Targeting 1: Deep or Under Cover

Paper 22

In “Proceedings of Exploration 17: Sixth Decennial International Conference on Mineral Exploration” edited by V. Tschirhart and M.D. Thomas, 2017, p. 305–318

Sea Floor Mining Exploration Technology and Methods

Kowalczyk, P.[1], Lum, B. [1] _________________________ 1. Ocean Floor Geophysics Inc., Burnaby, BC. Canada

ABSTRACT The world's oceans have always both challenged and helped humanity, providing food for our table, linking settlements around the globe and driving the weather systems that influence our daily lives. Now we are finding mineral and energy resources on the ocean floor that will be as important to mankind as the resources we presently exploit on land. The oil industry has already moved offshore. However, the methods and capital required to exploit offshore oil and gas are not appropriate to find and develop seafloor mineral and gas hydrate deposits. To find and exploit them, exploration methods used on land have been re-invented for the sea. Exploration activities must be economical and the scale of operations must be appropriate. In the past, humans have gone directly down to the seafloor to explore and map it, or have used crude machines to retrieve samples. This is expensive, high risk to human life, and not always successful in deep water. Today, robotic devices have replaced using people on the seafloor, increasing safety and increasing the periods of active operations near or at sea floor. Using robots and remotely operated sensors, we can map, sample, and mine the seafloor. Robotic systems provide economies independent of scale. The size of individual operations can be adjusted to provide a profit without building and operating gargantuan projects. Remotely operated and autonomous underwater vehicles (ROVs/AUVs) are central to this process. Ship borne sonar and multibeam systems map the seafloor at scales of about 1:50,000 or smaller. Sonar, and particularly synthetic aperture sonar, can map the seafloor at scales of 1:100 using AUVs/ROVs. Water chemistry sensors, magnetometers and cameras can be operated on vehicles tasked for other purposes. Sampling can be done using ROVs, and drilling done using robotic seafloor drills. 3D seismic cubes can be acquired using small hydrophone arrays easily deployed from small vessels. Electromagnetic systems and interpretation algorithms exist to map both shallowly buried and seafloor massive sulfide deposits and gas hydrate deposits. Positioning of subsea vehicles and installations is not simple, but off the shelf transponder systems are available, and subsea sonar based navigation systems allow multiple assets to be located within a transducer array. These systems are becoming smaller, less expensive, and deployable from general purpose vessels rather than specialized purpose built ships. The principal economic targets today are seafloor deposits of gas hydrates, submarine massive sulfides, and polymetallic nodules. The geologic signature of these deposits is understood. Best practices to find and exploit them are well known or are being developed. The impact of robotics is that “bigger is not necessarily better” and that exploration can be done cost effectively and safely using smaller, reliable, more capable, and less expensive equipment. Mining at sea, long delayed, is about to become an established industry.

MAJOR ECONOMIC TARGETS The major economic targets for seafloor exploration today are submarine massive sulfides (SMS), manganese nodules and manganese crusts, and gas hydrates (Figure 1). Also, significant potential exists for phosphate exploitation and diamond mining is ongoing off the west coast of Africa. The drivers for mineral exploration are twofold: the potential for highly profitable future mining operations as the technology develops to exploit known seafloor deposits, and a strategic desire by a number of nations to establish secure and stable supplies of energy and metals to support their national industrial base. Japan, in particular, has stated that it is a national priority to develop the technology to exploit the seafloor resources within their Exclusive Economic Zone (EEZ) (Figure 2). The country has a strong industrial base, but presently imports raw materials and oil and gas. Their objectives are to enable trial mining of SMS deposits in the Japanese EEZ this year, to enable

commercial exploitation of gas hydrate deposits in 2018, and to initiate exploration and build the technology for the exploitation of manganese nodules and crusts to supply manganese, cobalt, nickel, and rare earths to Japanese industry with a target date of 2025.

DEPOSIT TYPES AND EXPLORATION TECHNOLOGY

Submarine Massive Sulfides There is considerable crossover between methods applied for different deposit types. There is generally an association between methods and type of target, but usually the methodology is applicable across the full range of deposit types.

306 Targeting 1: Deep or Under Cover

Figure 1: a) Massive sulfide chimneys, Solwara 1, PNG; b) Sample box showing manganese nodules; c) Gas hydrate mound, Gulf of Mexico.

Figure 2: Potential seafloor resources within the Japanese EEZ that provide motivation for the development of seafloor exploration mining technology within Japan. SMS are a new deposit type first found on the East Pacific Rise in 1979 (Speiss et al. 1980). Since then they have been found around the world, with new deposits being discovered on a regular basis. They are attractive economic targets as they have high grades, with higher gold and copper grades located in deposits along the trend of the famous ring of fire of the Western Pacific, extending from Japan to New Zealand. The economics of mining these deposits will be quite different from that of a deposit on land. The capital cost, and risk profile of a mining investment is very different. The mining system, once built, can move and hence amortized over many deposits. The sequence of mining is highly flexible. Grades are high, and there is no stripping required. The infrastructure is modular and is built in a shipyard, and the mining operation is robotic and thus amenable

to a staged implementation using standard mining modules rather than requiring an upfront commitment to achieve economies of scale. Exploration for SMS deposits has been ongoing seriously since 2006. Ocean Floor Geophysics (OFG) has participated on more than 20 mineral exploration cruises of up to 3 months duration since 2007. Other ships and exploration teams have been at sea every year. They have been funded privately, by governments, and by parastatal organizations. This continuing work is motivated by a desire to start mining. It is not academically motivated. Figure 3 illustrates the distribution of known vent fields documented in the InterRidge Vents Database (Beaulieu, 2015). The highest grades of copper and gold presently have

Kowalczyk, P., and Lum, B. Sea Floor Mining Exploration Technology and Methods 307

been found in vent fields within the back arc environment of the Western Pacific. Most of these deposits lie within the EEZ of coastal nations. Hence the legal environment relating to tenements is more favourable for private companies, and most privately funded exploration has occurred in this area. Deposits in the central Indian Ocean, and along the Mid Atlantic Ridge are administered by the United Nations International Seabed Authority (ISA) and are being explored by national entities, for example, the EU, Russia, China, India and Japan within tenements issued by the ISA. Regional exploration for new vent fields is done using ship borne bathymetry, looking for favourable seafloor morphology, i.e. spreading arc centers, calderas, and seafloor volcanic features. This is followed by transects using towed sonars, and

water chemistry and water sampling systems such as the tow-yo water sampling system (Figure 4). In a tow-yo cast, continuous pH, Oxidation-Reduction Potential (ORP), turbidity and a discrete series of water sample measurements are made towing an instrument sled and simultaneously winching it up and down to search for hydrothermal plumes in the water column. Anomalous values provide a spatial indication of nearby hydrothermal venting. Turbidity and ORP anomalies are highly correlated with active hydrothermal venting (de Ronde, 2001), but are not necessarily indicators of economic mineralization. Also, detection of regional magnetic anomalies that disrupt the normal linear patterns of magnetization may indicate favourable areas that localize hydrothermal activity. This can be done using towed ship-based magnetometers (Barkhausen et al. 2016).

Figure 3: InterRidge Vents Database - location of hydrothermal vents and vent fields. The back arc spreading centre deposits of the Western Pacific are generally associated with high grades of copper and gold. Map after Stace Beaulieu, Woods Hole Oceanographic Institution.

Figure 4: (Left) Typical water chemistry anomaly passing through a hydrothermal plume. Note that methane and manganese measured in water samples are enhanced in the plume, while optical transmissivity falls. The optical transmissivity is a continuous measurement. (Right) Tow-yo image of turbidity combined with multibeam sonar backscatter in 3D visualization (picture courtesy Nautilus Minerals, rights reserved).

308 Targeting 1: Deep or Under Cover

Sonar Seafloor morphology indicated by sonar is used to localize SMS deposits once a target region has been identified using regional methods. Many different sonar systems have been deployed. Side-scan sonar has a longer range than multibeam, but multibeam provides bathymetric information while side-scan sonar generally provides backscatter only. Often a multibeam sonar will be used to map the bathymetry below the ROV or AUV, while side-scan will be deployed concurrently to map a swath on either side of the vehicle. Interferometric side-scan systems provide additional information as they can provide bathymetric information as well as backscatter in the image swath, widening the effective width of investigation. The gold standard is High Resolution Synthetic Aperture Sonar (HISAS), which uses synthetic aperture processing that combines multiple pings along the vehicle track into a synthetic ping with a very narrow beam width, to produce very high resolution sidescan imagery (Table 1) with constant resolution across the full swath width. In the Hugin AUV implementation (Figure 5), two receiver arrays on each side of the vehicle also produce very high resolution interferometric bathymetry data (Hansen, 2011). Figure 6 illustrates a SAS image collected using HISAS on a Hugin AUV. The mosaic was made by combining imagery from several AUV passes. It is suitable for geological interpretation and can be viewed in 3D when combined with high resolution bathymetric data collected concurrently. Denny et al. (2015) describes many typical morphological anomalies that can be used to aid the recognition of hydrothermal mineralization using high resolution AUV backscatter and bathymetric imagery. Figure 7 shows sonar imagery from the Gondou SMS field discovered by the Japanese Coast Guard in June 2015. This may be the largest SMS field discovered to date in the Japanese EEZ. The field was mapped in two days during two dives by the Japan Coast Guard AUV Gondou (Minami, 2015). The single chimney highlighted has a height of 23 m above the seafloor, a diameter of about 8 m, and a mound at the seafloor about 20 m in diameter. A very rough tonnage estimate for this chimney is about 7500 tonnes, which would provide a week's feed for a 1000 tonne/day mining and processing system.

Figure 5: Hugin AUV, capable of 3000 m depth of operation and equipped with a suite of sensors including HISAS.

Figure 6: AUV SAS image of two hydrothermal vents. Image resolution across the swath has been reduced to 4 cm x 4 cm from 2 cm x 2 cm. The vents are visible as mounds of hydrothermal debris and the "flares" of hot water distorting the acoustic image (Denny et al. 2015). The associated bathymetric map has been generated at 9 cm x 9 cm resolution (not shown).

Figure 7: Ship borne and AUV sonar imagery from the Gondou SMS discovery (Minami, 2015). First image (left) shows the ship borne Multibeam Echo Sounder (MBES) bathymetry image. Second image shows the sonar imagery after two AUV dives; the SMS chimney fields are clearly visible. Third image is a zoomed image. Fourth image (right) shows the multibeam imagery collapsed together for the small area indicated. Grey pixels are AUV MBES points from an AUV pass on the left, purple pixels are AUV MBES points from an AUV pass on the right. Together they define the shape of the individual chimney.

Kowalczyk, P., and Lum, B. Sea Floor Mining Exploration Technology and Methods 309

Table 1: Hugin HISAS sonar specifications. Synthetic Aperture Sonar Interferometric Bathymetry

Center frequency 100 khz Cross track resolution 10 cm Wavelength 1.5 cm Along track resolution 10 cm Bandwidth 30 kHz Relative bathymetric resolution 2 cm Total frequency range 50-120 kHz Along track resolution 2 cm Cross track resolution 2 cm Maximum Range @ 2m/sec 200 m Area coverage rate 2 km2/hr

Magnetometry

Magnetometry is a useful method of mapping geology in the SMS environment (Figure 8), as it is on land. However, there are challenges not faced on land. Doing a dedicated magnetic survey is not cost effective; data acquisition is efficiently done during dives purposed to multiple objectives. ROVs and AUVs are not good magnetometer platforms, being constantly modified and having significant and varying self magnetizations that interfere with the measurement of the ambient field. There is no place to put a base station on a ship – it is a large magnetic object. Also, the benefit of doing magnetic tie lines is generally deemed not worthwhile as the time taken to do this can be used to acquire other data. The cost of putting a geophysicist on a ship to manage the magnetic data acquisition is seen to be significant. Finally, the mounting of a magnetometer will generally be done by technicians and equipment mechanics not experienced in the art of magnetic data acquisition. Even so, it is possible to acquire very useful magnetic data that can provide geologic information of considerable value. Magnetic anomalies in the SMS environment are often large enough that compensated data acquired without diurnal corrections and tie lines can be used effectively. Interpretation methods taken from land based and airborne magnetic surveying in the mineral industry can be applied in the marine environment, particularly magnetic inversion methods. This takes the use of magnetic maps well

beyond the stage of looking at images to a quantitative analysis method that maps geological contacts, faults, magnetite rich zones and alteration envelopes associated with SMS mineralization. Initial compensation methods for ROV and AUV deployment were practical but crude, depending upon the ROV or AUV maintaining fixed headings and then building a table of adjustments to correct data combined to a common datum. Automatic magnetic compensation methods have now been developed that allow magnetometers to be mounted onto an ROV or an AUV of opportunity, and to collect useful data (Bloomer et al. 2014). Figure 9 shows uncompensated and compensated data collected using an AUV over an SMS prospect. The AUV was tasked with mapping the prospect with multibeam and sidescan sonar, and collected water chemistry measurements and magnetic measurements concurrently. The ability to collect compensated magnetic readings simply without impacting the crew size supporting a ROV or an AUV, or requiring specialized staff on the ship makes the routine collection of good magnetic data practical, and adds considerable value to an ROV or AUV dive purposed to other tasks. Some care still needs to be taken when mounting a magnetometer; this is a matter of good practice.

Figure 8: (Left) Two fluxgate magnetometers mounted on a work class ROV during a 2008 SMS survey deployment. The ROV was tasked with doing EM, video acquisition, and sampling. The fluxgates were 3-component +/- 2.0nT instruments. (Right) A magnetic map acquired this system.

310 Targeting 1: Deep or Under Cover

Figure 9: (Left) Uncompensated magnetic data from a magnetometer on an AUV. The data contains about 2400 nT of heading error. (Right) The same data after compensation (Bloomer et al. 2014). The heading errors have been reduced to about 10 nT and anomalies that can be usefully interpreted are evident in the data.

Figure 10: Magnetic inversion results of AUV magnetic data from the Solwara 1 deposit of Nautilus Minerals. Bathymetry is shown as white contours and AUV observation points as purple spheres. The green volumes are zones of reduced magnetization inferred from a 3D inversion of the magnetic data. Figure 10 is a visualization of a magnetic inversion of data from the Autonomous Benthic Explorer (ABE) AUV (Tivey et al. 2006) collected over the Solwara 1 deposit of Nautilus Minerals. The magnetic data was inverted using the MAG3D code developed at the Geophysical Inversion Facility at the University of British Columbia (Li and Oldenburg, 1996). The geological inference is that the regions of reduced magnetization are zones of alteration caused by the destruction of magnetite in the feeder zones of the mineralized mounds on the seafloor. A general relationship between the size of the alteration low and the size of mound can be expected as the size of the alteration zone is a proxy for the amount of hydrothermal fluid flow. This allows one to rank targets based on the volume of the magnetization low, at least in the geological case where such a low is observed. Mapping these alteration lows is a useful tool to locate blind mineralization shallowly buried by recent volcanic flows close to exposed hydrothermal mounds on the seabed.

Electrical and Electromagnetic Methods SMS deposits are conductive and present good targets for electromagnetic (EM) prospecting (Figure 11). The principal issues for the use of EM are the practical issues of deploying an EM prospecting system close to the sea bottom in the deep ocean, surrounded by a conductive medium. The volcanic terrain is typically quite rugged, the seafloor is a jagged and sharp so not easy to lay equipment on or to drag equipment across. As well, the mounting of an EM system on various ROV or AUV platforms requires a purpose-built system for each platform which is typically different for each mineral exploration campaign. Free swimming AUVs do not have much power available, and the available system geometries for an AUV system are limited as the attachment of anything to the exterior an AUV can impede the control of the AUV and will slow it down. ROVs have power and good communications to the surface vessel, generally through fibre. However, operating ROVs close to the seafloor in a survey geometry is challenging, and the ROV itself generates significant amounts of EM noise.

Figure 11: An electrical model of the TAG SMS deposit after Yu and Edwards (1996). Note the mound itself is conductive, surrounded by conductive seawater. Beneath the mound a feeder zone exists – this has intermediate conductivities, while the surrounding rocks, typically volcanics, are resistive in comparison to the sulfides and the ocean. These problems have been overcome by some systems that have been successfully deployed and show considerable promise. The first system to be successfully deployed is the Ocean Floor Geophysics (OFG) EM system (Kowalczyk, 2008). It was used at the Solwara 1 deposit of Nautilus minerals and mapped out the limits of the conductive mineralization. The bounds of the deposit mapped by EM were used in the NR43-101 securities report describing the deposit (Lipton, 2012). To the author's knowledge, this is the first and perhaps the only case where EM has been used as an integral part of defining an ore reserve. Towed time domain systems have since been developed such as the Waseda University system (Nakayama and Saito, 2016) (Figure 12) and the Golden Eye system (Muller et al. 2016) which can operate both in frequency domain and time domain. These systems are coincident loop systems that can either be towed by an ROV just above the seafloor, or can be landed on the seafloor to do EM soundings. The Waseda University system has been deployed during several campaigns mapping SMS mineralization in the Japanese EEZ and has successfully

Kowalczyk, P., and Lum, B. Sea Floor Mining Exploration Technology and Methods 311

detected known mineralization buried at a depth of 30 m (Figure 13).

Figure 12: Waseda University TEM system. This system can be landed on the seafloor, or towed about 1.5 m above the seafloor (photo Professor A. Saito, Waseda University). Controlled source electromagnetic (CSEM) using a towed EM electric dipole source and E-field receivers has recently been shown to be an excellent method to detect SMS mineralization. The method was developed for gas hydrate exploration, but is applicable to SMS exploration. Using a deep tow CSEM array very clear anomalies have been measured (Murton, 2017) over

known mineralization at the TAG field on the mid-Atlantic ridge. The disadvantage of a deep tow method is that it is difficult to navigate the array at the end of a long tow cable, it is difficult to maintain a low altitude over the rugged topography, the array can only be towed at slow speeds, and much of the survey time is spent in turns if multiple closely spaced lines are flown over a deposit to define its geometry. The use of fixed seafloor transmitters and a roving E-field sensor mounted on an AUV overcomes these problems. Ocean Floor Geophysics, working with Steven Constable at Scripps and Fukada Salvage and Marine Works has completed a CSEM-AUV survey over the Iheya deposit in the Japan EEZ (Figure 14 to Figure 16). Strong and interpretable EM anomalies were detected over the mineralized zones. Interestingly, mounting the E-field sensors on the AUV allows a highly sensitive self potential measurement to be made, something that is difficult to do using an ROV. Use of an AUV allows very accurate navigation, good altitude control, and tightly spaced traverse lines to be achieved efficiently. It is also possible to use all the other sensors on the AUV concurrently: sidescan and multibeam sonars, sub-bottom profiler, and water chemistry sensors. The result is a very cost effective survey. This system uses robust well tested equipment and provides significant cost savings over other EM systems designed for SMS exploration.

c Figure 13: TEM anomaly over the Izena SMS deposit, Okinawa trough, Japan, measured with the Waseda TEM system. Blue are early time windows, purple late time. The anomaly is clearly visible in the early and mid time windows. The left-hand portion of the anomaly is the response traversing over buried mineralization of up to 30 m depth, the right-hand portion is the response crossing the sulfide mound. Data is raw, unstacked data (A. Saito, pers. comm.).

Figure 14: AUV CSEM survey, Sea of Japan Nov 2016. (Left) AUV with electric dipole mounts on AUV. (Right) Scripps DUESI seafloor transmitter being deployed.

312 Targeting 1: Deep or Under Cover

Figure 15: Bathymetry and apparent resistivity collected during an AUV-CSEM survey. Grid lines are every 500 m. Water depth is about 1500 m. Note the strong conductivity anomalies associated with the general zone of mineralized mounds. The multi-transmitter, multi-frequency EM survey is amenable to 3D inversion to locate the burial depth and limits of the conductive zones.

Figure 16: Self Potential (SP) anomalies detected over the hydrothermal system shown in Figure 15 (Constable, 2017). Multiple passes of the AUV are plotted. As the SP voltage was collected using 3 components it is possible to draw quiver plots showing the direction of the SP currents. Note the well organized and repeatable directions of the voltage gradients pointing away from an SP source. Hydrothermal flow is hypothesised as the source of this anomaly.

Gravity Gravity responds to density changes, so is suited to mapping SMS deposits. Like EM, the difficulty has been the practical problems of making a gravity measurement on the seafloor. Evans (1996) made a gravity traverse across the TAG SMS mound that unambiguously indicated a gravity anomaly due to the massive sulfides, and for which it was possible to compute an excess mass (Figure 17) on the seafloor. The gravity meter was in the submersible with the operator. This proved the utility of the gravity measurements, but is not a practical method for routine surveys. There has been work done using ROV deployed gravimeters (Sasagawa et al. 2003) and

attempts to measure gravity using free swimming AUVs. ROV deployed gravimeters have excellent sensitivity, but low productivity and hence high costs for a seafloor minerals exploration application. Figure 18 shows a recent advancement, a self levelling ocean bottom gravimeter has been mated to a small "hopping" AUV that allows seafloor gravity to be collected from a ship primarily tasked to other purposes (A. Oshida, pers. comm.). The accuracy and ease of use of this system allows gravity surveys to be done readily to assist resource tonnage estimation during the assessment of an SMS deposit prospect.

Kowalczyk, P., and Lum, B. Sea Floor Mining Exploration Technology and Methods 313

Figure 17: (Left) Gravity traverse over the TAG SMS mound (Evans, 1996). (Right) The anomaly due to the massive sulfides is easily recognizable.

Figure 18: Seabed landing AUV with self levelling ocean bottom gravimeter (OBG). Photo A. Oshida/ KGE.

Seismic Seismic methods have high resolution and are the only method that can reliably map the top and bottom of an SMS deposit. However, it is difficult to acquire useful seismic data over an SMS deposit using normal high resolution marine seismic methods. The problem is the large size of the Fresnel zone, the small size of a deposit and the strong diffraction effects caused by the rough seafloor and often dipping and faulted volcanic stratigraphy. Also, the mobilization costs of bringing a seismic ship to survey a small area at high resolution is relatively high. These issues have been overcome by the Vertical Cable Seismic

method (VCS) (Asakawa, 2015). This method uses vertical receiver arrays anchored on the seafloor over the survey area, and a traversing source over a grid pattern on the surface. The receivers are close to the seafloor so a 3D data cube can be developed with high resolution. The system is not large, is readily shipped and can be deployed from a ship of opportunity. Again, like gravity, the method reduces the amount of drilling required to establish an ore reserve and provides detailed information about the limits and thickness of both seabed and shallowly buried deposits (Figure 19).

314 Targeting 1: Deep or Under Cover

Figure 19: 3D seismic section through the Izena Deposit, Japan. The data was acquired using a vertical cable seismic array and a surface source. The left hand edge of the section starts at the edge of a seafloor mound, the buried mineralization between VC2 and VC3 has been confirmed by drilling (Asakawa, 2015).

Figure 20: (Left) Multiscale imagery from the Iheya North field, Japan (Thornton et al. 2016). Note that imagery can be retrieved to show the entire prospect, or zoomed in to display a 40 cm image area with a crab in it. The yellow arrows indicate nested images at different scales. (Right) Mosaiced seafloor image built after automatic position adjustment of AUV track lines using image correlation (Woolsey and Woolsey, 2016).

Video Imagery Multiple video imagery streams are routinely acquired during ROV operation to aid piloting, and an AUV operating near the seafloor can also acquire seafloor imagery. Considerable long term value can be created from this data if it is indexed in a project database that allows interactive image review at different scales. It becomes more valuable as more data is acquired and can be used to plan subsequent work, to do environmental baseline work, and for change detection (Figure 20). The principal issue is the use of appropriate software to allow rapid access at different scales and that will remain scalable over time as the image database becomes very large. It is critical to be able

to reference images by both time and position, and at multiple scales. Also, notes attached to the images during acquisition need to be retained and accessible, for example noting the identification of fauna or flora that are of import to an environmental assessment.

Manganese Nodules Manganese nodules occur widely throughout the world on the ocean abyssal plains. As well, widespread occurrences of manganese crusts occur on the flanks of seamounts and other occurrences of volcanic rock such as ridges and plateaus (ISA, 2017). Nodules particularly are of economic interest both because of their large tonnages, and because they contain

Kowalczyk, P., and Lum, B. Sea Floor Mining Exploration Technology and Methods 315

significant credits of Ni, Co, and rare earths. At present, most exploration attention is being directed to the Clarion-Clipperton Fracture zone in the Eastern Pacific where numerous mineral tenements have been established under the authority of the United Nations ISA (Figure 21) to the present exploration, as well as the value of the contained Ni and Co, which may be greater than the manganese values. Rare earths are a strategic resource and the world supply is presently controlled by China. Outside China, the establishment of a seafloor mining capability may contribute to world security of supply. In the 1970s, three United States based consortia conducted mining tests in the Pacific using hydraulic mining systems. These efforts stopped when commodity prices fell, but they demonstrated that the nodules could be lifted to surface using hydraulic risers and that the nodules constituted a potential economic resource. Present exploration has advanced to the stage of doing engineering feasibility tests. Many national entities and at least one private company are actively doing nodule exploration. The principal objective of these programs is to establish the concentration of nodules across large areas prior to starting mining operations. Ship borne MBES and backscatter imagery is being used regionally to identify areas of interest. Sonar is useful mapping nodule concentration, but in detail it is difficult to link grades from the physical samples collected with box-cores to sea-floor photography, and then to sonar responses across large areas. AUV seafloor photography and high resolution MBES bathymetry are the principal tools used at the seafloor to extend estimates of ore tenor away from the locations of physical samples. Detailed bathymetry is being acquired

using AUV systems to map out areas suitable for the operation of seafloor collector systems, which need to be designed to traverse across the expected topography. In deep water, navigation and linking of seafloor sampling and AUV photography and sonar imagery generally requires the installation of seafloor transponders to provide accurate positioning.

Gas Hydrates It is perhaps debateable whether gas hydrates should be considered a mineral. They are a hydrocarbon resource, but are quite different from conventional oil and gas and different methods used to explore for them that have more in common with mineral exploration than traditional oil and gas exploration. Deposits of gas hydrate comprise frozen methane in the interstitial pore space of sea floor sediments. They occur worldwide. Their stability zone extends from the seafloor to some 200–300 m below the seafloor. In the arctic, they can be found close to surface, throughout the world they generally occur below at water depths below 300 m. They are important because there are huge amounts of methane existing as gas hydrates. Present estimates are that there are undiscovered conventional natural gas resources on the order of 104 Trillion Cubic Feet (TCF). The global gas hydrate resource is estimated at 105 to 108 TCF, that is 10 to 10,000 times greater (Ruppel 2011). Only a small portion of this will be recoverable but clearly it is an important resource, particularly for nations like Korea and Japan, and for emerging economies such as India that do not have substantial developed conventional resources.

Figure 21: Map of mineral tenements 2014 from the ISA. Green blocks surrounding the tenements are ecological reserves established around potential mining areas.

316 Targeting 1: Deep or Under Cover

Figure 22: Two styles of gas hydrate occurrence. Left shows a seismic section from the Nankai Trough, Japan (Fuji et al. 2015). The gas hydrate target is in sandy sediments at the bottom of the gas hydrate stability zone. Right shows a seafloor gas hydrate mound occurring close to a methane vent and occurring close to the seabed. Historically, economic gas hydrate targets have been targeted at the base of the gas hydrate stability field (Figure 22 – left image). These would be exploited by drilling, and warming of the gas hydrate in situ or by CO2 replacement. Japan has demonstrated production from this type of occurrence in the Nankai Trough using the drill ship Chikyu. Recently it has been shown that seafloor gas hydrate mounds (Figure 22 – right image) can be a part of larger bodies that contain large concentrations of gas hydrates immediately below the seafloor. These have been mapped effectively using the Scripps towed VULCAN CSEM method (Constable et al. 2016). Exploration for these near seafloor deposits comprises regional exploration for mounds, seafloor pockmarks, and methane venting and bubble streams using full water-depth multibeam imaging. A detailed seafloor investigation follows using AUV bathymetry mapping and the acquisition of sub-bottom profiler data. Favourable areas can then be mapped using the Vulcan CSEM. Resistive anomalies have been interpreted using the MARE2D EM code (Key, 2012) (Figure 23), or in 3D using the E3D code (Oldenburg et al. 2005). The successful use of these methods has been emphasized by a press release from METI, the Japanese

ministry funding gas hydrate exploration technology research, stating that resistivity was the best method after drilling to estimate near sea bottom gas hydrate resources (METI, 2016).

SUMMARY Geophysical methods have become key to the cost-effective exploration for minerals on the seafloor. Best practices and robust methods have been tested and are now being applied routinely. Geophysical systems can be deployed on AUV and ROV platforms. AUVs are becoming the main vehicle for geophysical surveys as they are more cost effective to deploy than ROVs, and can survey using multiple systems at the same time. They can be deployed from smaller ships than ROVs. However, for sampling in SMS terrains, and for the support of seafloor drills, ROVs remain critical. For some applications, towed systems are still effective and regularly used, e.g. tow-yo, towed sonar, and CSEM systems. However, these require larger surveys to be cost effective as lines cannot be turned around quickly, and significant support equipment and crew are required on the ship.

Figure 23: 2D inversion results of a VULCAN CSEM survey over a methane vent. The resistor is interpreted to be a gas hydrate body near the seafloor and close to the vent (Constable et al. 2016). The seawater is red in the image, and the VULCAN E-field readings were taken at the white points along the section. The blue resistive body in the result comes to the seafloor at the location of the known methane vent.

Kowalczyk, P., and Lum, B. Sea Floor Mining Exploration Technology and Methods 317

AUVs can navigate accurately in deep water and with closely spaced lines. Synthetic aperture sonar, multibeam sonar, and interferometric bathymetry provide excellent images. AUVs can also collect magnetic, CSEM, water chemistry, seafloor imagery, and sub-bottom profiler data concurrently with the operation of high resolution sonar systems. For manganese nodules, AUVs can be deployed to map nodule distribution on the sea bed. An advantage of AUV operations is that physical sampling such as box coring can be done from the AUV mother ship while the AUV survey is in progress ROVs can be utilized to acquire video, EM and magnetic data, and do seafloor sampling. These systems are now robust, tested systems than can be effectively incorporated into a mineral exploration program. Magnetic and EM methods can explore around known deposits for shallow buried extensions or sister deposits. Localized, low logistics gravity and seismic methods can delineate the tonnages and dimensions of deposits once located. These systems combine with new technologies in seafloor drilling to reduce the amount of drilling required to establish the grade and tonnages of deposits, and will allow a mining operator to rank their prospects and to plan their exploitation cost effectively.

REFERENCES Asakawa, E., 2015, Advances in Vertical Cable Seismic (VCS) for Seafloor Massive Sulfide exploration: AGU Fall Meeting, Abstract NS23A-1930. Barkhausen, U., L. Heyde, C. Rühlemann, U. Schwarz-Schampera, B. Schreckenberger, and K. Schwalenberg, 2016, Geophysical Exploration Tools for Deep Sea Mineral Resources: EAGE workshop on Deep Mineral Exploration. Beaulieu, S.E., 2015, InterRidge Global Database of Active Submarine Hydrothermal Vent Fields: prepared for InterRidge, Version 3.3, http://vents-data.interridge.org, accessed 2 June 2017. Bloomer, S., P. Kowalczyk, J. Williams, T. Wass, and K. Enmoto, 2014, Compensation of magnetic data for autonomous underwater vehicle mapping surveys, Autonomous Underwater Vehicles (AUV): Presented at IEEE/OES Autonomous Underwater Vehicles. Constable, S., P.K. Kannberg, and K. Weitemeyer, 2016, Vulcan: A Deep-Towed CSEM Receiver: Geochemistry, Geophysics, Geosystems, 17(3), 1042–1064.. Constable S., P. Kowalczyk, and S. Bloomer, 2017, Self potential and EM Studies using an Autonomous Underwater Vehicle, in preparation. de Ronde, C.E., E.T. Baker, G.J. Massoth, J.E. Lupton, I.C. Wright, R.A. Feely, and R.R. Greene, 2001, Intra-oceanic subduction-related hydrothermal venting, Kermadec volcanic arc, New Zealand: Earth and Planetary Science Letters, 193 (3), 359-369.

Denny, A.R., T.O. Sæbø, R.E. Hansen, and R.B. Pedersen, 2015, The use of synthetic aperture sonar to survey seafloor massive sulfide deposits: The Journal of Ocean Technology, 10 (1), 48-65. Evans, R.L, 1996, A seafloor gravity profile across the TAG mound: Geophysical Research Letters, 23 (23), 3447-3450 Fuji, T., K. Suzuki, T. Takayama, M. Tamaki, Y. Komatsu, Y. Konno, J. Yoneda, K. Yamamoto and J. Nagao, 2015, Geological setting and characterization of a methane hydrate reservoir distributed at the first offshore production test site on the Daini-Atsumi Knoll in the eastern Nankai Trough, Japan: Marine and Petroleum Geology, 66, 310-322 Hansen, R. E., 2011, Introduction to synthetic aperture sonar: Sonar Systems: InTech. International Seabed Authority (ISA), 2017, Deep seabed mineral resource, https://www.isa.org.jm/mineral-resources/55, accessed 10 July 2017. Minami, H., 2015, Discovery of the largest chimney field in Japanese E.E.Z. by AUV Gondou, http://www.ise.bc.ca/pdfs/JCG%20Chimney%20Field%20Discovery%20Paper.pdf , accessed 04 June 2017. Key, K., 2012, Marine EM inversion using unstructured grids: a 2D parallel adaptive finite element algorithm: 82nd Annual International Meeting, SEG, Expanded Abstracts, 1–5. Kowalczyk, P., 2008, Geophysical prelude to first exploitation of submarine massive sulphides: First Break, 26 (11). Li, Y., and D.W. Oldenburg, 1996, 3-D inversion of magnetic data: Geophysics, 61 (2), 394-408. Lipton, I., 2012, Mineral Resource Estimate, Solwara Project, Bismarck Sea, PNG: Technical report compiled under NI43-101. METI, 2016, ANRE Estimated the Amount of Shallow Methane Hydrate Resources and Verified the Estimation Results, http://www.meti.go.jp/english/press/2016/0916_03.html, accessed 2 June 2017. Müller, H., K. Schwalenberg, and T. von Dobeneck, 2016, Challenges in the deep sea: The Golden Eye electromagnetic induction profiler, Geophysical Exploration Tools for Deep Sea Mineral Resources: EAGE workshop on Deep Mineral Exploration. Murton, B., 2017, Cruise Report JC138: National Oceanography Centre, Southampton UK. Nakayama, K., and A. Saito, 2016, The Seabed TDEM Towed by ROV for the Ocean Bottom Hydrothermal Deposits: EAGE First Conference on Geophysics for Mineral Exploration and Mining - Near Surface Geoscience.

318 Targeting 1: Deep or Under Cover

Oldenburg, D.W., R. Eso, S. Napier, and E. Haber, 2005, Controlled source electromagnetic inversion for resource exploration: First Break, 23, 41-48. Ruppel, C., 2011, Methane hydrates and the future of natural gas: MITEI natural gas Report, supplementary paper on Methane Hydrates, 4, 25. Sasagawa, G.S., W. Crawford, O. Eiken, S. Nooner, T. Stenvold, and M.A. Zumberge, 2003, A new sea‐floor gravimeter: Geophysics, 68(2), 544-553. Speiss, F.N., K.C. Macdonald, T. Atwater, R. Bal, J. Francheteau, J. Guerrero, J. Hawkins, R. Hayi, R. Hessler, M.K. Juteau and B.L. Larson, B.L., 1980, East pacific rise: hot springs and geophysical experiments, Science, 207 (4438), 1421-1433. Thornton, B., A. Bodenmann, O. Pizarro, S.B. Williams, A. Friedman, R. Nakajima, K. Takai, K. Motoki, T. Watsuji, H. Hirayama, Y. Matsui, H. Watanabe, and T. Ura, 2016, Biometric assessment of deep-sea vent megabenthic communities using multi-resolution 3D image reconstructions: Deep Sea Research Part I: Oceanographic Research Papers, 116, 200–219. Tivey, M., W. Bach, J. Seewald, M.K. Tivey, and D. Vanko, 2006, Manus 2006: hydrothermal systems in the Eastern Manus Basin: fluid chemistry and magnetic structure as guides to subseafloor processes: Cruise Report R/V Melville MAGELLAN-06. Woolsey, M., and A. Woolsey, 2016, Using MB-System and Matlab to Generate Geographical Mosaics from Seafloor Images: OCEANS 2016 MTS/IEEE Monterey, 1-10. Yu, L., and R.N. Edwards, 1996, Imaging axi-symetric TAG-like structures by transient electric dipole seafloor electromagnetics: Geophysical Research Letters, 23, 3459-3462.