A boom lift is not a typical tool geophysicists use when scouting landscapes for mineral deposits, but it is something Ronald Bell needed in the summer of 2019 to allow him to perform a magnetic survey of a four-square mile patch of Canadian Shield south of the Eagle mine in the upper peninsula of Michigan – a survey that he conducted using a drone.
The terrain was treed, and the pilot was required to keep his unmanned aerial vehicle (UAV) in his view at all times. That had Bell, a senior geophysicist and geoDRONEologist at Denver-based International Geophysical Services, towing a rented lift behind his pickup truck to the survey site.
“I lifted the pilot above the canopy, so that he could maintain his line of sight during the data acquisition process.” Harnessed and buckled in, the pilot settled into his task above the tree tops. “It was just another day at work,” said Bell.
The area Bell had surveyed had been identified by an aircraft survey in the 1950s to have a magnetic anomaly, which was suspected to be a copper-nickel deposit. Due to the height that aircraft flew, the precise location of the anomaly could not be easily determined. The drone, however, could better than halve that distance, cruising over the treetops between 25 and 30 metres from the ground.
“We’re much more sensitive to variations in the magnetic field because the variations are bigger when we’re flying the sensor lower,” he said.
Proximity to the ground, easier access to hard-to-traverse territory and the relatively small expense as compared to traditional aircraft are just some of the reasons drone-mounted surveys are becoming more and more popular as exploration tools – even if there are still challenges and technical hurdles to overcome that require the type of creative thinking that leads to renting booms and to scientific advancements that shrink surveying equipment down to drone size.
Why drones?
Drones today range from the seven-tonne fixed-wing Global Hawk flown remotely by the United States Air Force on spying and bombing missions half a world away to the multi-rotor 500-gram toy your kids fly around the house. Geophysicists like something in between. Typically suitable for them are multi-rotor industrial drones designed for commercial applications that can carry a payload of up to five kilograms and stay in the air for up to 30 minutes before needing fresh batteries. A five kilogram limit permits their use with sensors like magnetometers, ground-penetrating radar (GPR), electromagnetic interference (EMI) conductivity meters, and LiDAR but (for now) eliminates others such as gravimeters.
It is not hard to imagine why geophysicists are turning to these drone platforms. UAVs can perform endless repetitive tasks without getting bored or tired, go places humans fear to tread, stay on track more precisely than a human with a hand-held GPS can, and do it faster and cheaper than a human can.
They are also becoming cheaper and more wide-ranging, with more models available from more vendors. The Business Research Company calculates that worldwide, the commercial drone market was worth US$3.45 billion in 2018 and, even with a COVID-19 related sales dip in 2020, it should reach US$7.13 billion in 2022.
Just as there are options available for drone types, there are options for the type of survey equipment (referred to collectively as sensors) they can carry. Typically these sensors weigh under five kilograms and measure around 15 centimetres by 20 centimetres by 20 centimetres.
Magnetic surveys are one of the most common geophysical applications. Magnetometers measure changes in the earth’s magnetic field (typically along fixed intervals of a grid pattern) to deduce areas of mafic and ultramafic rocks (such as magnetite), which have higher and more variable magnetic fields than rocks such as limestone. Magnetic surveys are useful for examining terrain where rocks and earth are hidden by groundcover.
UAVs are also currently used to carry LiDAR sensors to map terrain, ground-penetrating radar (GPR) equipment designed to image the near-surface, and induction (EMI) sensors for measuring near-surface conductivity.
In the near future, drones might carry seismic monitoring equipment or tools designed to take gravimetric readings (measuring changes in the Earth’s gravitational field to calculate density contrasts in surface rock and sediment, which can be used to detect deposits associated with nickel, diamond kimberlites, and iron).
Beyond their many applications, UAVs are relatively easy to fly. Between liftoff and landing, the craft is essentially on autopilot. “These are sophisticated robots with sensor packages using GPS for guidance,” said Bell. They are well-equipped for self-control and contain an internal measurement unit (IMU), accelerometers in the x, y, and z directions, a compass and a microprocessor. “So really the pilot’s job is just to lift it off and land it,” he said. “We usually have a visual operator and sensor operator out there with a radio to see things in the air space that the pilot may not be aware of and communicate that to the pilot.”
The limitations
The requirement for direct line-of-sight flying is not the only limitation faced. The size and weight of the surveying equipment is the major limiting factor, but as technology and science progress, that becomes less of a problem.
“The biggest obstacle in any drone instrumentation is the balance between size, cost and signal-to-noise ratio [SNR – basically a measure of usable data versus extraneous interference],” said Jan Francke, director at Vancouver-based Groundradar Inc., a company that specializes in the application of unique GPR technology for deep mineral exploration and geotechnical applications. “For example, we could build a very cheap (disposable) EM sensor that is super light, but completely unusable because it will record mainly noise. Geophysical instruments were designed for ground or fixed-wing use, and they are the sizes they are for a reason.” Making them lighter can have attendant drawbacks. He explained that, for example, companies developing MicroElectroMechanical System (MEMS) sensors for gravimetry realize that the level of SNR possible today might be okay for the oil and gas industry, but is likely not good enough for mineral exploration.
“Once you move to a bigger, more powerful instrument, it generally means the unit is much heavier,” said Bell. “This translates into requiring a UAV capable of lifting the heavier payload, which in turn means investing in a more capable and expensive UAV.”
Additionally, technology has not entirely solved the problem of the weather. “Perhaps the first challenge, which is also one of the biggest advantages of drone magnetic surveys, is learning to be flexible with local weather conditions,” said Bell. “While it is impossible to acquire good data in high wind conditions, one can acquire data in modest winds. In the end, the crew can wait out the wind at a much lower cost compared to a conventional pilot-on-board aeromagnetic survey.”
Ground-truthing UAVs
Near-surface conductivity measures variations in electrical conductivity found in different rocks and minerals, which can be analyzed to reveal potentially useful mineral deposits. It usually means walking with an electromagnetic induction (EMI) conductivity meter in hand to map lateral and vertical variations in subsurface electrical conductivity. According to Bell, EM variations can show different types of rock, weathering in the rock or even a contaminant in the soil (i.e., a situation-specific oil seep coming from depth).
Bell, who has over 30 low-altitude drone surveys under his belt, spent over six months in 2020 working to adapt a ground-based EM conductivity meter for drone use – this included upgrading his drone by adding a radar altimeter and planning for some more test flights this winter. During the past three years, he has already flown a number of magnetic surveys over a testing area in southern Colorado, where he knows that the site has between one and four per cent magnetite in the ground. He is hoping that the EM survey will help him improve and correct and understand the data he captured in his magnetic surveys.
He also wanted to prove what he long suspected about EM drone surveys: that an ultra-low-altitude drone EM survey would handily beat walking that line because of factors “such as reduced risk in highly contaminated areas [such as some mine slags] and the prospect of acquiring a sizable volume of high-definition data in comparatively less time.” He believed those advantages offset practical considerations such as how long a drone can stay aloft, how it can get around obstacles, how it navigates rugged terrain, how to deal with the impact of motion-induced noise and how to accurately follow precise geo-referencing.
Bell directly compared data collected via a sensor mounted on push-cart and data collected via an ultra-low altitude drone flight. To him, it was obvious that the drone delivered superior data – a conclusion he intends to present at an upcoming geophysical conference
“Even though I didn’t have a proper radar altimeter to control the altitude of my aircraft, I could see the data were not as impacted by the movement of the vehicle as it was when we had it on the cart. The wheels of the cart on the ground would cause the sensor, the acquisition system, to move around. When I had it hanging below the aircraft, as long as the aircraft was moving smoothly over the ground, that actually removed a lot of the noise. From that point of view, my determination was that I had better data and I didn’t have to correct for the movement of the sensor.”
From his extensive time in the field with UAVs – mostly in mag applications – Bell said there are a few fundamentals that should be followed, beginning with choosing a drone type. “Quad copters [drones with four rotors] are good drones,” he said. “But it is better to have six or eight rotors – in case one fails, you can land it safely.”
Bell said he generally takes four to six additional battery pack sets, with each set consisting of six batteries. “If we take four, the other three are usually charging from our portable generator.” To work with the short flight times, drone-enabled magnetic surveys are broken into a series of individual flight blocks, each containing one continuous data stream, which is parsed into individual flight lines.
As for coverage, Bell observed that the area surveyed will depend on the density of data needed. “In a mag ground survey, you are probably going to be at 25-metre line spacing, so we operate the drone at 25 metres,” he said. “[Other] mineral exploration projects generally have a little wider line spacing.”
Ground-penetrating radar
Drone-mounted ground-penetrating radar (GPR) has also become more common and shares most of the advantages and challenges with other applications.
As with traditional cart-carried GPR, airborne GPR units typically include a long, narrow and flat antenna that is mounted underneath the drone so it is horizontal to the ground and an on-board data recorder. The GPR emits a pulse of radio waves and the antenna picks up the echoes from the material below the ground.
Using LiDAR to create a high-resolution digital elevation map (DEM) in advance is a good way to enhance results using any geophysical method that calls for low-altitude flying, such as GPR.
“This usually requires first flying the region for photogrammetry [usually with a drone] to generate a very accurate DEM, followed by a drone magnetic survey, which can ‘terrain-hug’ at very low altitudes,” said Francke. “In the case of GPR, things get much more complicated because you need to be usually within one metre of the ground, so one needs sense-and-avoid technology (for obstructions) along with an accurate terrain [map].”
Hugging the terrain
The 2018 search for a missing WWII aircraft lost in the ice of Greenland illustrates the inherent efficiency of GPR equipped-drones, said Kristaps Brass, technical project manager at SPH Engineering.
“We managed to find what appeared to be the aircraft after the first flight. The team with the ground-based GPR later confirmed it.” After a few more cross-grid flights, they determined the approximate location of the aircraft which was confirmed by drilling.
The Greenland expedition emphasized some of the same requirements Francke noted about the need for drones to hug the terrain – an ability that is vital for GPR survey flights being conducted for mineral exploration.
“We found that it is critical to have true terrain-following when using GPR on a drone,” explained Brass. “So now we have made a system that includes an onboard data logger combining UAV GPS data with the GPR data as well as an altimeter for terrain following.” He said that system allows automatic start/stop GPR data logging and following the terrain at a pre-set altitude. “Otherwise, even if the terrain elevation data seems to be quite accurate, it can still have differences compared to the actual terrain,” he said. “So it is critical to be able to fly according to the terrain.”
Besides the need to follow the actual terrain, using drone-mounted GPR presents other challenges, with loss of depth being the primary one. “From our experience, using GPR on a drone decreases the penetration depth by about two times compared to using the same GPR on the ground,” explained Brass. Adopting larger more powerful antennas to overcome this shortens flight times by draining battery power, assuming they can be carried at all.
Still, for geophysicists who want to mount GPR sensors on drones, Brass offered some practical advice based on his field experience. “Optimal altitude depends on the antenna used,” he said. “Antenna altitude over the ground should [ideally] be about 0.6 metres.” He cautioned, however, that some antennas have a dead zone. “So we’d have to fly higher, about 2.5 metres, to avoid the dead zone from hiding some of the data near the ground surface.”
Brass explained that in areas with varying terrain elevations, such as flying up the slope of a hill or a mountain, SPH Engineering typically leaves more distance between the antenna and the ground, usually between 1.0 and 1.5 metres. He said that distance works as long as the drone is not flying too quickly, and noted that the ideal drone speed should be between one and two metres per second.
When it comes to choosing the right UAV, Brass pointed out the features he thinks critical. “The most important [factors] would be the autopilot used, the maximum payload capacity, as well as flight time,” he said, explaining that the GPR logging system needs to be compatible with the autopilot in order to ensure true terrain following can occur. For the autopilot Brass said, “the mission is planned using our UgCS [Universal ground Control Software] flight planning software. As for the altimeter, we are currently using either laser or radar (not to be confused with the GPR). Radar altimeters seem to perform better above reflective surfaces such as water, which is important for bathymetric surveys.” The company also supports the A3 autopilot system from Shenzhen, China-based DJI.
Drone usage in the future
Besides the magnetic, GPR, and near-surface conductivity surveying already taking place, there are surveying technologies with the potential to be mounted on drones, such as seismic sensors, gravimetric sensors and gamma ray spectrometers.
Bell said that he has been approached by an organization that wants him to start testing its gamma ray spectrometer within the next few months. “That has implications for mapping alteration geology,” he said. This type of surveying could prove useful for locating potential uranium, thorium and potassium deposits.
Something that Bell sees as being a little further off is gravimetry, as gravimeters generally weigh 10 kilograms or more. “Current airborne gravity meters are way too large and heavy, and they need more power,” said Bell. He added that a chip-based meter developed at the University of Glasgow and announced in 2016, with potential to be smart-phone sized may work for drones. “Power is always a big factor.” Providing corrections for motion – roll, pitch and yaw – also needs to be addressed because gravity surveys are much more sensitive to platform movement.
UAV-based seismic surveying is a work in progress, too. Envisioned for work in places like jungles and deserts, wireless vibration-sensing geophones can be dropped by drones from altitudes high enough to penetrate the ground. The geophones capture sound waves that have been directed into the earth and reflected back by the geological layers. The technology works, but current test projects are very labour-intensive requiring humans to find those half-buried geophones and bring them back after the spread is shot.
The energy company Total, has conducted experiments with both retrievable and disposable drones. In 2017, the company used one drone to drop approximately 60 seismic sensors in the form of partially biodegradable Downfall Air Receiver Technology (DARTS). Although these DARTs were retrieved, Total is looking for ways to make all of the parts biodegradable. The company also had plans to use five drones to release 4,000 DARTs in an Abu Dhabi desert location and retrieve them via an autonomous unmanned ground vehicle.
When it comes to positioning and placing the geophones, various concepts are envisioned, including some that are pie-in-the-sky at this point. One approach proposes deploying a swarm of thousands of drones. This is Total’s plan for a proposed 2022 drop, when the company hopes to release up to 100,000 drones in either a desert or forest environment.
The swarming approach is one that is being considered for more applications than seismic surveying, according to Francke. “Swarms have definite applications in situations such as complex transmitter/receiver geometries (e.g. EM, GPR), or collecting data from multiple flight lines or multiple altitudes at once for magnetics,” he predicted. “They will definitely need to be used for the concept of ‘grasshopper’ surveys, where sensors are dropped off and picked up on a pre-defined grid.”
That, however, may be a while. In most countries, a pilot is required for each drone, but Francke is optimistic: “The software is absolutely ready to go for swarms in geophysics,” he said.
Bell is enthusiastic about the future, too. He foresees drones with several hours of available flight time with a navigation system that allows for beyond visual line-of-sight flight operations becoming available within the next 12 to 18 months. Moreover, “the advent of a workable unmanned traffic management system will measurably encourage more use of drones for geophysical data acquisition and geoscientific mapping,” he predicted.
Por Graham Chandler
Fonte: CIM Magazine
