present invention; Fig.15 is a graph of preferred offset distance D (m, solid line) to maximize flood volume and provide adequate immersion for freeze-bonding, with level ice characteristic length l _c (m) on the x-axis and offset distance on the y-axis (m). Also graphed are related offsets for evaluating platform design: a. Offset distance D _min (m, dashed line) in which deposition of rafted ice is approximately up to the channel edge; b. Offset distance Del-vmax (m, hidden line) in which a maximum volume of floodwater would be produced if the level ice remained fully elastic and did not crack under the linear rafted ice load; c. Offset distance D _cr-vmax (m, hidden line) in which the level ice may crack due to the hogging moment from the linear rafted ice load; d. Offset distance D _cr (m, phantom line) in which a crack parallel to the channel forms due to the hogging moment from the linear rafted ice load. Fig.16 is a rear perspective of the platform of Fig.1, rafting ice and creating a disturbance micro polynya and corridor; Fig.17 is a plan view of the platform in Fig.7 and Fig.16, rafting ice and creating a disturbance corridor with corridor extents indicated; Fig.18 is a longitudinal outboard profile of the platform rafting ice in Fig.19; Fig.19 is an enlarged plan view of the platform rafting ice in Fig.17; Fig.20 is a transverse section through the pier 120 of the platform in Fig.19; Fig.21 is a transverse section through jacking mechanism of the platform in Fig.19; Fig.22 is a transverse section through the shearing zones 48 of the bow in Fig.19; Fig.23 is an enlarged axonometric section of rafted pack ice with the freeze-bond formed; (Ice structure adapted from Fig.1 of Schwarz and Weeks (1977); Fig.24 is an enlarged section of skeletal bottom ice in Fig.23; Image adapted from Fig.1.2, Lund-Hansen et al. (2020), p.4; Fig.25 is an enlarged section of skeletal bottom ice after freeze-bonding in Fig.23; (Image adapted from Fig.1.2, Lund-Hansen et al. (2020), p.4; Fig.26 is an enlarged section of a brine channel wall with some typical bottom ice community members in Fig.23; Image adapted from Fig.7.10 of Thomas et al. (2008), p. 194; Fig.27 is a transverse section of the micro polynya channel near field zone in Fig.17, prior to freezing; Fig.28 is an enlarged detail of the section in Fig.27 of the rafted pack ice set back from the channel edge; Fig.29 is a transverse section of the micro polynya channel’s far field zone cut in a manner similar to that of the near field in Fig.27, but further downdrift and with new ice production in the channel; Fig.30 is an enlarged detail of the section in Fig.29 of the rafted pack ice set back from the channel edge with freeze-bonding and new ice production underway; Fig.31 is a transverse section through the frozen channel and rafted, freeze-bonded ice slabs in the further far field with their induced snow drifts, and two subnivean lairs constructed by ringed seals; Fig.32 is an enlarged detail of two subnivean lairs from Fig.31; Fig.33 is a plan view of the leeward snow drift and two subnivean lairs from which the section in Fig.31 is taken; Fig.34 is a transverse section cut in a manner similar to Fig.32 through a canted- rafted 230 slab and a freeze-bonded flat rafted 226 slab formation in the case of no snow for ringed seals, or for other pagophilic phocids that do not construct lairs but that may benefit from the added shelter; Fig.35 is a plan view of platform in Fig.1 during summer melt in which rafted, freeze-bonded remnant floes (248 and 250) derived from the ice disturbance corridor support the semi-aquatic marine mammals and other aforementioned species; Fig.36 is a section through the water column over shelf waters during summer melt in which isolated floes of rafted, freeze-bonded remnants (248 and 250) from the disturbance corridor drift and disperse sea ice micro-, meio- and macrofauna and flora into the water column 204, which support the benthic community upon which many of the aforementioned species depend; Fig.37 is an enlarged section of a rafted, freeze-bonded floe supporting walrus from the section in Fig.35 and Fig.36; Fig.38 is an enlarged section of a rafted, freeze-bonded floe with advanced rotting of the ice and reduced capacity still able to support pagophilic phocid seals, seabirds and sea ducks, from the section in Fig.35 and Fig.36; Fig.39 is an axonometric view of a few typical Arctic shelf benthic community members with a foraging sea duck (Image adapted from Conlan et al.1998, p.13); Fig.40 is a free body diagram of extracted ice segments driven over a ramp with 2D radius of curvature for the derivation of criterion ^ to quantify bucking instability. Fig.41 is a graph of Thawing Degree-Days (TDD)/Day (°C-day/day) vs Ordinal Date for the air temperature above 0°C in the Chukchi Sea region adapted from TDD per month data from Francis et al. (2016, p.23, Fig.1.15). DESCRIPTION OF THE EMBODIMENTS [0081] Reference will now be made in detail to embodiments of the invention, examples of which are illustrated in the accompanying drawings. The terminology used in the description of the invention herein is only for the purpose of describing particular embodiments of the invention and is not intended to be limiting of the invention. [0082] As used in the description of the invention and the appended claims, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. The words "and/or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items. The words "comprises" and/or "comprising," when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof. [0083] The words "right," "left," "lower" and "upper" designate directions in the drawings to which reference is made. The words "inwardly" and "outwardly" refer to directions toward and away from, respectively, the geometric center of the assembly, and designated parts thereof. The terminology includes the words noted above, derivatives thereof and words of similar import. [0084] As used herein, the word "about" when preceding a numerical value is intended to mean that the disclosed and/or claimed numerical value can be any value within a range of minus 10% to plus 10% of the stated value. [0085] The following description is directed towards various embodiments of the sea ice habitat restoration platform in accordance with the present invention. The preferred area of deployment for the below described embodiments is the Beaufort and Chukchi Seas. Accordingly, the sea ice properties commonly found in the Beaufort and Chukchi Seas, and shown in Table 2, provide the basis for determining the preliminary design parameters for the preferred embodiments described below. Table footnotes are numbered in parentheses. FY Pack Ice Property Value Mean thickness, h (m) Chukchi Sea (1) 0.5-2.0 Beaufort Sea (2) 1.5-2.3 Flexural strength, σ _f (kPa) (winter) (3) 400-800 Density, ρi (kg/m ³ ) (4) 910 Seawater density, ρ (kg/m ³ ) (4) 1025 Elastic modulus (GPa) (5) 3.5-6.5 Poisson’s ratio, ^ ^ ^ ^ ^ ^ 0.30 Characteristic length, lc (m) For 0.5 m thickness (7) 9 For 2.0 m thickness (7) 25 Average annual drift speed (m/s) Chukchi Sea (8) 0.1-0.5 Beaufort Sea (9) 0.08 Table 2: Pack ice properties for Beaufort and Chukchi Seas assumed for preliminary design Notes: 1. ISO 19906 (2019), Annex B, Chukchi Sea, FY ice 2. ISO 19906 (2019), Annex B, Beaufort Sea, FY ice 3. Tatinclaux (1984), p.6; Weeks (2010), p.249 Fig.10.18, p.270 Fig.10.33 4. Weeks (2010), p.223 5. Weeks, 2010, p.269, Fig.10.32 6. Tatinclaux (1988), p.7 7. Tatinclaux (1988), p.6 8. ISO 19906 (2019), Annex B, Chukchi Sea, Table B.8-4 9. ISO 19906 (2019), Annex B, Beaufort Sea, Table B.7-4 [0086] However, the disclosure is not intended to be exhaustive or to limit the invention to the precise values of the design parameters. Those skilled in the art will appreciate that changes could be made to the embodiments described below for sea ice habitat restoration platforms to be deployed in areas other than the Beaufort and Chukchi Seas and having different sea ice properties without departing from the broad inventive concept thereof. FIRST PREFERRED EMBODIMENT [0087] Referring to the drawings in detail, where like numerals indicate like elements throughout, there is shown in Figs.1-41 a first preferred embodiment of the sea ice habitat restoration platform, generally designated 10 and hereafter referred to as the “habitat restoration platform 10” in accordance with the present invention. The habitat restoration platform 10 extracts oncoming pack ice 208 having a drift speed v, drift direction 214, a nominal thickness h, a characteristic length l _c , and a flexural strength σ _f , and deposits the extracted sea ice on passing pack ice 210 to form rafted pack ice (226 with freeze-bonding 234; doubly rafted 228 without freeze-bonding; and canted-rafted 230 with partial freeze- bonding) providing an augmented habitat for semi-aquatic marine mammals and the other ice-obligate and ice-associated species aforementioned. [0088] The sea ice habitat restoration platform 10 comprises a marine platform 12 having a design water line (DWL) 200, a beam 14, a bow 18, a stern 20, a longitudinal centerline 22, a port side 24, a starboard side 26, a platform longitudinal axis Ax extending from the platform stern 20 to the bow 18, a platform transverse axis A _y orthogonal to the platform longitudinal axis Ax, and a platform vertical axis Az, centered on the supporting pier 120, orthogonal to the platform longitudinal axis Ax and the platform transverse axis Ay. [0089] A symmetrical cantilevering double-inclined-plane ice ramp 28 is supported by the marine platform 12. The double inclined-plane ice ramp 28 has a port ice-ramp portion 30 and a starboard ice-ramp portion 32. The port ice-ramp portion 30 has a forward port ice- ramp portion 34 and an aft port ice-ramp portion 36. The starboard ice-ramp portion 32 has a forward starboard ice-ramp portion 38 and an aft starboard ice-ramp portion 40. The forward port ice-ramp portion 34 and the forward starboard ice-ramp portion 38 form a forward ice- ramp portion 42. The aft port ice-ramp portion 36 and the aft starboard ice-ramp portion 40 form the aft ice-ramp portion 44. [0090] In some embodiments, the port ice-ramp portion 30 and the starboard ice-ramp portion 32 have a transverse slope δ progressively increasing from about zero degrees at the bow 18 to about negative 15° amidship. [0091] The forward ice-ramp portion 42 has a forward ice-ramp portion inlet 46 with a forward ice-ramp portion inlet half-width, R, based on Equation 6, and has forward ice-ramp portion shearing zones 48 aft of and contiguous with the forward ice-ramp portion inlet 46. The aft ice-ramp portion 44 has port and starboard aft ice-ramp portion outlet zones 50 aft of and contiguous with the forward ice-ramp portion shearing zone 48. [0092] In some embodiments, the bow 18 of the marine platform 12 has a leading edge 134 beveled downwardly at an angle ^ of about 45° at a bow centerline 52 below the design waterline 200 and has a general shape corresponding to a truncated frustrum of an ellipsoid configured to break sea ice having a pressure ridge. A filleted edge 136 with radius of curvature ^ of about 2.5 m between pressure ridge breaking inlet surface 134 and forward ice-ramp portion inlet 46 facilitates weathervaning due to the fillet’s fair curve inducing less resistance to pack ice lateral flow over inlet portion 46 – when rotating from port to starboard and vice versa. The preferred stem angle ^ is about 17° at bow centerline 52. [0093] A rafting wedge 54 is supported by the marine platform 12 and is generally located amidships. The rafting wedge 54 bisects 44 the aft ice-ramp portion into the aft port ice-ramp portion 36 and the aft starboard ice-ramp portion 40. The aft port ice-ramp portion 36 extends from a rafting-wedge prow 56 aftwardly along the port side 24 of the marine platform 12 and is cantilevered outboard from the port side 24 of the marine platform 12. The aft starboard ice-ramp portion 40 extends from the rafting-wedge prow 56 aftwardly along the starboard side 26 of the marine platform 12 and is cantilevered outboard from the starboard side 26 of the marine platform 12. [0094] The rafting wedge 54 has a port rafting wedge wall 58 extending beyond the port side 24 of the marine platform 12 and a starboard rafting wedge wall 60 extending beyond the starboard side 26 of the marine platform 12. The port and starboard rafting wedge walls 58, 60 have a rafting wedge wall width, W, which is based on Equation 5, relating offset distance, D, for depositing the rafted ice, and ramp inlet width, R. In some embodiments, the rafting wedge 54 has a prow half angle ε of about 25°; the rafting wedge prow 56 has a negative vertical slope; and the port and starboard rafting wedge walls 58, 60 have a negative vertical slope and an arclength extending beyond the beam 14 of the marine platform 12 and having a radius of curvature corresponding to a portion of a right circular cone. [0095] The rafting wedge walls 58, 60, in cooperation with the cantilevered port and starboard aft ice-ramp outlet zones 50 deposit the extracted sea ice 216 an offset distance D on average from a free edge 220 of a micro polynya channel 236 formed in the marine ecosphere aft of the marine platform 12 to an approximate centerline of the deposited rafted ice. Preliminary Design Equations for the First Preferred Embodiment [0096] As preliminary guidance for estimating a preferred offset distance D from the channel edge with adequate immersion for freeze-bonding 234 and maximal floodwater 232 for maximal new ice 239 production, it is estimated using upper and lower boundary curves. The upper boundary curve Dcr-vmax is the offset distance from channel edge that produces adequate immersion for freeze-bonding and maximum floodwater 232 volume for a fully cracked level ice condition, the cracking 237 due to the hogging moment from the linear rafted ice load. Floodwater volume created varies with the offset distance from the channel edge 220 at which the rafted ice is deposited, the characteristic length lc of the ice, and whether it is fully cracked parallel to the channel edge or remains uncracked, in this case by the load of the rafted ice strip (226, 228, 230 and any smaller amounts of rubble ice). The upper boundary curve is derived with finite element modeling (FEM) using a fully cracked portion of level ice supporting a strip of rafted ice load, with the crack parallel to the channel at about a distance Dcr from the channel edge as defined in Equation 8. Since this portion of level ice is conservatively assumed completely separated by the crack 237 from the rest of what is idealized as a semi-infinite level ice plate, it has a reduced load bearing area to support the rafted ice, leading to deeper immersion of the rafted ice and greater flood volume than the uncracked (lower boundary) condition. [0097] The upper boundary curve D _cr-vmax of the fully cracked level ice condition may be estimated by the empirical mathematical relationship, D _cr-vmax ≅ ^ l _c ^1/2 , (Eqn.1) and the sea ice characteristic length l _c as determined by the mathematical relationship, l _c = [Eh ³ /12 ^g(1 - ^ ² )] ^1/4 (Eqn.2) where: l _c is characteristic length (m); h is mean ice thickness (m); ^ ^ is seawater density (kg/m ³ ); g is acceleration of gravity (m/s ² ); E is the ice elastic modulus (Pa); and ^ ^ ^ ^is Poisson’s ratio (-). [0098] The characteristic length is the reciprocal of the “characteristic” of the floating ice plate system, which relates the elastic foundation modulus of the sea water, ^g , to the flexural rigidity of the ice plate floating on it, Eh ³ /12(1- ^ ² ) (Hetényi, 1946, pp.1-4; Timoshenko and Woinowsky-Krieger, 1959, pp.4-6). Characteristic length is a concise way to quantify the load bearing contributions of both parts of the level ice system, and includes the contributions of level ice thickness, elastic modulus, and Poisson’s ratio, and eliminates the need to track of the many possible combinations separately. For example, the deflection curves of two floating ice plates with different thicknesses and elastic moduli, will be the same if they have the same characteristic length. This is analogous to the effect of flexural rigidity EI on a prismatic beam’s deflection curve. [0099] The lower boundary curve in contrast to the upper is based on a hypothetically uncracked uniform, semi-infinite level ice plate supporting the same strip of rafted ice (‘hypothetically’ since nearly all FE models of level ice indicated some level of flexural overstress due to rafted ice loading). Similar to the upper boundary, the lower boundary curve is based on ordered pairs of characteristic lengths and offset distances that maximize floodwater volume and provide adequate immersion for freeze-bonding, and may be approximated with the empirical mathematical relationship, Del-vmax ≅ R/2+10 sinh ³ [ ^(60- lc )/120] cos ² [ ^(60- lc )/120], (Eqn.3) where: Del-vmax is the distance (m) from channel edge to centerline of rafted ice to generate the maximum flood volume and adequate immersion depth for freeze-bonding; R is inlet half-width (m) defined in Equation 6 below; and lc is characteristic length (m) defined in Equation 2 above. [00100] For a partially cracked condition assumed to occur in the level ice supporting the strip of rafted ice, the preferred offset D from channel edge to the centerline of deposited rafted ice may be approximated by a mean of the upper and lower boundary offsets such that, D = ^{^} ଶ ( Dcr-vmax + Del-vmax ). (Eqn.4) Given the level of idealization in this preferred offset distance, refined values of D must eventually be confirmed with ice tank testing of scale models, and evaluation of all other design requirements for the area of deployment, including those outlined in the section on the five basic functions of an ice disturbance corridor. [00101] In some embodiments, the rafting wedge wall width W may be determined by a mathematical relationship, W ≅ D + R/2, (Eqn.5) which is derived from the relationship D + R ≅ W + R/2 based on the plan geometry and dimensions indicated on the first preferred embodiment, where the offset distance D is determined by considering the results of Equations 1 through 4 set forth above and where the forward ice-ramp portion inlet half-width R is determined by Equation 6 set forth below. As a minimum offset case, where D ≅ R/2 (hence, W = R using Equation 5), the rafted ice 226, 228, 230 is placed approximately right at the channel edge 220, which is not preferred as it may create a large step or barrier that impedes movement of semi-aquatic marine mammals in or out of the water. Rafted ice deposited at the edge is also more likely to fall back into the channel. Regarding other dimensional relationships of the rafting wedge and the ramps, the preferred maximum cantilever span distance S is about 1.2 times W, and the preferred wedge wall height U is about 1.5 times the mean design ice thickness ^ ^ ^ ^2.33 ^ ^(for 99 ^th percentile ice thicknesses from -∞ to z, where z is the z-score selected by the designer), with ^ ^ and ^ ^ based on pack ice data from the area of deployment. ^ ^ [00102] The double inclined-plane ice ramp 28 has a vertex 62 extending from the forward ice-ramp portion inlet 46 to the rafting-wedge prow 56. In some embodiments, the vertex 62 has a positive linear longitudinal vertex slope α preferably about 21°. The vertex 62 applies a hogging moment M to the extracted sea ice 216 exceeding the flexural strength σf of the extracted sea ice as the extracted sea ice longitudinally traverses 218 the forward ice-ramp portion 42 at a ramp velocity about equal to the drift speed v of the pack ice. [00103] In some embodiments, the preferred forward ice-ramp portion inlet half-width R to break the sea ice 216 under its own weight in flexure on the vertex 62 of the double- inclined plane ice ramp 28 may be determined by the following mathematical relationship, ^ ^{ఙ^ ^/ଶ} ^^ ൌ ^ _{ଷ ఘ^ ^} ^ (Eqn.6) where: R is inlet half-width (m); h is sea ice thickness, as noted below (m); σf is sea ice flexural strength (Pa); ρi is sea ice density (kg/m ³ ); and g is acceleration of gravity (m/s ² ). [00104] To achieve a desired confidence interval of cracking (e.g., a 95 ^th percentile from - ∞ to z) and assuming an approximately normal distribution for preliminary design, the thickness for designing the minimum ramp width may be estimated with the z-score method as follows: h _pctl = ^ + z ^ (Eqn.7) ^ where: h _pctl is percentile ice thickness (m); ^ ^ is mean design ice thickness (m); z is z-score (-) corresponding to the percentile rank; and ^ ^ is standard deviation (m). [00105] A design ice thickness hpctl based on the statistical distribution of level ice thickness for the region of deployment is preferred in order to achieve a selected confidence of complete cracking. For example, assuming the sea ice flexural strength is 800 kPa ice with mean thickness ^ = 0.45 m and standard deviation ^ of 0.17 m, the anticipated 95th percentile thickness is: h ₉₅ = ^ + 1.65 ^ = 0.45 + 1.65(0.17) = 0.73 m [00106] Accordingly, a minimum ramp inlet half-width R of about 4.7 m is required to crack sea ice that may range up to 0.73 m thickness with the aforementioned characteristics based on Equations 6 and 7. If no standard deviation is used, on average 50% of the ice extracted will not crack on the vertex 62 under its own weight, again assuming a normal distribution. [00107] As noted above, the rafted ice linear load may induce a crack 237 via hogging moment in the level ice parallel to the channel edge, which may be useful for estimating flood volume, immersion depth, extents of the disturbance corridor ecotone, and other effects on corridor habitat properties. Assuming an idealized uniform thickness and strength of level ice prior to cracking, the offset distance Dcr from channel edge to the crack 237 centerline may be estimated by the following empirical mathematical relationship: Dcr ≅ ^ lc ^3/4 + Del-vmax (Eqn.8) where: Dcr is distance (m) from channel edge to crack centerline; l _c is characteristic length (m) defined in Equation 2; and Del-vmax is lower boundary curve (m) defined in Equation 3. [00108] To check for buckling instability of the extracted ice due to eccentric end-to-end compression between ice segments 216 driven along a ramp, and as generalized with ice segments 267, 268, 269 and 270 in Fig.40, which are driven by a drift force F over a 2D convex ramp 139, a buckling criterion ^ as derived in Appendix A is evaluated. As background, a related class of equations have in the past been proposed to evaluate ice riding up discontinuous ramps - not fair curves - though with the intent of inducing instability and pile-up in order to protect built property from encroaching ice (Cammaert and Muggeridge, 1988, pp.215-219). The ^ ^criterion on the other hand estimates the smallest radius of curvature of a fair-curve ramp that may put the ice at risk of buckling in order to prevent a rubble pile-up on a ramp for a given level ice characteristic length lc. Accordingly, the local radius of curvature and ice characteristic length should be constrained to satisfy: ^ ^ ^ = 4 (-2 ^sn ^ ± [4( ^sn ^) ² +1] ^1/2 ) > lc (Eqn.9) b r     where: ^ is the ice sheet buckling criterion (-); r is radius of curvature (m) aligned with drift force F; ^ _s is coefficient of static friction (-); n (-) is number of ice slab segments upstream of the potential buckling zone resisting sliding; ^ is safety factor (-) against out-of-plane buckling; b (-) is the minimum of A/l _c or ^/4; A is length (m) along ramp vertex from point aligned with strip shearing away from level ice up to prow; and lc is characteristic length (m) as determined by Equation 2.   Sample Calculations Using the First Preferred Embodiment [00109] As an example of Equations 1 through 9 applied to the first preferred embodiment depicted in the figures, consider a hypothetical deployment in Arctic pack ice with the following properties: ^ ^ ^= 1.00 m mean ice thickness; ^ ^ = 0.36 m standard deviation; z = 1.65 for 95 ^th and 2.33 for 99 ^th percentile ice thicknesses for z: -∞ to z; ^ ^f = 550 kPa (Weeks, 2010, p.270, Fig.10.33); ^ ^i = 910 kg/m ³ (sea ice, Table 2); ^ ^ = 1025 kg/m ³ (seawater, Table 2); E = 5.5 GPa (Weeks, 2010, p.269, Fig.10.32); and ^ ^ = 0.30, with mean level ice thickness ^ being the maximum mean anticipated for the region of deployment just prior to the onset of melt season. [00110] Based on Equation 7, the 95 ^th and 99 ^th percentile ice thicknesses are: h ₉₅ = ^ + 1.65 ^ = 1.00 + 1.65(0.36) = 1.59 m; and h99 = ^ + 2.33 ^ = 1.00 + 2.33(0.36) = 1.84 m, where values for ^ and ^ ^were chosen for the convenience of a simple example from Shirasawa et al., 2009, Table 2, ice station 040521, sampled May 21, 2004 (roughly the start of melt season based on Francis et al., 2016, p.23, Fig.1.15, and thus approximately the maximum FY ice thickness), in an area south of Hanna Shoal, Chukchi Sea. [00111] The ramp inlet half-width R required to crack and extract on average 95% of the oncoming level ice is then: R = (h ₉₅ ^ _f /3 ^ _i g) ^1/2 = [(1.59)(550,000) / 3(910)(9.81)] ^1/2 = 5.7 m [00112] Based on Equation 1, the offset distance D _cr-vmax to achieve maximum floodwater volume and adequate immersion, assuming the level ice supporting the rafted ice is separated from the remaining semi-infinite ice plate by a crack at a distance D _cr is: Dcr-vmax = ^ ^lc ^1/2 = ^ ^15.0 ^1/2 = 12.2 m where: l _c = [Eh ³ /12 ^g (1- ^ ^ ² )] ^1/4 l _c = [5.5x10 ⁹ (1.0 ³ )/12(1025)(9.81)(1-0.30 ² )] ^1/4 = 15.0 m (Note that ^ is used to calculate lc.) [00113] Likewise using Equation 3, the offset distance to maximize floodwater volume assuming the level ice behaves as an uncracked semi-infinite plate on elastic foundation is: D _el-vmax ≅ R/2 + 10 sinh ³ [ ^(60 - l _c )/120] cos ² [ ^(60 - l _c )/120] Del-vmax ≅ 5.7/2+10 sinh ³ [ ^(60 - 15.0 )/120] cos ² [ ^(60 - 15.0 )/120] = 7.5 m [00114] Per Equation 8, the offset distance to a crack that may occur in the level ice parallel to the channel due the hogging moment from the strip of rafted ice load is roughly: Dcr ≅ ^ lc ^3/4 + Del-vmax ≅ ^ (15.0 m) ^3/4 + 7.5 = 31.4 m The preferred offset for maximal flooding and adequate immersion for preliminary design is then approximated with the mean of Equation 4: D = (Dcr-vmax + Del-vmax)/2 = (12.2 + 7.5)/2 = 9.9 m [00115] The preferred wedge width per Equation 5 is then: W ≅ D + R/2 = 9.9 + 5.7/2 = 12.8 m [00116] The preferred minimum wedge wall height is: U = 1.5( ^ + 2.33 ^) = 1.5(1.84) = 2.8 m [00117] Regarding ice sheet buckling potential, the smallest radius of curvature lies on the upper boundary 66 - estimated at 13.9 m and aligned with sine wave crest 103. The second smallest radius of curvature – 30 m – lies on the lower boundary curve 82 and is aligned with transition 78. Given the ramp surfaces are situated between these, possible paths 218 taken by the extracted ice pass over radii of curvature between 13.9 - 30 m. Thus, the quickest, and most conservative approach is to check the smallest radius against the ^ criterion: ^ ^ ^ = 4 (-2 ^ _s n ^ ± [4( ^ _s n ^) ² +1] ^1/2 ) > l _c (Eqn.9) b r  [00118] As noted in Appendix A, the maximum length of ice segments riding up the ramps is limited by two cases: [00119] Case 1) Failure of the extracted ice strip 216 with respect to weak axis bending (in combination with torsion, shear, compression, and other loads that for simplicity will be conservatively assumed as negligible). This weak axis failure length is assumed equal to that of a floating cantilever ice beam with an upward tip load provided by contact with the ramp (42), so that L = ^l _c /4 or 11.8 m in this example (Hetényi, 1946, p.24; Daley, 2020, p.86). [00120] Case 2) Failure of the strip with respect to strong axis bending due to lateral prying by prow 56, which limits L to about a length of A - the inclined length of about 8 m along vertex 62 starting from a point on the vertex aligned with shear points 48, up to the base of the prow. If a strip of extracted ice is not broken in Case 1, the first it will be exposed to as it is lifted from the water, the strip should be broken in flexure against its strong axis as the prow pries it laterally, rotating it about its base near the notch formed at shear point 48. [00121] Since the shortest length of the two cases governs due to the constriction due to ramp geometry, variable b in Equation 9 may be determined such that: b = MIN(A/lc , ^/4) = MIN(8/15, ^/4) = 8/15 Substituting and solving for buckling criterion ^: ^ ^ ^ = 4 (-2 ^ _s n ^ ± [4( ^ _s n ^) ² +1] ^1/2 ) b where: ^s = 0.7 for ice on steel; n = 2 ice segments upstream on ramp; and ^ ^ = 1 (for simplicity). ^ ^ ^ ^ = 4 (-2 ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ± [4((0.7)(2)(1)) ² +1] ^1/2 ) (8/15)     ^ ^ ^ = 7.5 (-2 ^ ^ ± 2.97) = 1.30 (positive root) Checking the smallest ramp radius of curvature of the platform against ^: l _c      = 15.0 m = 1.08 < ^ ^= 1.30 OK r 13.9 m [00122] In summary, the first preferred embodiment in comparison to the above derived performance criteria indicates that: R = 7.2 m provided meets and exceeds R95 = 5.9 m minimum; W = 13.5 m provided reasonably satisfies a preferred W = 12.8 m; U = 2.9 m provided satisfies a preferred U = 2.8 m; D ≅ W–R/2 =9.9 m provided satisfies D = 9.9 m; and lc/r = 1.08 provided satisfies the buckling criterion ^ = 1.3. Therefore, the first preferred embodiment meets this abridged list of preliminary design requirements. Further Platform Geometry Considerations [00123] In some embodiments, the symmetrical cantilevering double inclined-plane ice ramp 28 is a ruled surface bounded by a lower bounding curve (or directrix) 64 and an upper bounding curve (or directrix) 66 and rulings of the ruled surface are oriented about 90° transverse to the platform longitudinal axis, A _x . [00124] The lower bounding curve 64 is formed by the intersection of a horizontally oriented surface 68 and a vertically oriented surface 70. The horizontally oriented surface 68 is extruded from a first curve 72 on the platform longitudinal axis, A _x , having a forward portion 74 inclined at a slope 76 of about 10° over a length of about 14.4 m transitioning via an arc 78 having a radius of curvature of about 30 m and a cord of about 6.3 m to an aft portion 80 inclined at a slope of about negative 2° and having a length of about 6.8 m. [00125] The vertically oriented surface 70 is extruded from a second curve 82 in a horizontal plan. A forward portion 84 of the second curve 82 is a portion of an ellipse with a major axis of about 10.83 m and a minor axis of about 5.50 m, with major axis parallel to portion 86, and which transitions tangentially at the co-vertex to a mid-portion line segment 86 having a slope of about 30° over a length of about 15.72 m. The mid portion 86 transitions via a portion of an ellipse 88 having a major axis of about 6.21 m and a minor axis of about 4.24 m, with major axis parallel to portion 86 to an aft portion 90, which is a line segment having a slope of about 82° from the major axis of the portion of an ellipse and having a length of about 3.67 m. [00126] The upper bounding curve 66 is formed by an intersection of a frustrum 92 of a right circular cone with a sinusoidal surface 94. A top 96 of the frustrum 92 has a radius of about 25 m with a draft angle ^ ^ 98 of about 21° and is centered about 25 m off the platform longitudinal axis, Ax, and offset aft from the base of the vertex 62 by about 5.86 m. The sinusoidal surface 94 is extruded along the platform traverse axis A _y from a curve 100 having an aft portion 102 defined by a sine wave z = -7cos(2πx/62) joined at the base of the wedge prow 101 to a forward portion 104 defined by a line coincident with the vertex 62 and having a slope ^ of about 21° tangentially joining the sine wave. [00127] In some embodiments, the included angle ^ between wedge walls (58 and 60) and the adjacent ramp surface (36 and 40) at every point along the wall length is less than about 85° and greater than about 75°. (Note that ^ may be controlled by altering the draft angle ^ ^of cone 92 and the transverse ramp angle ^.) In this condition, the ice 216 riding up the ramps (36 and 40) receives a downward clamping force component from the normal force exerted by the walls, which keeps the ice from lifting off the ramp and potentially riding up the wall, which can damage the ice including rubbling, and lead to a more disordered deposition of the extracted ice (which in some embodiments may be desired as described further below). As a minor point of geometry, pairs of ^ and ^ angles at each point along the wedge wall may for geometric simplicity be oriented to lie in planes parallel to the plane of the Ay-Az axes, and thus parallel to the lines forming the ruled surface; however, the downward clamping force component from the walls is related to the included angle normal to the wall surface, which is aligned along radius of the frustrum 92 and thus are not coplanar with Ay-Az. [00128] In some embodiments, the marine platform 12 has an aft haulout deck 107 atop the platform afterbody 108 and adjacent to an ice rudder 109. The ice rudder 109 has lateral and aft surfaces (110, 112) each of which are canted to maintain ice center of pressure in- plane with the level pack ice, at about 0.4h below the design waterline (DWL) 200, given that the centroid of the level pack ice is about h/2 from the top ice surface and that the bottom is about 0.9h below DWL. The aft haulout deck 107 with sides sloping down into the water provides a temporary haulout platform for semi-aquatic marine mammals including walrus 308, pagophilic phocid seals 300, and polar bears 310 when sea ice is of insufficient strength or distribution. In open water season the cantilever ramp system 28 may also serve as a haulout platform. [00129] A mooring assembly 118 is situated within and rises out of the rafting wedge 54, with the portion rising out of the wedge designated the sail 124. The mooring assembly 118 is configured to receive and retain a pier 120 securing the marine platform 12 to the seafloor 351 and allowing the marine platform 12 to weathervane 114 and heave 116 to accommodate changes in sea level, sea state, and any coincident weather and ice conditions. Preferably, the mooring assembly 118 has a rack-and-pinion jacking mechanism 126 disposed therein to lower the platform 12 onto the pier 120 at the deployment site, the pier 120 having been craned onto the sea floor with standard industry methods. The rack-and-pinion lifting mechanism 126 is substantially the same as the mechanisms found in jack-up rigs for raising or lowering the legs of a jack-up rig platform with the exception that the pinions 128 may be retracted from engaging the racks 127 when the platform is afloat, thus enabling the platform 12 to heave 116 and weathervane 114. The two racks 127 are located on opposite sides of the pier 120. The two groups of retractable pinions are located on the forward and aft portions of the mooring assembly 118 attached to the platform 12. [00130] In situations where the platform is desired to be jacked out of the water – e.g., for maintenance, relocation to another site, to wait until ice is sufficiently thick to raft, or to pause rafting – it may be aligned with the use of a tunnel thruster 130 or similar propulsion while adjusting the ballast, so that the pinions 128 may be reengaged into the racks 127. In some embodiments, the marine platform 12 may be provided with an ice shield 132 protecting the submerged portions of the pier 120 from ice impact. In open water season, the ice rudder 109 functions as a typical vane subject to the forces of currents. [00131] The orientation of the platform 12 such that the bow 18 always turns straight into the oncoming pack ice 208 is accomplished with the aid of weathervaning 114, i.e., yawing about the Az axis 360° clockwise or anticlockwise. In one embodiment, weathervaning may be achieved passively with an ice rudder 109 attached to the afterbody 108, which under ice drift forces normal to the Ax axis generate a moment about the Az axis that is always greater than the opposing moment generated by the forebody 29 assuming uniform level ice conditions. Passive weathervaning is a classic mechanism for orienting and is commonly used today in FPSO (floating production storage offloading) vessels, which typically have a turret mooring to weathervane about their station. As an alternative to passive weathervaning, a tunnel thruster 130 or similar propulsion situated in the afterbody108 may be used exclusively or in concert with the ice rudder. However, to keep ambient noise from the platform in the range of natural pack ice ambient noise, thrusters should be used extremely sparingly since vessel noise is a known disturbance, for example to Arctic cod, a keystone species (Ivanova et al., 2020, p.1). Ridge building events, as one example for context, can be quite noisy, and ridges in FY ice can comprise 6% of pack ice area (Dyer, 1988, p.517). [00132] The platform 12 has a bow leading edge which is beveled down at an angle ^ approximately 45° to a depth of about 4 m below the design waterline 200 for the purpose of breaking through ice pressure ridges that a platform operating in the Arctic will regularly encounter. The geometry is adapted from the truncated conical base of a typical Confederation Bridge pier, which has since the late 1990’s successfully broken FY sea ice ridges composed of 0.5 – 2 m thick rubble ice – approximately the same design ice thickness as that of the platform (Brown, 2006, Conclusions and Fig.5). Production of Habitat for Snow Lair and Ice Lair Making [00133] The preferred embodiment of the platform 12 produces some amount of doubly rafted ice slabs (228 on top of 226 on top of level ice 210) as well as canted-rafted slabs (230) partly supported on a flat rafted slab 226 and partly on level ice 210. These and any other rafted configurations, including singly rafted slabs 226, induce snow drifting given sufficient ice thickness to obstruct the wind, snowfall and wind speed. The resultant drifts enhance snow lair making opportunities 304 for ringed seals 303 and may provide shelter for the other pagophilic phocids 300 that do not make lairs. [00134] In some embodiments, the aforementioned doubly rafted and canted-rafted ice configurations may be produced more frequently by setting the wedge wall angle ^ to greater than 90°. In this configuration, the ice 216 riding up the ramp 28 does not receive a downward clamping force from the wedge walls (58 and 60) and will tend to ride up the walls and fall back onto itself (or onto the top of wedge 54 if the walls are not high enough). As noted, an acute wall angle favors better control by applying a small downward force component that keeps the ice from lifting off the ramps as they are force laterally by the wedge walls, resulting in a higher level of singly rafted ice 226 with a greater chance of freeze-bonding 234 while lying flat in floodwater 232 on the level ice 210. Conversely, a wall angle ^ that is too acute may lead to binding of the ice 216 and may result in jamming and excessive damage to the ice. The range of void sizes provided by the various rafted ice configurations may provide some immediate shelter for seals 300 and potentially other species including seabirds and sea ducks (e.g., 314, 316), polar bears 310 and Arctic fox. [00135] In general, the wedge wall angle ^ must be adjusted to the habitat restoration needs of the focal species selected and the ice conditions available. As an example, singly rafted and freeze-bonded ice offers a higher habitat restoration potential for large marine mammals as they attempt to haulout and rest. Walrus (308 and 309) require some of the thickest ice, on the order of at least 60 cm as noted, and they obviously require no lairs. In some embodiments, the wedge walls may be actuated so that the angle ^ may be varied according to rafted ice configuration required, level ice conditions, season or other requirement. [00136] As warming temperatures, less snowfall and more frequent rains further hamper the production and maintenance of subnivean lairs by ringed seals 303 (Hezel et al., 2012; Stirling and Smith, 2004, pp.59 and 65; Ferguson et al., 2005, p.1), ice lair habitat produced by the platform 12 may provide some alternative to the increasing levels of exposure of adults and pups 303 to wind 261 and rain 266. While arguably not as thermally insulating as an intact subnivean lair 304, an ice lair’s (306) ‘canopy of solid ice’ (230) may provide significant protection from predators such as polar bears 310 that readily break through subnivean lairs from above (McLaren, 1958, pp.56-57). Fay (1974, p.394) also notes that the young of the spotted seal and harp seal (300) are known to take refuge among ice blocks of naturally occurring pressure ridges. Thus, there is natural attraction for pagophilic phocids to take refuge around or within dynamically deformed sea ice. However, unlike fully enclosed voids that may at times be provided by pressure ridges (which phocids may then repurpose into snow or ice lairs), ice lairs constructed with the platform are often open on one or two sides. Thus, seal pups 300 within may not be fully protected from predation by similarly small Arctic fox, a common predator. Increased Durability with Extended Range [00137] In addition to improvement in structural capacity afforded by the rafted ice of the platform 12, a more durable ice habitat is created due to the thicker ice composite’s capacity to absorb more heat, and to the reduced surface-to-volume ratio of the rafted and freeze- bonded assembly (226 on 210 during freeze up, and remnants 248 and 250 during break-up and melt season). While not incorporating variables of surface area or volume, a simple empirical formula for comparison of sea ice durability as derived by Bilello (1980, p.30) with original variable names adapted to this disclosure is as follows: hf = hi – a (TDD) (Eqn.10) where: h _f is final sea ice thickness (m) after melting due to accumulated thawing degree days (TDD); h _i is initial ice thickness (m) prior to melting; TDD is accumulated thawing degree days (°C-days) based on air temperature above 0 °C; and a is an empirically derived melt rate constant (m/°C-days). [00138] Though derived from floating landfast ice rather than drifting pack (but including observations from stations in northwest Alaska on the Chukchi coast), Equation 10 provides a first approximation and empirical check relating ice thickness to thawing degree days and via a graph of TDD/day vs ordinal (or Julian) days, can in turn be related to days of duration. As a caveat, Bilello notes that Equation 10 does not include effects of wind, current, waves, solar insolation, or advected warm water, which may at times be significant, but for a comparison of rafted versus unrafted ice floes which must be based on similar climatic exposure to have meaning, they are assumed to have no differentiating effect. Equation 10 is also being applied here to individual rafted and unrafted ice floes that may be more exposed to waves, heat, and other stresses than they would in the fully developed level ice fields that Bilello started with. However, these fully developed ice fields would necessarily pass through a state of break up, dispersion and complete melting, so the melting regimes of the individual rafted vs unrafted floes does reside within the scope of Bilello’s data on which Equation 10 is based. [00139] Estimated TDD for the entire melt season in the Beaufort Sea is assumed to be about 440 °C-days, which is quite close to the 460 °C-days in the Chukchi based on data collected from 1991-2013 (Francis et al., 2016, p.11 (Fig.1.4) and p.23 (Fig.1.15)). For an approximate melt rate, Bilello suggests (in 1980) a value of a = 0.93 cm/TDD. Since the Beaufort and Chukchi have similar total melt season TDD values, this comparison will use the more stringent Chukchi value. [00140] Solving Equation 10 for the number of TDD that can be absorbed by singly rafted pack ice (two layers) of maximum thickness available in the Chukchi Sea and permitting it to melt down to about 0.60 m - the aforementioned limit estimated to support adult walrus - the initial thickness h _i required is about: hi = hf + a (TDD) = 0.60 m + (0.0093 m/TDD)(460 TDD) = 4.9 m [00141] This would require rafting and freeze-bonding of FY level pack that is approximately 2.4 m thick (4.9/2), which is about 0.4 m over the maximum still available in the Chukchi (or Beaufort; ISO19906, 2019, p.424: Table B.7-4; p.431: Table B.8-4). Alternatively, double rafting would be required, i.e., 3 layers of pack ice, each about 1.7 m thick, which may be available in some areas. Using the thickest ice available, near the end of winter, the estimated durability of singly rafted 2.0 m level pack in the Chukchi is approximately: TDD = (hi - hf)/a = [2(2.0 m) – 0.60]/0.0093 = 365 °C-days [00142] For comparison, the estimated durability of unrafted ice is: hf = hi – a (TDD) TDD = (hi - hf)/a = [2.0 m – 0.60]/0.0093 = 151 °C-days [00143] Translating TDD into approximate days of duration with the aid of Fig.41, the estimated time span of the total 460 °C-days accumulation during summer in the Chukchi Sea based on 1991-2013 data from Francis et al. (2016) is roughly from June 1 (ordinal date 152) to September 22 (ordinal date 266) – or about 114 days (170 to 172 in Fig.41). Note that the area under the curve and above the 0 TDD/day horizontal line (173) is equal to the total 460 TDD annual accumulation in the Chukchi Sea region. [00144] In comparison, the 365 TDD heat absorbing capacity of singly rafted 2 m ice melting down to about 0.6 m is about 81 days duration – lasting to about August 21 (i.e., 233 – 152 ordinal days; 180 and 181 in Fig.41). For the 151 TDD capacity estimated for the 2 m thick unrafted ice melting down to 0.6 m (177 and 178 in Fig.41), its duration is about 42 days, lasting to about July 13 (194 – 152). This implies walrus hauled out on remnant pack at Hanna Shoal, would need to swim for shore roughly in mid-July, which appears to align with a mass haulout reported on July 25 ^th at Point Lay, Alaska in 2019 (Rosen, 2019). [00145] As another empirical check, forty-two days of melting from 2 down to 0.6 m roughly aligns with an observation by Pounder (1965, p.143) that it is possible for 2-meter thick Arctic sea ice to completely melt within 6 weeks, though if ‘completely’ is to be taken literally as zero ice remaining, Equation 10 predicts nearly 8 weeks, though 6 may be plausible by altering the melt rate, which is chosen here for comparison. Based on the above, it is suggested that singly rafted 2-meter ice (two layers) may last about 81/42 or about twice as long as unrafted. Notice that if the TDD/day rate were constant rather than climbing nearly to the end of August – around ordinal day 235 – the rafted ice would perform still better. [00146] While the prospect of an additional 5 or 6 weeks of pack ice habitat remnants may seem modest, in the context of losing about 10.4 days per decade of freeze-up season in the Chukchi and Beaufort Seas (Markus et al., 2009, pp.10-11), this may be significant enough to locally counter decline, even if only in local patches, which may be enough for minimum viable populations to survive for some extended time. [00147] Assuming the above calculations and reasoning provide an initial indication for rafting and freeze-bonding to improve endurance of ice floes, it follows that there will be an extension of drift range when compared to un-rafted ice, and broader spatial and temporal distribution of the ice, fauna and flora on and within. The eventual melting of the ice, and transfer of much of these organisms (including 330 to 342) and accompanying pack ice nutrients into the water column 204 may then be dispersed in a pattern more consistent with conditions prior to the recent and projected effects of climate change. SECOND PREFERRED EMBODIMENT [00148] Referring to Fig.15, there is shown a flow diagram for a preferred embodiment of a method for restoring a sea ice habitat for semi-aquatic marine mammals and the other aforementioned species, generally designated 400 and hereafter referred to as the “habitat restoration method, 400 in accordance with the present invention. [00149] As a first step, the habitat restoration method 400 has a mooring step 410 in which a sea ice habitat restoration platform 10 is moored in a marine ecosphere having pack ice. In a preferred embodiment of the habitat restoration method 400, the sea ice habitat restoration platform that is moored in the marine ecosphere is the sea ice habitat restoration platform 10 described above. [00150] In an extracting step 412 after the mooring step 410, a sea ice slab 216 is extracted from the drifting pack ice 208. In a preferred embodiment of the habitat restoration method 400, the pack ice slab 216 is extracted using the forward ice-ramp portion 42 of the sea ice habitat restoration platform 12. [00151] In a cracking step 414 a bending moment is applied to the extracted sea ice slab 216 to crack the sea ice slab. In a preferred embodiment of the habitat restoration method 400, the hogging step 414 cracks the extracted sea ice slab under its own weight with the vertex 62 of the forward ice-ramp portions 42 of the sea ice habitat restoration platform 12 as the extracted pack ice slab longitudinally traverses the forward ice-ramp portions 42 at a ramp velocity about equal to the drift speed v of the oncoming drifting pack ice 208. [00152] In a channel forming step 413, occurring simultaneously with the extracting step 412, cracking step 414, and deposition step 418, a micro polynya channel is formed in the marine ecosphere aft of the habitat restoration platform 12. The micro polynya channel 236 is formed by passing pack ice 210 having a free edge 220 forming a bounding edge of the channel. In a preferred embodiment, the micro polynya channel has a port channel free edge 220 spaced from a starboard channel free edge 220 a channel width apart. In some embodiments, the channel width corresponds to the beam 14 of the marine platform 12. [00153] In a depositing step 416, the extracted sea ice slab 216 is deposited as rafted pack ice 226 on passing pack ice 210 setback from the free edge 220 of the pack ice to form a terraced ice edge 229 facilitating semi-aquatic marine mammal haulout and maximizing floodwater 232 volume. In a preferred embodiment of the depositing step 416, the extracted sea ice slab 216 is deposited as rafted pack ice (226 with freeze-bonding 234; doubly rafted 228 without freeze-bonding; and canted-rafted 230 with partial freeze-bonding) on passing pack ice 210 setback from the free edge 220 of the pack ice aft ice-ramp portion outlet zones 50 of the double inclined-plane ice ramp 28 of the marine platform 12. In the case where one of the focal species for the restoration program includes ringed seals 303 and improved lair- making opportunities is one of the habitat restoration goals, the included angle ^ may be set > 90°. If the case where the focal species is walrus, then ^ may be set to < 85° and greater than about 75°. [00154] Freeze-bonding 234 of the rafted pack ice on passing pack ice provides a structurally enhanced, tiered icescape amenable to semi-aquatic marine mammal haulout onto the icescape with the passage of sufficient freezing degree days after the deposition of the rafted pack ice. If freeze-bonding 234 does not substantially take place, structural enhancement still occurs, though to a lesser degree than that provided by the composite action of a freeze-bonded assembly. Flooding 232 of the level ice 210 due to rafting surcharge loads also provides a method of new ice habitat production (239, 240 and up through FY or MY ice limits for the region) given sufficient freezing degree days (FDD) thereafter. [00155] The foregoing detailed description of the invention has been disclosed with reference to specific embodiments. However, the disclosure is not intended to be exhaustive or to limit the invention to the precise forms disclosed. Those skilled in the art will appreciate that changes could be made to the embodiments described above without departing from the broad inventive concept thereof. Therefore, the disclosure is intended to cover modifications within the spirit and scope of the present invention as defined by the appended claims.

APPENDIX A. DERIVATION OF ICE SEGMENT BUCKLING CRITERION ^ Introduction [00156] Preliminary ice tank testing by this author demonstrates that the low slope ramps of the first preferred embodiment that produce the least damage to the extracted ice in the rafting process also tend to be more liable to rubble pileups, which may be triggered by buckling of ice segments off of the ramp surfaces. A simple method to quantify ramp curvature with respect to a critical buckling load induced has been found helpful in troubleshooting or assessing if a design is developed enough to be worth the time and cost of a scale model for tank testing. [00157] To this end a criterion ^ ^to estimate onset of ice segment buckling is defined where the upward out-of-plane component F _y of drift force F applied to an ice segment moving over a 2D convex surface is resisted by the sum of frictional forces Fs induced by upstream ice. As idealized in the free body diagram of Fig.40 and the accompanying equations below, as the radius of curvature r decreases from infinity - that of a plane, the out- of-plane buckling force Fy increases to the point where the self-weight of the ice segment (free body 269) no longer resists it and a pair of segments - 268 and 269 - lift off the surface (dashed) rotating about opposing hinge points of ice segments still in contact with ramp 139. [00158] The ratio of characteristic length l _c to radius of curvature r measured tangent to the direction of ride-up (for example, along a path 218 for any portion of a ramp in the first preferred embodiment) must be less than a maximum stable ratio established by ^ such that: ^ ^ ^ = 4 (-2 ^ _s n ^ ± [4( ^ _s n ^) ² + 1] ^1/2 ) > l _c (Eqn.9) b r  where: ^ is the ice sheet buckling criterion (-); r is radius of curvature (m) aligned with drift force F; ^s is coefficient of static friction (-); n (-) is number of ice slab segments upstream of the potential buckling zone resisting sliding; ^ is safety factor (-) against out-of-plane buckling; b (-) is the minimum of A/l _c or ^/4; A is length (m) along ramp vertex from point aligned with strip shearing away from level ice up to prow; and lc is characteristic length (m) as determined by Equation 2. Assumptions [00159] The following assumptions in the free body diagram will tend to hold true for full- scale sea ice and be less true for models in which the scale factor ^ is greater than 40 (i.e., very small models relative to full-scale prototypes which are generally avoided by ice tank test facilities; See Tatinclaux, 1988, p.11): 1. Ice flexural rigidity EI is sufficiently high such that any bending of ice segments under their own weight to the point that they begin to conform to the convex curve may be neglected. 2. Surface tension (Van der Waals forces) between the ice segments and ramp surface are negligible. The higher the scale factor ^ (the smaller the model), the more surface tension of the liquid phase in which the ice floats may play a role in the segments clinging to the curved surface. 3. Loading rate is quasistatic. Maximum Ice Segment Length [00160] The maximum length L of extracted ice segments (267 to 270) riding up the ramp 139 is estimated to be the smaller length resulting from two load cases that crack the strip transverse to its direction of ride-up: [00161] Case 1) Failure of the extracted ice strip with respect to weak axis bending (in combination with torsion, shear, compression, and other loads that will be conservatively assumed as negligible). This weak axis failure length will be assumed equal to that of a floating cantilever ice beam with an upward tip load provided by contact with the ramp (e.g., 42 or further up the ramp in the first preferred embodiment), which is equal to about ^l _c /4 (Hetényi, 1946, p.24; Daley, 2020, p.86). [00162] Case 2) Failure of the strip with respect to strong axis bending due to lateral prying by a prow or wedge wall, which results in a length L of about A - the inclined length of the ramp starting from a point on the vertex aligned with shear points (48 in the first preferred embodiment), up to the base of the prow or wedge wall first contacted. [00163] If a strip of extracted ice is not broken in Case 1, the first it will be exposed to as it is lifted from the water, the strip should be broken in flexure against its strong axis as the prow pries it laterally, rotating it about its base near the notch formed at the shear point. Failure may also occur at some other stress riser along its length such as a crack or other discontinuity, but then the segment length L will be less than A, and pose less of a buckling risk. Thus, ice segment maximum length L is defined such that: L ≅ b l _c = MIN(A, ^l _c /4) where: L is maximum length (m) of a typical ice segment measured along ride-up path of a ramp 139 (path 218 in the first preferred embodiment); MIN() is a function that returns the lowest input value; b is a constant (-) as determined below; A is length along the ramp from point of strip shearing away from level ice to prow (m); lc is characteristic length (m) as defined in Equation 2; Variable b may also be expressed directly as: b = MIN(A/l _c , ^/4) Static Analysis [00164] For forces in the x-direction, using the xy coordinate plane of Fig.40, the horizontal component Fx of drift force F is assumed just equal to the resisting force Fs of the ice segments upstream on the ramp. The resisting force is the sum of the products of the coefficient of static friction and the normal force of each segment. These two opposing horizontal components are then:  Fx = F cos2 ^ and Fs = n ^sN = n ^s ^ig Ldh where: ^ ^is an angle (radians) from ice segment midpoint to edge n is number of ice slab segments upstream of the potential buckling zone that resist sliding; ^s is coefficient of static friction (-); ρi is sea ice density (kg/m ³ ); g is acceleration of gravity (m/s ² ); L is ice segment length (m) defined above; d is segment depth (m) normal to section, equal to 1; and h is ice segment thickness (m); The ice segment weight w is used as the normal force N since the portion of the ramp in the diagram on which it bears is level. The normal force for any given curve may be derived to suit a particular condition, though assuming the upstream curve to be level is a conservative, simple estimate suggested for a preliminary calculation. Setting the horizontal drift force Fx equal to the frictional resistance Fs gives: F cos2 ^ ^= n ^s ^ig Ldh For static equilibrium in the y-direction, vertical force component Fy is assumed just equal and opposite to the tributary self-weight of ice slab 269 that it supports, w/2, such that: F sin2 ^ ^=  ^ig Ldh 2  where vertical component of drift force Fy and slab weight w are: Fy = F sin2 ^ ^ ^ ^ ^ ^and ^ ^ ^ ^w =  ^ig Ldh ^ To provide a factor of safety against uplift of slab 269, its weight is divided by ^ which has some value greater than 1 as selected by the designer such that: W =  ^ig Ldh 2 ^ Dividing the x-direction equality by the y-direction equality yields:  F sin2 ^   =   ^ _i g Ldh/2 ^   ^ F cos2 ^       ^ _s n ^ ^ig Ldh Simplifying yields: tan2 ^     = 1/2 ^ _s n ^ ^ Substituting 1/ ^ _s n ^ ^for ^tan2 ^ into ^the double-angle identity yields: tan2 ^ = 2 tan ^ ^ (1 - tan ² ^) 1 = 2 tan ^ ^ 2 ^sn ^ (1 - tan ² ^) Given that tan ^ ^ ^ L/2 based on the free body diagram, and substituting L/2r for tan ^ ^gives ^ r 1 = 2 (L/2r) ^ 2 ^ _s n ^ 1- (L/2r) ² Cross multiplying: 1-(L/2r) ² = 4 ^ _s n ^ (L/2r) Rearranging into quadratic form: (L/2r) ² + ^ ^ ^ ^sn ^) (L/2r) - 1 = 0 Letting x = L/2r: x ² + ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^sn ^) x - 1 = 0 Solving for x such that x = (-b ± (b ² - 4ac) ^1/2 )/2a: x = -4 ^ _s n ^ ± [(4 ^ _s n ^) ² - 4(1)(-1)] ^1/2 2(1) x = -4 ^ _s n ^ ± [16( ^ _s n ^ ^ ² + 4] ^1/2 2 Substituting L/2r back into the solution for x: L = -4 ^ _s n ^ ± [16( ^ _s n ^ ^ ² + 4] ^1/2 2r 2 Substituting the ice segment length based on characteristic length where L = bl _c : blc = -4 ^sn ^ ± [16( ^sn ^) ² + 4] ^1/2 2r 2 Isolating r and lc as a ratio yields: lc = (2/b)(-4 ^sn ^ ± [16( ^sn ^) ² + 4] ^1/2 ) r Substituting the definition of characteristic length and introducing scale factor ^ ^into all dimensions and elastic modulus to ensure similitude is maintained across scales (Tatinclaux, 1988, p.5; Note that coefficient of friction is not scaled) yields: [(E/ ^)(h ³ / ^ ³ )/12 ^g(1 - ^ ² )] ^1/4 = (2/b)(‐4 ^ _s n ^  ±  [16( ^ _s n ^) ²  + 4] ^1/2  )  (r/ ^ ^ ^ ^ (1/ ^ ^{^} ) ^1/4 [Eh ³ /12 ^g(1 - ^ ² )] ^1/4 = (2/b)(‐4 ^ _s n ^  ±  [16( ^ _s n ^) ²  + 4] ^1/2  )  (1/ ^) r Thus, ^ ^ ^ ^ at top and bottom cancel indicating similitude is maintained.     Setting ^ equal to the righthand side and requiring that it be greater than the ratio of l _c /r ensures buckling resistance with a factor of safety  ^: ^ = 4 (‐2 ^ _s n ^  ± [4( ^ _s n ^) ²  + 1] ^1/2  )  >   lc     Q.E.D.   b r 

REFERENCE SIGNS LIST PLATFORM - SURFACE GEOMETRY AND MOVEMENT 0-199 [00165] 10. Sea ice habitat restoration platform [00166] 12. Marine platform [00167] 14. Beam [00168] 18. Bow [00169] 20. Stern [00170] 22. Longitudinal centerline [00171] 24. Port side [00172] 26. Starboard side [00173] 28. Double-inclined-plane ice ramp [00174] 29. Forebody [00175] 30. Port ice-ramp portion [00176] 32. Starboard ice-ramp portion [00177] 34. Forward port ice-ramp portion [00178] 36. Aft port ice-ramp portion [00179] 38. Forward starboard ice-ramp portion [00180] 40. Aft starboard ice-ramp portion [00181] 42. Forward ice-ramp portion [00182] 44. Aft ice-ramp portion [00183] 46. Forward ice-ramp portion inlet [00184] 48. Forward ice-ramp portion shearing zones [00185] 50. Aft ice-ramp portion outlet zones [00186] 52. Bow centerline [00187] 54. Rafting wedge [00188] 56. Rafting wedge prow [00189] 58. Port rafting wedge wall [00190] 59. Top curve of port rafting wedge wall [00191] 60. Starboard rafting wedge wall [00192] 61. Top curve of starboard rafting wedge wall [00193] 62. Vertex [00194] 64. Lower bounding curve (or directrix) [00195] 66. Upper bounding curve (or directrix) [00196] 68. Horizontally oriented surface [00197] 70. Vertically oriented surface [00198] 72. First curve from which horizontally oriented surface is extruded [00199] 74. Forward portion of first curve [00200] 76. Slope of forward portion of first curve [00201] 78. Arc of first curve [00202] 80. Aft portion of first curve [00203] 82. Second curve from which vertical oriented surface is extruded [00204] 84. First portion of second curve [00205] 86. Mid portion of second curve [00206] 88. Arc of second curve [00207] 90. Aft portion of second curve [00208] 92. Frustrum of right circular cone [00209] 94. Sinusoidal surface [00210] 96. Top of frustrum 92 [00211] 98. Draft angle of frustrum 92 [00212] 100. Curve from which sinusoidal surface 94 is extruded [00213] 101. Base of wedge wall prow joining curves 102 and 104 [00214] 102. Aft portion of curve 100 [00215] 103. Peak of sine curve 102 [00216] 104. Forward portion of curve 100 [00217] 105. Aft wedge prow [00218] 106. Aft wedge walls [00219] 107. Aft haulout deck [00220] 108. Afterbody [00221] 109. Ice rudder [00222] 110. Lateral surface of ice rudder [00223] 112. Aft surface of ice rudder [00224] 114. Weathervane – yaw in either direction around Az axis [00225] 116. Heave - rising and falling with surrounding waves along A _z axis [00226] 118. Mooring assembly [00227] 120. Pier [00228] 122. Pier canopy [00229] 124. Sail [00230] 126. Rack-and-pinion jacking mechanism [00231] 127. Rack anchored to pier 120 [00232] 128. Retractable pinions [00233] 130. Tunnel thruster [00234] 132. Ice shield [00235] 134. Beveled down ramp inlet for pressure ridge breaking [00236] 136. Filleted edge between pressure ridge breaking surface 134 and forward ice-ramp portion inlet 42 [00237] 138. Bow stem [00238] 139. Generalized ramp section with 2D convex curvature of radius r [00239] 160. Curve of thawing degree days per day (TDD/day) for Chukchi Sea [00240] 170. Approximate start of Chukchi Sea melt season (ordinal day) [00241] 172. Approximate end of Chukchi Sea melt season (ordinal day) [00242] 173. Horizontal line between thawing degree days/day (above line) and freezing degree days/day (below), crossing the y-axis of Fig.41 at 0 TDD/day. OCEAN, ICE AND ATMOSPHERE 200-299 [00243] 200. Design waterline (DWL) of platform [00244] 201. Waterline (WL) of floating objects other than the platform [00245] 202. Sea surface [00246] 204. Water column [00247] 205. Ice-water interface [00248] 206. Design wave for design sea state [00249] 208. Pack ice drifting towards platform [00250] 210. Pack ice drifted by either side of the platform [00251] 212. Rubble ice – randomly piled ice slabs from mechanical deformation [00252] 214. Pack ice drift direction [00253] 216. Pack ice riding up ramp impelled by pack ice drift pressure [00254] 218. Pack ice ride-up and traversal paths along and off of ramp surfaces [00255] 220. Free edge of pack ice micro polynya created by rafting [00256] 221. Refrozen edge of pack ice channel downdrift of micro polynya [00257] 222. Ice flexure crack from hogging moment at ramp vertex 62 [00258] 224. Ice slab sliding off ramp - either driven by pack ice drift force, sliding under gravity, or combination thereof [00259] 226. Pack ice slab rafted flat onto level ice [00260] 228. Top ice slab of doubly rafted ice slabs [00261] 229. Terraced edge formed by rafted ice set back from channel edge [00262] 230. Canted-rafted ice slab partly supported on flat rafted slab and partly on level ice [00263] 232. Floodwater due to submersion of level ice under rafted ice strip load [00264] 234. Frozen floodwater freeze-bonding of rafted slab to level ice [00265] 235. Pack ice disturbance corridor [00266] 236. Micro polynya channel [00267] 237. Ice flexure crack parallel to channel due to rafted ice load on level ice [00268] 238. Ocean-atmosphere exchange: heat flux and frost smoke [00269] 239. New ice – newly formed sea ice crystals including frazil, nilas, grease ice less than 10 cm [00270] 240. Young ice – newly formed sea ice between 10-30 cm; successor of new ice [00271] 242. Consolidated frazil ice – first layer to form, randomly oriented crystals [00272] 244. Columnar, congelation ice – second layer to form below frazil [00273] 245. Brine channels [00274] 246. Skeletal ice – leading edge of bottom ice growth [00275] 248. Melting rafted ice floe (in spring and summer) - remnant of pack ice disturbance corridor from winter [00276] 250. Rotting rafted ice floe remnant (248 in advanced state of melting) [00277] 251. Brine flux into the water column from sea ice formation [00278] 252. Melt pond [00279] 259. Snow transformed into ice by rafting, immersion and freeze-bonding [00280] 260. Snow, which may or may not overlay the sea ice [00281] 261. Wind streamline [00282] 262. Windward snow drift [00283] 264. Leeward snow drift [00284] 265. Snow drift contour line (heights of equal elevation) [00285] 266. Rain [00286] 267. Idealized ice slab segment of thickness h and length L [00287] 268. Idealized ice slab segment about to buckle off ramp [00288] 269. Idealized ice slab segment (the free body) about to buckle off ramp with hinging opposite to 268 [00289] 270. Idealized ice slab segment exerting resistive force Fs due to static friction HABITAT AND PACK ICE COMMUNITIES 300-399 [00290] 300. Pagophilic phocid seal(s) hauled out, including: bearded seal (Erignathus barbatus) – ice-obligate ringed seal (Phoca hispada) – ice-obligate harp seal (Pagophilus groenlandicus) – ice-associated hooded seal (Cystophora cristata) – ice-associated ribbon seal (Histriophoca fasciata) – ice-associated spotted seal (Phoca largha) – ice-associated All prey on Arctic cod (Boreogadus saida), at times capelin (Mallotus villosus), other fishes, and crustaceans [00291] 301. Pagophilic phocid species swimming [00292] 302. Breathing hole of pagophilic phocid [00293] 303. Ringed seal (Phoca hispada) – constructs subnivean lairs for shelter, birthing and nursing when snow is available [00294] 304. Subnivean lair constructed by ringed seal and pup [00295] 306. Ice lair of pagophilic phocid – with or without snow drifting [00296] 307. Walrus (Odobenus rosmarus) swimming – ice-obligate, key predator of bivalve clams [00297] 308. Walrus hauled out [00298] 309. Walrus hauling out on or diving from pack ice [00299] 310. Polar bear (Ursus maritimus) – ice-obligate, key predator of pagophilic phocid seals, especially ringed seals, bearded seals, occasionally walrus or carrion [00300] 314. Spectacled eider (Somateria fischeri) – ice-obligate sea duck that dives to 70 m for mollusks and other benthic prey; also forage at bottom ice [00301] 316. Black guillemot (Cepphus grylle mandtii) – ice-obligate seabird [00302] 317. Ivory gull (Phagophila eburnea) – ice-obligate seabird [00303] 318. Arctic cod (Boreogadus saida) – ice-associated keystone species [00304] 330. Top ice community – including flagellates, ice algae, bacteria, virus (Arrigo et al., 2010, p.284-285; van Leeuwe et al., 2018, Fig.4) [00305] 331. Top ice community transformed by rafting into an internal community [00306] 332. Internal ice community – generally associated with frazil ice layer near waterline, including sea ice algae, flagellates, bacteria, virus (Arrigo et al., 2010, p.284-285; van Leeuwe et al., 2018, Fig.4) [00307] 334. Bottom ice community – including sea ice algae (predominately pennate diatoms by spring), ciliates, flagellates, bacteria, virus, fungi, nematodes, rotifers, flatworms, polychaetes, nauplii, amphipods, harpacticoid copepods (van Leeuwe et al., 2018, Fig.4; Lund-Hansen et al., 2020, p.4) [00308] 335. Bottom ice community transformed by rafting into internal community [00309] 336. Strand community – including colonial diatoms such as Melosira arctica (Arrigo et al., 2010, p.291) [00310] 337. Sea ice algae especially diatoms of genera Nitzschia, Fragilariopsis and Navicula (Arrigo et al., 2010, p.290) [00311] 338. Sea ice copepods (typically 0.5-2mm) that may graze sea-ice algae - including Cyclopina gracilis, Cyclopina schneideri, Harpacticus spp., Halectinosoma spp., Tisbe furcate. Cyclopina schneideri and Tisbe furcata may migrate to the benthos during open water season [00312] 339. Sea ice amphipods including Onisimus nanseni, Gammarus wilkitzkii and Apherusa glacialis – key prey of Arctic cod [00313] 340. Sea ice nematodes including Theristov melnikovi [00314] 342. Tuuli's hydroid (Sympagohydra tuuli) [00315] 350. Sea ice-derived organic carbon flux, enhanced with rafting and new ice growth – a component of marine snow – flakelike aggregates of detritus composed of decaying organisms or fecal matter drifting to the seafloor. It is a transfer of nutrients produced in the photic zone to the relatively aphotic benthos, and is part of the biological carbon pump that sustains the benthos, which in turn sustains walrus, ice-obligate diving seabirds, gray whales and other species. [00316] 351. Seafloor of the Arctic continental shelves [00317] 352. Benthos – community of continental shelf organisms broadly categorized as epifauna and infauna [00318] 354. Epifauna – benthic organisms living on or moving over the seafloor [00319] 356. Infauna – benthic organisms living within the upper layer of the seafloor [00320] 358. Brittle star species including Ophiura sarsii and Ophiocten sericeum [00321] 360. Common sunstar (Crossaster papposus) [00322] 362. Bivalve mollusk species including blunt gapper (Mya truncata), Greenland smooth cockle (Serrepes groenlandica), Arctic Hiatella (Hiatella arctica), Chalky macoma (Macoma calcarea); all circumpolar in distribution and typical are prey of walrus (Fay, 1982) [00323] 364. Burrowing polychaete including Phyllodoce groenlandica [00324] 366. Oligochaete worm [00325] 368. Sea cucumber including orange-footed sea cucumber (Cucumaria frondose) [00326] 370. Benthic crab species including Arctic lyre crab (Hyas coarctatus), Snow crab (Chionoecetes opilio), which are occasional prey of pagophilic phocids and walrus (molted forms) [00327] 372. Sieve kelp species including Agarum cribrosum [00328] 374. Benthic amphipods including Eusirus holmii (which may also reside in sea ice and the water column), and Gammarus setosus (benthic only). All are prey of benthivorous seabirds, sea ducks and gray whales. [00329] 375. Benthic fishes including Arctic staghorn sculpin (Gymnocanthus tricuspis), Hamecon (Artediellus scaber) [00330] 376. Bioturbation of benthos and seafloor sediments RAFTING METHODS 400-499 [00331] 400. Habitat restoration method [00332] 410. Mooring step [00333] 412. Extracting step [00334] 413. Micro polynya channel forming step [00335] 414. Cracking step [00336] 416. Depositing step

MATHEMATICAL SYMBOLS LIST [00337] a Sea ice melt rate constant (m/°C-days) [00338] A Maximum extracted ice segment length (m) based on ramp geometry [00339] A _x Longitudinal axis of platform [00340] Ay Transverse axis of platform [00341] A _z Vertical axis of platform [00342] b Constant real number (-) [00343] B Beam of platform (m) [00344] d Depth of ice segment (m), normal to section plane [00345] D _min Offset distance (m) from centerline of rafted ice to channel edge where rafted ice is deposited up to the channel edge [00346] D _cr Offset distance (m) of centerline to crack in pack ice due to hogging moment induced by linear rafted ice load to channel edge [00347] D _cr-vmax Offset distance (m) of rafted ice centerline to channel edge for the fully cracked condition producing maximum floodwater volume and adequate immersion depth [00348] Del-vmax Offset distance (m) of rafted ice centerline to channel edge for the uncracked, elastic condition producing maximum floodwater volume and adequate immersion [00349] D Preferred offset distance (m) of rafted ice centerline from free edge of channel [00350] E Ice elastic modulus (Pa) [00351] F 2D force vector (kN) [00352] Fs 2D force vector (kN) equal to the product of the coefficient of static friction ^ _s (-) and normal force (kN) due to ice self-weight [00353] FDD Freezing degree days (°C-days) [00354] g Acceleration of gravity (m/s ² ) [00355] h Nominal level pack ice thickness (m) [00356] hf Final sea ice thickness (m) after melting [00357] h _i Initial ice thickness (m) prior to melting [00358] hpctl Ice thickness (m) of a specified percentile rank [00359] H Height (m) from design waterline to bottom of cantilever ramp tip [00360] H _s Mean height (m) of pack ice pressure ridge for area of deployment [00361] lc Level sea ice characteristic length (m) [00362] L Length (m) of idealized ice slab segment riding up or rafting from ramp [00363] M Hogging moment (kN-m) [00364] N Normal force (kN) [00365] r Radius of curvature (m) [00366] R Forward ice-ramp portion inlet half-width (m) [00367] S Maximum ramp cantilever tip distance (m) from platform centerline [00368] T Draft (m) [00369] TDD Thawing degree days (°C-days), negative of FDD [00370] U Height (m) of rafting wedge wall [00371] v Sea ice (pack ice) drift speed (m/s) [00372] w Ice slab self-weight (kN) [00373] W Rafting wedge wall width (m) [00374] x Independent variable [00375] z Z-score of a normal distribution [00376] α Slope of double incline-plane ice ramp vertex (deg) with respect to horizontal [00377] ^ Slope of pressure ridge breaking surface at bow centerline (deg) [00378] ^ Included angle between wedge wall and adjacent ramp surface (deg) [00379] δ Transverse slope of port and starboard ice-ramp portions (deg) [00380] ε Rafting wedge prow half angle (deg) in plan view [00381] ^ Fillet radius of curvature (m) at beveled down edge at bow centerline [00382] ^ ^ Ice segment buckling criterion (-) [00383] ^ Scale factor of model to full-scale prototype (-) [00384] ^ Wavelength (m) [00385] ^ ^ Mean design level pack ice thickness (m), not including ridges [00386] ^s Coefficient of static friction (-) [00387] ^ Poisson’s ratio (-) [00388] ^ Right circular cone draft angle (deg) [00389] ρ Seawater density (kg/m ³ ) [00390] ρ _i Sea ice density (kg/m ³ ) [00391] ^ ^ Standard deviation (same dimensions as mean) [00392] ^^ _^ Sea ice flexural strength (Pa) [00393] ^ Bow stem angle (deg) [00394] ^ Angle (radians) between centerline and edge of ice segment of length L on 2D ramp with radius of curvature r [00395] ^ Safety factor (-) ABBREVIATIONS LIST [00396] CL Centerline [00397] DWL Design waterline of platform [00398] FY First year (ice) [00399] MIZ Marginal ice zone [00400] MY Multiyear (ice) [00401] WL Waterline of floating objects other than the platform CITATIONS LIST Patent Literature [00402] Braley, W. (1984), Ice-Breaking Hull, U.S. Patent No.4,436,046. [00403] Stangebye, C. (1907), Boat for Ice Breaking and Other Purposes, U.S. Patent No. 857,766. [00404] Waas, H. (1976), Icebreaker Vessel, U.S. Patent 3,985,091.

Non-Patent Literature [00405] Al-Harthi, A. A., Al-Amri, R. M., & Shehata, W. M. The porosity and engineering properties of vesicular basalt in Saudi Arabia. Engineering Geology, Vol.54, no.3-4, (Oct.1, 1999), pp.313-320. [00406] Alonso‐González, I. J., Arístegui, J., Lee, C., Sanchez‐Vidal, A., Calafat, A., Fabrés, J., ... & Benítez‐Barrios, V. Role of slowly settling particles in the ocean carbon cycle. Geophysical Research Letters, Vol.37, no.13, (Jul.15, 2010). [00407] Anderson, D. L. Growth rate of sea ice. Journal of Glaciology, Vol.3, no.30, (1961), pp.1170-1172. [00408] Arias, P., Bellouin, N., Coppola, E., Jones, R., Krinner, G., Marotzke, J., ... & Zickfeld, K. Climate Change 2021: The Physical Science Basis. Contribution of Working Group14 I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change; Technical Summary, (2021). [00409] Arrigo, K. R., Mock, T., & Lizotte, M. P. “Primary Producers and Sea Ice.” in: Thomas, D.N. and Dieckmann, G.S., Sea Ice, 2 ^nd Ed., (Wiley-Blackwell, 2010), pp.283- 325. [00410] Arrigo, K. R. The changing Arctic Ocean. Elementa: Science of the Anthropocene, 1. (Jan.1, 2013). https://online.ucpress.edu/elementa/article/doi/10.12952/jou rnal.elementa.000010/112329/Th e-changing-Arctic-OceanThe-changing-Arctic-Ocean [00411] Bailey, E., Feltham, D. L., and Sammonds, P. R. A model for the consolidation of rafted sea ice. Journal of Geophysical Research: Oceans, Vol.115, No. C4, (Apr.9, 2010). https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2008JC00 5103 [00412] Baveye, P., Vandevivere, P., Hoyle, B.L., DeLeo, P.C., and Sanchez de Lozada, D., Environmental Impact and Mechanisms of the Biological Clogging of Saturated Soils and Aquifer Materials, Critical Reviews in Environmental Science and Technology, Vol.28, no.2 (1998), pp.123-191. DOI: 10.1080/10643389891254197 [00413] Belikov, S., Boltunov, A., and Gorbunov, Y. DISTRIBUTION AND MIGRATION OF POLAR BEARS, PACIFIC WALRUSES AND GRAY WHALES DEPENDING ON ICE CONDITIONS IN THE RUSSIAN ARCTIC, 17th Symposium on Polar Biology, (1996), pp.263-274. [00414] Bilello, M. A. Maximum thickness and subsequent decay of lake, river, and fast sea ice in Canada and Alaska, US Army, Corps of Engineers, Cold Regions Research and Engineering Laboratory, Vol.80, no.6 (Feb.1980), pp.29-34. [00415] Bluhm, B.A., Gradinger, R.R., and Schnack-Schiel, S.B. “Sea Ice Meio- and Macrofauna.” in: Thomas, D.N. and Dieckmann, G.S., Sea Ice, 2nd Ed. (Wiley-Blackwell, 2010), pp.357-394. [00416] Brown, T. G. Confederation Bridge – An innovative approach to ice forces, in: Proceedings of the Annual Conference of the Transportation Association of Canada, Prince Edward Island (2006). https://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.10 53.9873&rep=rep1&type=pdf [00417] Brown, T. A., Alexander, C., Yurkowski, D. J., Ferguson, S. H., & Belt, S. T. Identifying variable sea ice carbon contributions to the Arctic ecosystem: A case study using highly branched isoprenoid lipid biomarkers in Cumberland Sound ringed seals. Limnology and Oceanography, Vol.59, no.5 (2014), pp.1581-1589. https://aslopubs.onlinelibrary.wiley.com/doi/pdf/10.4319/lo. 2014.59.5.1581 [00418] Brown, T. A., Galicia, M. P., Thiemann, G. W., Belt, S. T., Yurkowski, D. J., & Dyck, M. G. High contributions of sea ice derived carbon in polar bear (Ursus maritimus) tissue. PloS one, Vol.13, no.1 (Jan.23, 2018). e0191631. https://journals.plos.org/plosone/article?id=10.1371/journal .pone.0191631 [00419] Dyer, Ira. "Speculations on the origin of low frequency Arctic Ocean noise." in: Kerman, B.R., Sea Surface Sound (Dordrecht, Kluwer Academic Publishers, 1988), pp.513- 532. [00420] Burns, J. J., Frost, K. J., & Lowry, L. F. (Eds.). Marine Mammals Species Accounts. Alaska Department of Fish and Game, Division of Game, Mar.1985. [00421] Cammaert, A., and Muggeridge, D. B. Ice Interaction with Offshore Structures. New York, Van Nostrand Reinhold, 1988. [00422] Cameron, M. F., Bengtson, J. L., Boveng, P. L., Jansen, J. K., Kelly, B. P., Dahle, S. P., ... & Wilder, J. M. Status review of the bearded seal (Erignathus barbatus), U.S. Department of Commerce, NOAA Technical Memorandum NMFS-AFSC-211 (Dec.2010). https://www.researchgate.net/profile/Brendan-Kelly- 6/publication/270571869_Status_review_of_the_bearded_seal_Er ignathus_barbatus/links/56 abd42708ae19a385115764/Status-review-of-the-bearded-seal-Eri gnathus-barbatus.pdf [00423] Chen, J., Blume, H. P., & Beyer, L. Weathering of rocks induced by lichen colonization - a review. Catena, Vol.39, no.2, (Mar.1, 2000), pp.121-146. [00424] Conlan, K. E., Lenihan, H. S., Kvitek, R. G., & Oliver, J. S. Ice scour disturbance to benthic communities in the Canadian High Arctic. Marine Ecology Progress Series, Vol. 166, (May 28, 1998), pp.1-16. [00425] Daley, C. Sea Ice Engineering: theory and application. Lecture note for EN8674/9096. Memorial University of Newfoundland, 2020. https://www.engr.mun.ca/~cdaley/8074/IceTextv10j.pdf [00426] Dieckmann, G. S., and Hellmer, H. H. “The Importance of Sea Ice: An Overview.” in: Thomas, D.N. and Dieckmann, G.S., Sea Ice, 2 ^nd Ed., (Wiley-Blackwell, 2010), pp.1-22. [00427] Den Boer, P. J. Spreading of risk and stabilization of animal numbers. Acta biotheoretica, Vol.18, no.1 (Mar.1968), pp.165-194. [00428] Eisendle‐Flöckner, U. and Hilberg, S. Hard rock aquifers and free‐living nematodes–an interdisciplinary approach based on two widely neglected components in groundwater research. Ecohydrology, Vol.8, no.3 (Apr.2015), pp.368-377. [00429] Fay, F. H. “The role of ice in the ecology of marine mammals of the Bering Sea.” in: Hood, D.W. and Kelley, E. K., Oceanography of the Bering Sea (Institute of Marine Science, Univ. Alaska, Fairbanks, 1974). pp.383-399. [00430] Fay, F. H. Ecology and biology of the Pacific walrus, Odobenus rosmarus divergens Illiger. U.S. Department of the Interior, Fish and Wildlife Service, American Fauna, No.74 (1982), pp.1-279. [00431] Ferguson, S. H., Stirling, I., and McLoughlin, P. Climate change and ringed seal (Phoca hispida) recruitment in western Hudson Bay. Marine Mammal Science, Vol.21, no.1 (Jan.2005) pp.121-135. [00432] Fischbach, A. S., Monson, D. H., and Jay, C. V. Enumeration of Pacific walrus carcasses on beaches of the Chukchi Sea in Alaska following a mortality event, September 2009. U.S. Department of the Interior, U.S. Geological Survey, Open-File Report 2009–1291. (2009). https://pubs.usgs.gov/of/2009/1291/pdf/ofr20091291.pdf [00433] Forman, R.T.T., and Godron, M. Landscape Ecology. New York, Wiley & Sons, 1986. [00434] Forman, R. T.T., Land Mosaics: The ecology of landscapes and regions. Cambridge, Cambridge University Press, 1995. [00435] Francis, O., Metzger, A.T., Statscewich, H. & Winsor, P. Physical Oceanographic and Meteorological Data for the Beaufort and Chukchi Seas to Support Reliability-Based Design Criteria for Arctic Offshore Oil and Gas Structures. US Dept. of the Interior, Bureau of Safety and Environmental Enforcement (BSEE). BSEE contract number: E13PC00014. (Feb.2016). https://catalog.northslopescience.org/dataset/2492/resource/ 23b64223-e73c- 4f72-8afe-f31ad103013c [00436] Garcia-Soto, C., Cheng, L., Caesar, L., Schmidtko, S., Jewett, E. B., Cheripka, A., ... & Abraham, J. P. An Overview of Ocean Climate Change Indicators: Sea Surface Temperature, Ocean Heat Content, Ocean pH, Dissolved Oxygen Concentration, Arctic Sea Ice Extent, Thickness and Volume, Sea Level and Strength of the AMOC (Atlantic Meridional Overturning Circulation). Frontiers in Marine Science. (Sep.21, 2021). https://www.proquest.com/openview/1f69010a5ee5eb0a9a25ae7f5b 26c5f8/1?pq- origsite=gscholar&cbl=2049538 [00437] Golden, K.M., Mathematics of Sea Ice. The Princeton companion to applied mathematics. Princeton: Princeton University Press, 2014. [00438] Grebmeier, J. M., L. W. Cooper, H. M. Feder, and B. I. Sirenko. Ecosystem dynamics of the Pacific‐influenced Northern Bering and Chukchi Seas in the Amerasian Arctic. Progress in Oceanography, Vol.71 (2006), pp.331‐361. [00439] Grebmeier, J. M., and Barry, J. P. “Benthic Processes in Polynyas.” in: Smith, W.O. Jr and Barber, D.G., Polynyas: Windows to the World, Elsevier oceanography series 74. (Amsterdam, Elsevier, 2007), pp.363-390. [00440] Grebmeier, J. M., Bluhm, B. A., Cooper, L. W., Danielson, S. L., Arrigo, K. R., Blanchard, A. L., ... & Okkonen, S. R. Ecosystem characteristics and processes facilitating persistent macrobenthic biomass hotspots and associated benthivory in the Pacific Arctic. Progress in Oceanography, Vol.136 (2015), pp.92-114. [00441] Hahn H.J. and Matzke D.A. Comparison of stygofauna communities inside and outside groundwater bores. Limnologica, Vol.2, no.35(1-2), (May, 2005), pp.31-44. [00442] Hancock, P.J., Boulton A.J., and Humphreys W.F. Aquifers and hyporheic zones: towards an ecological understanding of groundwater. Hydrogeology Journal. Vol.13, no.1, (Mar.2005), pp.98-111. [00443] Harper, J.L., Population Biology of Plants. London, Academic Press, 1977. [00444] Hess G.R., and Fischer R.A. Communicating clearly about conservation corridors. Landscape and Urban Planning. Vol.55, no.3 (Jul 30, 2001), pp.195-208. [00445] Hetényi, M. Beams on Elastic Foundation: Theory with applications in the fields of civil and mechanical engineering. Ann Arbor, University of Michigan Press, 1946. [00446] Hibler W.D. “Arctic Ice-Ocean Dynamics.” in: Herman, Y., The Arctic Seas. (New York, Van Nostrand Reinhold, 1989), pp.47-91. [00447] Hilty, J. A., Lidicker Jr, W. Z., and Merenlender, A. M. Corridor Ecology: The Science and Practice of Linking Landscapes for Biodiversity Conservation. Washington, Island Press, 2006. [00448] Horner, R. A. "Arctic Sea-ice Biota." in: Herman, Y., The Arctic Seas. (New York, Van Nostrand Reinhold, 1989), pp.123-146. [00449] ISO19906, I. S. O. Petroleum and Natural Gas Industries–Arctic Offshore Structures.2 ^nd Ed. Geneva, ISO, Sep.2019. Reference number ISO 19906:2019(E) [00450] Ivanova, S. V., Kessel, S. T., Espinoza, M., McLean, M. F., O'Neill, C., Landry, J., ... & Fisk, A. T. Shipping alters the movement and behavior of Arctic cod (Boreogadus saida), a keystone fish in Arctic marine ecosystems. Ecological Applications, (2020). Vol. 30, no.3, e02050. [00451] Jay, C. V., Farley, S. D., and Garner, G. W. Summer diving behavior of male walruses in Bristol Bay, Alaska. Marine Mammal Science, Vol.17, no.3 (Aug.26, 2006), pp. 617-631. [00452] Jay, C. V., Fischbach, A. S. and A. A. Kochnev. Walrus areas of use in the Chukchi Sea during sparse sea ice cover. Marine Ecology Progress Series (Nov.14, 2012) Vol.468, pp.1–13. doi: 10.3354/meps10057. https://www.int-res.com/articles/feature/m468p001.pdf [00453] Jizu, X., S. Qingzeng, S. An, F. Yunlin, and L. Tongkui. Sea Ice Engineering in China, Journal of Coastal Research, Vol.7, no.3 (Summer 1991), pp.759-770. [00454] Johnston, C. A., & Naiman, R. J. Boundary dynamics at the aquatic-terrestrial interface: the influence of beaver and geomorphology. Landscape Ecology, Vol.1, no.1 (1987), pp.47-57. [00455] Karnovsky, N. J., and Gavrilo, M. V. “A feathered perspective: the influence of sea ice on Arctic marine birds.” in: Thomas, D. N., Sea Ice, 3 ^rd Ed. (Chichester UK, Wiley- Blackwell, 2017), pp.555-569. [00456] Keighley, X., Olsen, M. T., Jordan, P., & Desjardins, S. P. (Eds.). The Atlantic Walrus: Multidisciplinary Insights into Human-Animal Interactions. London, Academic Press, 2021. [00457] Kelly B.P. and Wartzok D. Ringed seal diving behavior in the breeding season. Canadian Journal of Zoology, Vol.74, no.8 (Aug.1, 1996), pp.1547-1555. [00458] Kerr, A. D. The bearing capacity of floating ice plates subjected to static or quasi- static loads: a critical survey. Vol.333 (Corps of Engineers, US Army, CRREL, 1975). [00459] Kohlbach, D., Schaafsma, F. L., Graeve, M., Lebreton, B., Lange, B. A., David, C., ... & Flores, H. Strong linkage of polar cod (Boreogadus saida) to sea ice algae-produced carbon: evidence from stomach content, fatty acid and stable isotope analyses. Progress in Oceanography, Vol.152 (Mar.2017), pp.62-74. [00460] Krembs, C., Eicken, H., and Deming, J. W. Exopolymer alteration of physical properties of sea ice and implications for ice habitability and biogeochemistry in a warmer Arctic, Proceedings of the National Academy of Sciences, Vol.108, no.9 (Mar.1, 2011), pp. 3653-3658. [00461] Leontowich, K. A study of the benthic faunal distribution in the subtidal zone of Turton Bay, Igloolik Island, Nunavut. Faculty of Graduate Studies and Research, University of Regina. (2004). https://ourspace.uregina.ca/bitstream/handle/10294/13127/MQ9 2856.pdf?sequence=1 [00462] Lønne, O. J. “On Productivity in Ice-Covered Polar Oceans.” in: Wettlaufer, J.S., Dash, J.G., and Untersteiner, N., Ice Physics and the Natural Environment (Berlin, Springer, 1999), pp.209-218. [00463] Lund-Hansen, L. C., Søgaard, D. H., Sorrell, B. K., Gradinger, R., and Meiners, K. M. Arctic Sea Ice Ecology. Cham, Switzerland, Springer International Publishing, 2020. [00464] Markus, T., Stroeve, J. C., and J. Miller. Recent changes in Arctic sea ice melt onset, freezeup, and melt season length, Journal of Geophysical Research: Oceans, Vol.114, no. C12024 (Dec., 2009). https://agupubs.onlinelibrary.wiley.com/doi/pdf/10.1029/2009 JC005436 [00465] Marchenko A and Chenot C., Regelation of ice blocks in the water and on the air, Proceedings of the 20th International conference on Port and Ocean Engineering under Arctic Conditions (POAC). Luleå, Sweden, June 9-12, 2009. [00466] McLaren, I. A. The biology of the ringed seal (Phoca hispida Schreber) in the eastern Canadian Arctic. Vol.118. Ottawa (Canada), Fisheries Research Board of Canada, 1958. [00467] McMahon S., Parnell J., Ponicka J., Hole M., and Boyce A. The habitability of vesicles in martian basalt. Astronomy & Geophysics. (Feb.2013) Vol.54, no.1, pp.1-7. https://doi.org/10.1093/astrogeo/ats035 [00468] Mellor, M. “Mechanical Behavior of Sea Ice.” in: The Geophysics of Sea Ice. (New York, Plenum, 1986), pp.165-281. [00469] Melnikov, I. A. The Arctic Sea Ice Ecosystem. Amsterdam, Gordon and Breach. 1997. [00470] Moore, S. E., and Huntington, H. P. Arctic Marine Mammals and Climate Change: Impacts and Resilience. Ecological Applications. (Mar.2008), Vol.18, no. sp2, pp. S157-S165. [00471] Mundy, C.J. and Meiners, K.M., “Ecology of Arctic Sea Ice.” in: Thomas, D.N., Ed. Arctic Ecology. Hoboken, N.J., John Wiley-Blackwell, 2021. [00472] Nelson, C. H. Gray whale and Pacific walrus benthic feeding grounds and sea floor interaction in the Chukchi Sea. Menlo Park, California, US Department of the Interior, Minerals Management Service, Alaska Outer Continental Shelf Region, 1994. [00473] Peacock, E., Derocher, A. E., Lunn, N. J., and Obbard, M. E. “Polar Bear Ecology and Management in Hudson Bay in the Face of Climate Change.” in: Ferguson, S.H. et al., A Little Less Arctic. (Dordrecht, Springer, 2010), pp.93-116. [00474] Perovich, D. K., Richter‐Menge, J. A., Jones, K. F., and Light, B. (2008). Sunlight, water, and ice: Extreme Arctic sea ice melt during the summer of 2007. Geophysical Research Letters, Vol.35, no.11 (Jun.3, 2008). [00475] Perovich et al. Arctic report card: Update for 2019: Arctic ecosystems and communities are increasingly at risk due to continued warming and declining sea ice. NOAA. https://arctic.noaa.gov/Report-Card/Report-Card-2019/ArtMID/ 7916/ArticleID/841/Sea-Ice [00476] Petrich, C., and Eicken, H. “Growth, Structure and Properties.” in: Thomas, D.N. and Dieckmann, G.S., Sea Ice, 2 ^nd Ed. (Wiley-Blackwell, 2010), pp.23-78. [00477] Pittman, S. J. (Ed.). Seascape Ecology. Hoboken, NJ, John Wiley & Sons, 2017. [00478] Pounder, E. R. The Physics of Ice. Oxford, Pergamon Press, 1965. [00479] Sousa, W.P. “Disturbance and Patch Dynamics on Rocky Intertidal Shores.” in: Pickett, S. T. A. and White, P. S. The Ecology of Natural Disturbance and Patch Dynamics. (Orlando, Florida, Academic Press, 1985). [00480] Pritchard, R. S. and Thomas, D. R.1985. Chukchi Sea Ice Motions 1981-82. Flow Industries Inc., Research and Technology Division, Mar.1985. [00481] Ray, G. C., and McCormick-Ray, J. Marine Conservation: Science, Policy, and Management. Malaysia, Wiley Blackwell, 2014. [00482] Robards, M. D. Perspectives on the Dynamic Human-Walrus Relationship.: A Dissertation for the Degree of Doctor of Philosophy. Fairbanks, University of Alaska, 2008. [00483] Rosen, Y. Chukchi Sea walruses are massing on shore earlier than ever. Arctic Today Business Journal. August 5, 2019. https://www.arctictoday.com/chukchi-sea-walruses-are-massing -on-shore-earlier-than-ever/ [00484] Saar, M. O., and Manga, M. Permeability‐porosity relationship in vesicular basalts. Geophysical Research Letters, Vol.26, no.1 (Jan.1, 1999), pp.111-114. [00485] Schwarz, J. and Weeks, Wilford. Engineering Properties of Sea Ice. Journal of Glaciology, Vol.19, no.81 (1977), pp.499-531. [00486] Shirasawa, K., Eicken, H., Tateyama, K., Takatsuka, T., & Kawamura, T. Sea-ice- thickness variability in the Chukchi Sea, spring and summer 2002–2004. Deep Sea Research Part II: Topical Studies in Oceanography, Vol.56, no.17 (Aug.1, 2009), pp.1182-1200. [00487] Smith, T. G., and Stirling, I. The breeding habitat of the ringed seal (Phoca hispida). The birth lair and associated structures. Canadian Journal of Zoology, Vol.53, no.9 (Sep.1, 1975), pp.1297-1305. [00488] Solomon, S., Plattner, G. K., Knutti, R., & Friedlingstein, P. (2009). Irreversible climate change due to carbon dioxide emissions. Proceedings of the national academy of sciences, (Feb.10, 2009) Vol.106, no.6, pp.1704-1709. [00489] Sousa, W.P. “Disturbance and Patch Dynamics on Rocky Intertidal Shores.” in: Pickett, S. T. A. and White, P. S. The Ecology of Natural Disturbance and Patch Dynamics. (Orlando, Florida, Academic Press, 1985). [00490] Stirling I. The biological importance of polynyas in the Canadian Arctic. Arctic, (Jun.1, 1980), Vol.303, no.15. [00491] Stirling, I. Polar Bears. Ann Arbor, University of Michigan Press, 1998. [00492] Stirling, I., and Smith, T. G. Implications of warm temperatures and an unusual rain event for the survival of ringed seals on the coast of southeastern Baffin Island. Arctic, (Mar.2004), pp.59-67. [00493] Stroeve, J. C., Markus, T., Boisvert, L., Miller, J., & Barrett, A. (2014). Changes in Arctic melt season and implications for sea ice loss. Geophysical Research Letters, Vol. 41, no.4, (Feb.22, 2014), pp.1216-1225. https://agupubs.onlinelibrary.wiley.com/doi/pdfdirect/10.100 2/2013GL058951 [00494] Stroeve, J. C., Schroder, D., Tsamados, M., & Feltham, D. Warm winter, thin ice?. The Cryosphere. Vol.12, no.5 (May 30, 2018), pp.1791-1809. [00495] Tatinclaux, J. C. Model tests in ice of a Canadian Coast Guard R-class icebreaker. Special Report 84-6. U.S. Hanover, New Hampshire, Army Corps of Engineers (Apr.1984). https://erdc-library.erdc.dren.mil/jspui/bitstream/11681/118 69/1/SR-84-6.pdf [00496] Tatinclaux, J. C. Ship model testing in level ice: An overview. Report 88-15, U.S. Hanover, New Hampshire, Army Corps of Engineers (Oct.1988). https://erdc-library.erdc.dren.mil/jspui/bitstream/11681/907 3/1/CRREL-88-15.pdf. [00497] Tchesunov, A. V., and Riemann, F. Arctic sea ice nematodes (Monhysteroidea), with descriptions of Cryonema crassum gen. n., sp. n. and C. tenue sp. n. Nematologica, Vol. 41, no.1-4 (Jan.1, 1995), pp.35-50. [00498] Thomas, D. N. Frozen Oceans: The Floating World of Pack Ice. Buffalo, NY, Firefly Books, 2004. [00499] Thomas, D. N., Fogg, G. E., Convey, P., Fritsen, C. H., Gili, J. M., Gradinger, R., ... and Walton, D. W. The Biology of Polar Regions. Oxford, OUP, 2008. [00500] Thomas, D.N., Papadimitriou, S., and Michel, C. “Biogeochemistry of Sea Ice.” in: Thomas, D.N. and Dieckmann, G.S., Sea Ice, 2 ^nd Ed., (Wiley-Blackwell, 2010), pp.425- 467. [00501] Timco, G. W., and Weeks, W. F. A review of the engineering properties of sea ice. Cold Regions Science and Technology, Vol.60, no.2 (2010), pp.107-129. [00502] Timoshenko S., and Woinowsky-Krieger, S. Theory of Plates and Shells. New York, McGraw-Hill, 1959. [00503] Toimil, L. J. Ice-gouged Microrelief on the Floor of the Eastern Chukchi Sea, Alaska: A Reconnaissance Survey. Open-File Report 78-693. US Geological Survey, 1978. https://pubs.usgs.gov/of/1978/0693/report.pdf [00504] Tynan, C. T., Ainley, D. G., and Stirling, I. “Sea ice: A critical Habitat for Polar Marine Mammals and Birds.” in: Thomas, D.N. and Dieckmann, G.S., Sea Ice, 2 ^nd Ed., (Wiley-Blackwell, 2010), pp.395-424. [00505] Udevitz, M.S., Jay, C.V., Taylor, R.L., Fischbach, A.S., Beatty, W.S., Noren, S.R. Forecasting consequences of changing sea ice availability for Pacific walruses. Ecosphere, Vol.8, no.11 (Nov.2017), pp.1-30. [00506] Van Leeuwe, M. A., Tedesco, L., Arrigo, K. R., Assmy, P., Campbell, K., Meiners, K. M., ... and Stefels, J. Microalgal community structure and primary production in Arctic and Antarctic sea ice: A synthesis. Elementa: Science of the Anthropocene, Vol.6 (Jan.1, 2018), pp.1-25. [00507] Wadhams, P. Ice in the Ocean. Gordon and Breach, 2000. [00508] Wang, M., and J.E. Overland. A Sea Ice Free Summer Arctic within 30 Years? Geophysical Research Letters, Vol.36 (2009)L07502, doi: 10.1029/2009GL037820. [00509] Wang, M. and J. E. Overland. A sea ice free summer Arctic within 30 years: An update from CMIP5 models. Geophysical Research Letters, Vol.39 (2012), L18501, doi:10.1029/2012GL052868. [00510] Weeks, W. On Sea Ice. Fairbanks, AK, University of Alaska Press, 2010. [00511] Wiig, Ø., Gjertz, I. and Griffiths, D. Migration of Walruses (Odobenus rosmarus) in the Svalbard and Franz Josef Land Area. Journal of Zoology, Vol.238, no.4 (Apr.1996), pp.769-784. [00512] U.S. Fish and Wildlife Service (USFWS). Conservation Plan for the Pacific Walrus in Alaska. Marine Mammals Management. Anchorage, AK, FWS, (Jun.1994). [00513] Worsley, T. R., & Herman, Y. Episodic Ice-free Arctic Ocean in Pliocene and Pleistocene Time: Calcareous Nannofossil Evidence. Science, Vol.210, no.4467 (1980), pp. 323-325. [00514] Yurkowski, D. J., Brown, T. A., Blanchfield, P. J., and Ferguson, S. H. Atlantic walrus signal latitudinal differences in the long-term decline of sea ice-derived carbon to benthic fauna in the Canadian Arctic. Proceedings of the Royal Society B, Vol.287, no.1940, (Dec.9, 2020), pp.1-10.