All-Seeing Eye: World Experience in Remote Sensing of the Earth

Photographic progression

The first methods of remote sensing of the earth appeared at the beginning of the 20th century using aerial photography. Back in 1858, the French photographer Felix Nadar made the first flight in a hot air balloon with a photographic camera. He took a series of photographs of Paris from above, which became the first in the history of aerial photography. A similar achievement in Russia was made by the Russian military commander, military electrical engineer Alexander Matveevich Kovanko in 1886. He photographed St. Petersburg from a height of 800, 1200 and 1350 meters. At first, aerial photography was used mainly for military reconnaissance, but the resolution of the images was low, and data processing was labor-intensive. Nevertheless, this method remained very effective for many years: in Russia, it was used to study forest resources, make contour plans and maps at scales of 1:2000-1:50000.

The launch of the first artificial Earth satellite, produced in the USSR in 1957, marked the beginning of the space era in the history of remote sensing. Although Sputnik 1 itself was not equipped with photographic elements, aerial photography was used for optical observation of it and subsequent calculation of its orbit. The NAFA camera, a night automatic aerial camera, was adapted for these purposes. Two years later, the Americans sent their Explorer 6 satellite into orbit and received the first photograph of the Earth. Following this, in the 1960s, with the help of space images, a large oil field in Western Siberia was discovered for the first time in the USSR. The images captured geological structures, including depressions, faults, and changes in vegetation, which are characteristic of oil and gas deposits. Later, this field played a central role in the structural reorganization of the economy and energy balance of the USSR. By 1969, the Soyuz-6, Soyuz-7 and Soyuz-8 spacecraft were already studying geological formations on the eastern coast of the Caspian Sea. Soyuz-9 received a large number of images of geological and geographical objects in the southern regions of the European part of the USSR, in Kazakhstan and Western Siberia.

Since 1972, the Landsat program was launched in the United States - the longest-running project to create satellite images of the Earth. The devices were equipped with multispectral, panchromatic video cameras and scanners. Multispectral remote sensing is basically based on the fact that rocks and minerals have different spectral characteristics at different wavelengths. For example, iron ore has specific spectral features in the visible and infrared ranges of the spectrum, which can be used to detect iron deposits. In contrast, panchromatic cameras make black and white images using only one spectral channel. They usually have a high resolution with clear detail and visible textures on the ground.

In 1999, the era of high resolution began in Earth remote sensing. The commercial satellite IKONOS was launched from the Vandenberg Space Center in California, the first in the world capable of obtaining images up to 0.82 meters in panchromatic mode and up to 3.2 meters in multispectral mode. In 2001, the American group was replenished with QuickBird with a wide coverage band, high metric accuracy and resolution of up to 0.6 and 2.4 meters in various spectral modes. Despite the high quality of the images, the satellites were still dependent on the time of day and weather conditions. It was almost impossible to make a detailed image in fog or at night until Synthetic Aperture Radar technology appeared. It had been worked on in the USA since the 1950s, based on the Doppler effect. According to the same principle as the sound of a passing car changes depending on its approach to or distance from the observer, each object in the radar beam has a slightly different speed relative to the antenna recording it. Accurate frequency analysis of these reflections allows us to build a detailed image. The effectiveness of this method was tested by NASA back in 1978 when studying the oceans from space, including their waves and currents. In 1991-1995, the European Space Agency (ESA) used SAR technology to monitor glaciers and measure the deformation of the earth's surface.

The use of Doppler satellites was truly put on stream in the 21st century. In 2007, the German TerraSAR-X satellite was launched from the Baikonur Cosmodrome, which for the first time created high-precision digital models of the Earth's relief, and also studied tectonic movements and landscape changes. Subsequently, similar satellites were sent into space by ESA (Sentinel-1) and Japan (ALOS-2), which was interested in studying the effects of tsunamis and earthquakes.

According to the developer of software solutions in the field of geoinformatics Rakurs, in 2024, every third launch vehicle had Earth remote sensing spacecraft on board. In total, according to the UCS database, there are currently 1,192 remote sensing satellites in orbit. Among them are both government devices and commercial ones used for agriculture, ecology, climatology and cartography. At the same time, their number has grown by 13% over the year, and optical visualization satellites remain the invariable leader, representing just over 40% of the fleet.

Big Brother Watches the Earth

The United States has some of the most advanced remote sensing expertise in the world for geological exploration. Landsat, launched by NASA and the USGS, is the longest-running and most effective remote sensing program. After Landsat 5 was shut down in early 2013, Landsat 7 remained the program's only operational satellite; the eighth was launched on February 11, 2013. Four years later, researchers used Landsat 8 data to identify previously unknown glacial lakes in Antarctica, which contributed to a better understanding of ice melt processes and their impact on sea level rise. In 2019, the data was used to create the first-ever map of anthropogenic changes in land resources. The satellite's data also made it possible to estimate the metallogenic potential of the southern part of the Greater Kuril Ridge, including Kunashir Island. They contain concentrations of trace elements (molybdenum, bismuth, cadmium, indium, germanium, etc.), precious metals (gold-silver alloys), and non-ferrous metals (zinc, copper, nickel, zinc, lead).

The satellite also explored the Thai Nguyen Province in Vietnam. Scientists were able to identify areas containing high concentrations of clay minerals, including kaolinite, montmorillonite, illite, and others. The satellite data also helped to accurately localize areas with significant accumulations of iron oxide, a valuable raw material for the metallurgical industry. Statistical processing of Landsat-8 data made it possible to construct maps of the distribution of hydrothermal changes in the Central Part of the Malouralskaya Zone. The study showed that potentially gold-bearing and polymetallic areas are concentrated along transregional fault zones that intersect favorable structures with ore mineralization. In addition, areas with high levels of iron oxides (II and III), and sometimes hydroxide- (Al-OH, Mg-OH) and carbonate-containing minerals were identified.

The European Sentinel-2 program made it possible to visually assess the scale and features of the Elga coal deposit in Yakutia, the largest coking coal deposit in Russia. The detail of the images allows engineers and designers to develop plans for the expansion of quarries, construction of roads and other infrastructure, and scientists can study the geological features of the region, analyze soil erosion and degradation processes, and model future changes.

The RADARSAT program, developed by the Canadian Space Agency (CSA), is one of the country's key initiatives in the field of remote sensing. RADARSAT satellites use synthetic aperture radar (SAR), which allows data to be obtained regardless of weather conditions and time of day. This is especially useful for monitoring hard-to-reach regions, such as the northern territories of Canada. Although the satellite is designed to monitor the environment rather than search for minerals, its data can be used to analyze geological structures and tectonic features that indicate possible deposits of oil, gas, or metals. However, at present, only the satellite’s environmental achievements are known: in April–June 2012, it recorded oil slicks on the surface of the Caspian Sea, and also obtained data on the displacement and deformation of the earth’s surface and structures above the Zhezkazgan copper deposit in Kazakhstan. Most of the data collected is used to assess the state of the ice in the Arctic. For example, the radar tracks changes in ice thickness, their reduction in area, and monitors the ice situation around Canada’s oil infrastructure. Its images are used to create maps of the ice cover and to plot the routes of the Northern Sea Route.

Developing Asian countries are also actively implementing Earth remote sensing (ERS) technologies for geological exploration, but have serious limitations on resources and infrastructure. India, Indonesia, Vietnam and the Philippines actively use data from the Landsat (USA), Sentinel (EU) and MODIS satellite programs - a moderate-resolution spectroradiometer. India is also developing its own ISRO program, which includes the Resourcesat and Cartosat series satellites used for agriculture and mapping.

The leading position among Asian countries is occupied by China - one of the world leaders in the field of Earth remote sensing. Its Gaofen program includes a series of satellites providing high-resolution data. The fifth series of devices has become the flagship of observation, since it allows detecting objects on Earth, including identifying enemy equipment for military purposes. Another Earth remote sensing program - ZiYuan, which is translated from Chinese as "resources" - was developed for geological and agricultural purposes. The satellites, first launched in 1999, have been used to locate coal, iron ore, copper, and gold deposits. For example, new rare earth element deposits, which are of strategic importance to high-tech industries, have been discovered in the provinces of Xinjiang, Qinghai, and Yunnan. They have also been used to pinpoint the exact locations and volumes of rare earth deposits in the Inner Mongolia and Jiangxi regions. In the Tarim Basin (Xinjiang) and offshore in the South China Sea, satellite data has helped identify promising areas for oil and gas exploration. New uranium deposits have been discovered in Hunan Province, which is essential for China’s rapid development of nuclear energy. China has launched satellites with Brazil as part of the CBERS program to study mineral deposits in South America, including iron ore, bauxite, and copper.

Thanks to its developed space industry, scientific base and rich natural resources, Russia with 68 launched devices also occupies one of the leading positions in remote sensing. For example, the satellites of the Resurs series are used to study deposits of various minerals, including oil, natural gas, and ores. The data obtained with the help of these devices helped to discover diamond deposits in the Arkhangelsk region. In addition, space images have found application in geological exploration work in the artesian basin of the Leningrad region. Based on the data of the Kanopus-V series satellites, developed by the VNIIEM Corporation together with the British company Surrey Satellite Technology Limited, it was possible to build a predictive model of the diamond content of the Ermakovskaya-7 kimberlite pipe on the Kola Peninsula. Back in 1986, during an expedition there, 131 small diamonds were discovered. Research into the deposit is still ongoing. The Condor satellites are used to monitor oil and gas fields in Western Siberia (Samotlor, Priobskoye, Urengoy). Spacecraft of this series also help to study promising areas for oil and gas exploration in hard-to-reach areas of the Arctic - on Yamal and the Gydan Peninsula.

Race at Quantum Speeds

Earth remote sensing technologies are advancing rapidly, with hundreds of research institutes around the world taking part. Hyperspectral sensors offer a completely new perspective on space observation. These sensors can collect data in hundreds of narrow spectral bands, greatly improving the ability to identify materials and minerals on the Earth’s surface. In fact, the technology is a combination of traditional spectroscopy and a modern imaging system. The first of its kind, designed to demonstrate the capabilities of this technology, NASA’s Hyperion was installed on the Earth-Observing-1 satellite and was decommissioned in 2017. Its sister satellite, the Italian Space Agency’s PRISMA, was launched on March 22, 2019. The innovative electro-optical equipment, which combines a hyperspectral sensor with a medium-resolution panchromatic camera, collects images with a resolution of 30 meters within scenes measuring 30 by 30 kilometers. In 2024, the private space company Sputnix and the Samara University named after Korolev were noted for their remote sensing satellite with a hyperspectrometer with a resolution of seven meters per pixel. The creators call the resolution of the device a record for its compact size.

Small satellites - cubesats - are becoming increasingly popular due to their low cost and the ability to launch large groups. They allow you to receive data with a high update rate, which is important for monitoring dynamic processes. Their basic size is 10 × 10 × 10 cm with a mass of no more than 1.33 kilograms. For example, SkySat devices of the American company Planet Labs are engaged in photography and recording short videos lasting 90 seconds at a speed of 30 frames per second. At the moment, the resolution of the satellites has reached 50 meters, for which the company lowered their orbit. This made it possible to get a more accurate picture of changing conditions on the spot and add detailed context to what is happening on Earth.

Artificial intelligence and machine learning are used to automate the analysis of remote sensing satellite data. AI can automate many tasks, from processing space images to creating analytical reports on various topics. Thus, neural networks can identify cloud formations, their shadows, as well as fog and other interference in images, and then make the necessary adjustments to the original data. Various software packages are used to analyze and interpret remote sensing data, including ERDAS IMAGINE from the American corporation Intergraph, ScanEx Image Processor, the Russian company SCANEX, ENVI ITT Visual Information Solutions from the USA. These programs are based on computer interpretation algorithms, and artificial intelligence also helps classify objects in images. By dividing satellite images into segments corresponding to different types of terrain, including forests, water bodies and cities, scientists can observe changes in land use. In the process of processing this data, convolutional neural networks are used, which are trained on extensive arrays of information and learn to recognize patterns and features that are important for accurately separating objects in satellite images. In addition, artificial intelligence helps to assess the condition of crops, identify plant diseases and monitor vegetation growth.

Hybrid clusters are used to efficiently process Earth remote sensing data. The technology is a combination of data from various sensors (optical, radar, laser), which allows for a more complete picture of the objects being studied. As a rule, standard computing modules are supplemented with graphics accelerators (GPU). An example is the hybrid cluster NKS-ZOT+GPU used at the Siberian Supercomputer Center. The SSCCIP system is used to process data, which allows for the integration of remote multiprocessor machines into the process. Geospatial platforms such as Google Earth Engine or Roscosmos Digital Earth are gaining wide popularity. These technologies combine and analyze multiple sources, including satellites, drones and ground sensors.

Quantum sensors are being developed that can significantly improve the accuracy of magnetic field and gravity measurements. The quantum gravity gradiometer, created by researchers from the University of Birmingham, makes it possible to study underground structures without using invasive techniques. This device records the slightest fluctuations in gravitational fields, allowing us to analyze objects of different sizes and compositions hidden underground, including artificial structures. One promising direction is the use of sensors based on neutral atoms. They use the quantum mechanical properties of atoms to measure physical parameters such as magnetic field, gravity and temperature. Atoms are placed in a special optical trap, where they are held by laser radiation. Laser light controls their state, transferring atoms between different energy levels. Being in a certain state, atoms begin to react to external fields - magnetic or gravitational. Any changes in these fields lead to a change in the frequency of transitions between the energy levels of the atom, which is subsequently recorded by devices.

Sensors based on neutral atoms have a number of significant advantages over classical remote sensing methods. Firstly, they are highly sensitive, which allows them to detect even the smallest changes in physical fields, making them indispensable for detecting weak signals. Secondly, thanks to the use of quantum mechanical effects, these sensors demonstrate exceptional measurement accuracy. In addition, since the atoms are protected from external influences, the noise level is significantly reduced, which has a positive effect on the quality of the collected data. Finally, the versatility of these sensors, which allows them to measure a variety of physical fields, significantly expands the scope of their practical application. The development of such sensors and their components is currently underway at several research institutes, including the Max Planck Institute in Germany, the National Institute of Standards and Technology in the United States, and the Moscow Institute of Physics and Technology (MIPT). Singaporean Atomionics, French muQuans, and American ColdQuanta are also working on quantum sensors.

Modern technologies in the field of remote sensing of the Earth are aimed at increasing the accuracy, regularity and availability of data. They create new prospects for exploration of deposits, environmental monitoring, agricultural development and efficient management of natural resources. The implementation of these technologies requires an integrated approach that combines the achievements of the space industry, artificial intelligence and big data. The United States today holds the palm due to the large number of private companies. They are followed by China, whose private initiatives in the space sector are estimated at $ 1.5 billion annually. Russia currently occupies the third line of this world ranking. Only in 2024, the State Duma approved the federal law on the use of concession agreements and public-private partnership mechanisms in the field of space activities. In addition, the Russian government allocated 1.4 billion rubles for the implementation of the project "Development of high-tech areas "Promising space systems and services". Part of these funds were used to purchase ready-made Earth remote sensing data under the first forward contract concluded at the end of the year between Roscosmos and the Sputnix company. According to a number of space industry experts, what Russia currently lacks is primarily serial production of launch vehicles on board which satellites are delivered. In addition, Russian space should be commercialized in order to reduce government influence on its support. First of all, this concerns the delivery of foreign equipment into orbit by domestic rockets, and for this it is necessary to restore international cooperation at least in the scientific sphere.