Earth observation

Earth observation (EO) is the gathering of information about planet Earth’s physical, chemical, and biological systems. It involves monitoring and assessing the status of, and changes in, the natural and man-made environment. EO is sometimes described as gathering information about the Earth through remote sensing technologies often involving some form of digital imagery. However, EO also includes direct data collection using in-situ instruments; examples include analyzing soil samples or processing information from maps or numerical measurements taken by thermometers or wind gauges.

Modern EO is performed using a wide range of instruments:

• Floating buoys for monitoring ocean currents, temperature, and salinity
• Land stations that record air quality and rainwater trends
• Sonar and radar for estimating fish and bird populations
• Seismic and Global Positioning System (GPS) stations
• Aerial platforms such as chartered planes, drones, or satellites

Since the first photo of Earth from space was taken from a V-2 rocket in 1946, and the first artificial satellite, Sputnik 1, was launched in 1957, space-based technologies for EO have found significant use, providing significant scientific data and new views of Earth from space.

The first photograph of Earth taken from space, taken by a 35mm motion picture camera on a V-2 rocket at an altitude of 65 miles.

Data from the Union of Concerned Scientists (UCS) satellite database found there were 1,052 active satellites with the main purpose of EO or Earth science on December 31st, 2021. EO satellites deliver reliable and repeat-coverage datasets, that when combined with the development of appropriate methods, provide a unique means for gathering information concerning the planet. Examples include the monitoring of the state and evolution of our environment and the ability to rapidly assess situations during crises such as extreme weather events or during times of human conflict.

With the growing impact human activities are having on the environment, observations of the Earth are becoming valuable to understand, assess, and mitigate their effects. EO data can also be used to exploit new opportunities, such as sustainably managing natural resources. Specific applications of EO include the following:

• Weather forecasts
• Biodiversity and wildlife trends
• Understanding land-use change (such as deforestation)
• Monitoring and responding to disasters, including fires, floods, earthquakes, and tsunamis
• Managing energy sources, freshwater supplies, and agriculture
• Addressing emerging diseases and other health risks
• Predicting, adapting to, and mitigating climate change
• Ocean Science
• Rural and Urban development
• Geology & Geomorphology
• Defense & security
• Maritime surveillance

Remote sensing allows observers to obtain the physical properties of an area from a distance. For EO, it allows users to capture, visualize, and analyze objects and features on the Earth’s surface, collecting imagery for analysis. The EO sensors required are typically mounted on specialized platforms aboard airplanes, helicopters, satellites, UAVs, or balloons. While satellites can capture data on a global scale, drones offer a simpler solution for obtaining data across smaller regions.

EO remote sensing or imagery can be classified based on multiple factors, including passive vs active and the wavelength of the electromagnetic spectrum measured. Given the electromagnetic transparency of the atmosphere observations of the Earth's surface can be made using wavelengths from the visible spectrum, parts of the infrared spectrum (from 0.70 to 14 μm), and the radio wave range (from 1 cm to 11m).

Atmospheric transmission of electromagnetic radiation.

A number of key parameters define the final EO imagery produced by a particular system. Different systems and types of remote sensing favor certain parameters depending on the science behind their operations or the particular application and data they are looking to obtain.

Spatial resolution defines the size of the pixels and the area they represent on the Earth's surface. High spatial resolution means a greater level of detail can be retrieved from the image. Spatial resolution can be measured in multiple ways. One of the most common is ground sample distance (GSD), the distance between the center of adjacent pixels measured on the ground.

Example of varying levels of spatial resolution.

Spectral bands are the groups of wavelengths being measured by the sensor categorized by their frequency or wavelength.

Spectral resolution is the amount of spectral data in or the width of a band. High spectral resolution means narrow bands while low spectral resolution means broad bands covering more of the electromagnetic spectrum.

EO sensors on satellites are characterized by their swath width, the area of the ground it can image simultaneously from orbit. The swath depends on the design of the sensor and the satellite's orbit. Generally the higher the spatial resolution, the lower the swath of the instrument (i.e., either capable of imaging a small area in high detail or a large area in less detail).

Revisit time is the time between subsequent observations of the same area. EO data can be needed for a particular deadline (e.g., planning road repairs in a remote area), urgent (e.g., responding to a natural disaster), or periodic (e.g., monitoring crops). For satellites, revisit time depends on the type of orbit. Most EO satellites are in low Earth polar orbits that are sun-synchronous with altitudes and inclinations such that they observe the same scene over time with the same angle of illumination from the sun.

A related parameter is temporal resolution, or the time it takes for a satellite to complete a full orbit. While terrestrial aircraft are flexible, satellites orbit the Earth in set paths.

• Bit depth—the number of bits per band for each pixel. Also known as radiometric resolution, increasing bit depth increases the detail in an image's bright and dark regions.

• Off-nadir angle—high-resolution sensors generally do not image the Earth's surface directly above the target (the "nadir," or looking straight down on the target). The angle at which the target is imaged is known as the off-nadir angle, where 0 degrees is directly above. A low off-nadir angle is generally preferable, especially in areas of high relief or tall buildings to reduce the "building-lean" effect. A typical maximum off-nadir angle is 30 degrees. Imaging at the nadir of the target can limit the information gathered with only the top of the object visible to the sensor.

Demonstration of the building-lean effect caused by the off-nadir angle.

• Sun-elevation angle—the angle of the sun above the horizon affects the imagery collected. During low sun elevation periods, data can be too dark to use. A typical minimum sun-elevation angle is 30 degrees. Many EO satellites are in sun-synchronous orbits with little control over the time of day an area can be imaged.

• Cloud cover—the presence of clouds above an area of interest affects optical imagery.

Passive sensors detect electromagnetic emissions from constituents of the Earth's surface and atmosphere. This could be emissions locally produced (e.g., thermal radiation from vegetation in the infrared spectrum) or the result of reflected sunlight in the visible spectrum. As passive sensing does not provide an external source of electromagnetic signal, it is usually dependent on the day-night cycle and can be degraded by perturbations from unwanted sources of emission or cloud cover. Examples of well-known EO satellites passively imaging the Earth include Landsat and Sentinel.

Panchromatic images contain light from a wide range of the electromagnetic spectrum, allowing for more energy to be collected and improved resolution (up to 30cm for the best commercially available satellite instruments). A typical example of a panchromatic image could measure the light from Earth's surface across the entire visible spectrum, displaying the final image as shades of grey to signify intensity. As visible spectra require reflecting sunlight, this type of imagery can only be obtained during the daytime.

Example of a panchromatic image.

Another example of a panchromatic might be detecting thermal infrared energy to find information about the objects emitting the radiation. As infrared radiation is constantly emitted, IR satellite imagery is possible even when the surface is not illuminated by the sun.

Multispectral instruments measure the observed scene across multiple narrow bands of the electromagnetic spectrum. Restricting the wavelength of the radiation contributing to the final image means multi-spectral instruments typically produce lower resolution images compared to panchromatic and have to collect photons across larger spatial extents. A common example of multi-spectral images includes "natural color" images that combine three bands of the visible spectrum (narrow bands centered around blue, green, and red wavelengths). This is what standard consumer cameras do. Many other wavelength band combinations are used, depending on the information looking to be found:

• Shortwave infrared, near-infrared, and green—used to show floods or newly burned land
• Blue and two different shortwave infrared bands—used to differentiate between snow, ice, and clouds
• Blue, near-infrared, and mid-infrared—used to image water depth, vegetation coverage, soil moisture content, and the presence of fires

Pan-sharpening is the process of merging multi-spectral and panchromatic images to provide higher resolution color images.

Hyperspectral sensors generate images where each pixel contains a large number (from tens to several hundred) of contiguous narrow spectral bands. Whereas multispectral images are generally made up of a small number of wide bands, hyperspectral images contain a large number of narrow bands. Due to this large number of bands, every pixel in a hyperspectral image has a nearly continuous spectrum associated with it. With spectral resolution, hyperspectral images allow for the detection, identification, and quantification of surface materials, as well as inferring biological and chemical processes.

Microwave radiometers (MWRs) measure atmospheric water vapor column and cloud liquid water content primarily used for correcting radar altimeter signals. MWR measurements can also be used for the determination of surface emissivity and soil moisture over land.

Active imaging uses a transmitter to send out specific signals that interact with the Earth, producing a new signal that is measured by a sensor. Active sensors do not depend on solar illumination. An example of an active sensor is Radarsat-2, which uses synthetic aperture radar.

Demonstration of an active sensor satellite.

The most common active sensor for EO is the synthetic aperture radar (SAR). SARs transmit electromagnetic pulses towards the Earth's surface where they interact with surface features, reflecting or scattering. The instrument then detects the return pulses, recording the intensity and time the pulses take to return to generate SAR imagery.

Example of a SAR image showing Isla Isabella in the West Galapagos Islands.

SAR imaging is not dependent on sunlight and generally is immune to meteorological effects. The wavelength of the radio band used affects the level at which the incident radiation backscatters influencing the final image. There are a variety of SAR applications:

• Ship detection
• Oil spill detection
• Sea ice monitoring
• Forest monitoring
• Soil moisture measurements
• Assessing critical infrastructure

A technique known as SAR interferometry can produce accurate measurements of geophysical factors including topography, ground deformation/subsidence, and glacier movements. SAR interferometry compares the phase of multiple complex radar images acquired at slightly different positions or at different times. The phase of each SAR image pixel provides range data accurate to a small fraction of the radar wavelength used and it is possible to detect path length differences with high precision (centimeters or millimeters). Comparing the range information from multiple radar images produces differential SAR interferometric measurements where subtle (millimeter) changes in the range can be detected.

Light detection and ranging (Lidar) for EO uses the same principle as SAR for shorter wavelength electromagnetic radiation (IR, visible, or UV). Lidar provides precise measurements of topographic features, monitoring the growth or decline of glaciers, profiling clouds, measuring winds, studying aerosols, and quantifying various atmospheric components. Examples of satellites with lidar instruments include the following:

• The Atmospheric Lidar (ATLID) on ESA's EarthCare mission providing vertical profiles of aerosols and thin clouds. The instrument utilizes 355nm wavelength radiation.
• The Atmospheric Laser Doppler Lidar Instrument (ALADIN) on ESA's Aeolus-ADM mission measuring line-of-sight wind profiles across different levels of the atmosphere. Again, the instrument operates at a wavelength of 355nm.

Radar altimeters use the ranging capability of radar to measure the surface topography profile along the satellite's track. It provides precise measurements of a satellite's height above the ocean by measuring the time interval between the transmission and reception of short electromagnetic pulses. Multiple parameters can be inferred from radar altimeter measurements including time-varying sea-surface height (ocean topography), the lateral extent of sea ice and altitude of large icebergs above sea level, as well as the topography of land, ice sheets, and the sea floor.

Radar Altimetry derived data from ESA.

A relatively new category of satellite navigation is GNSS reflectometry (GNSS-R), which entails a method of remote sensing by receiving and processing microwave signals reflected from various surfaces and extracting information about those surfaces. The process requires a GNSS satellite to act as the transmitter and an airplane or Low Earth Orbit (LEO) satellite as the receiving platform. For altimetry applications, a GNSS-R receiver can also be placed on the land. An advantage of GNSS-R remote sensing is the ubiquity of signal sources, including GPS, Galileo, GLONASS, and Beidou/Compass. Possible applications include wide-swath altimetry, sea-wind retrieval, and measurement of seawater salinity and ice-layer density, as well as humidity measurements over land.

Radar scatterometry utilizes a microwave radar sensor to measure the reflection or scattering effect produced when scanning the surface of the Earth from an aircraft or a satellite. It measures wind speed and direction near the sea surface from the backscatter of small waves at the sea surface, at skew incidence angles. Assessing the sea roughness also allows the wind vector at 10 m height to be ] calculated. Radar scatterometer data is an important factor for numerical weather prediction (NWP), oceanography, and climate studies.

The chemistry of the Earth's atmosphere can be monitored with various techniques and instruments that measure different parts of the electromagnetic spectrum. Atmospheric chemistry sensors can be either active or passive. Atmospheric gases are characterized by their “absorption” and “emission” spectra, which describe how the molecules interact with different frequencies of radiation. EO remote sensing instruments exploit these spectra to provide information on atmospheric composition from measurements over a range of wavelengths between UV and microwave.

Atmospheric absorption is typically dominated by water vapor, carbon dioxide, and ozone, with smaller contributions from methane and other trace gases. While relatively broadband instruments can measure dominant gases, high spectral resolution sensors are required to make measurements of weaker signal molecules. Atmospheric chemistry instruments are generally operated in nadir-viewing mode (looking vertically directly down to measure the radiation emitted or scattered) or in a limb-viewing mode (scanning positions beyond the horizon to observe paths through the atmosphere at different altitudes). Nadir-viewing instruments provide high spatial resolution in the horizontal direction but limited vertical resolution, whereas limb-viewing instruments provide a high vertical resolution (a few km) and limited horizontal resolution (tens of km at best).

Gravity field measurements define the "geoid," the surface of equal gravitational potential at mean sea level. Space-based gravity field measurements rely on three main techniques:

• Deriving gravity or gravity gradient information through one or more accelerometers
• Determining satellite orbits precisely (through ground navigation systems and laser ranging systems) and separating satellite motion caused by the Earth's gravitational field from other factors (e.g., solar radiation, aerodynamic drag, etc.)
• Measuring the relative speed fluctuations of two satellites caused by gravitational forces, typically by GPS or microwave link

The geoid shows the irregularities in the Earth's gravity field at the surface caused by the inhomogeneous mass and density distributions of its interior. Accurate models of the geoid over time provide important data for absolute ocean currents and how they transport heat/other properties, for estimating the thickness of polar ice sheets, and for estimating the mass/volume of freshwater to understand the hydrological cycle.

The first global gravity model from ESA's GOCE satellite.

EO data is used to produce 3D models of the Earth's surface for a variety of applications:

• Military operations
• Flight simulation
• Disaster management
• Mapping buildings
• Cadastral databases

Many processes can be used to create 3D models, including combining multiple images from the same scene taken at different angles to reconstruct depth data.

Data from the Union of Concerned Scientists (UCS) suggests there were 1,052 satellites active on the 31st of December, 2021, with the primary purpose of EO and/or Earth science. This is an increase of 8.34% from 30th April 2021, (971 active satellites. While some satellites have multiple purposes, of the 1,052 active satellites, their operations include the following:

• Optical imaging—426 satellites
• Meteorology—170 satellites
• Electronic intelligence—113 satellites
• Radar imaging—90 satellites
• General earth observation (no detailed purpose listed)—79 satellites
• Earth science—75 satellites
• Hyperspectral/Multispectral imaging—41 satellites
• Automatic identification system (AIS)—19 satellites
• Other purposes—17 satellites
• Infrared imaging—12 satellites
• Video—10 satellites

Two hundred nineteen organizations are registered as operating EO satellites; 155 only operate a single satellite, and 41 operate two. The top operators account for almost 55 percent of active EO satellites. These operators are:

1. Planet Labs Inc—188
2. Spire Global Inc—119
3. Chinese Ministry of National Defense—85
4. America's National Reconnaissance Office (NRO)—44
5. Chang Guang Satellite Technology Co. Ltd—29
6. Russia's Ministry of Defense—21
7. Satellogic S.A.—20
8. 17 satellites are operated by the China National Academy of Sciences (CNSAS) and Indian Space Research Organization (ISRO)
9. 12 satellites are operated by BlackSky Global, EUMETSAT (European Organization for the Exploitation of Meteorological Satellites), and ICEYE Ltd.

Seventy countries are listed as controlling EO satellites. The USA has the most, operating 43 percent of the world's active EO satellites, with China second with 25 percent. While satellites can have multiple users, the UCS lists that 495 satellites have commercial users, 310 have government users, 220 have military users, and 27 have civil users.

Different orbits are used depending on the measurements the satellite is making. The altitude of an orbit corresponds with its orbital period—the higher the altitude, the longer it takes to complete an orbit. Orbits can be divided into low earth orbit (LEO), medium earth orbit (MEO), and high earth orbit (HEO). Important orbits to consider for EO include the following:

• Geostationary orbits that match the Earth's rotation to remain at a fixed point in the sky above a single location
• Sun synchronous orbits that maintain a consistent angle of sunlight on the surface of the Earth
• Polar orbits that pass above or near the poles of the Earth
• Elliptical orbits with a high eccentricity deviating from a circular path such that the distance from the Earth varies significantly

As of the 31st of December, 2021, EO satellite orbits include:

• 989 in low earth orbit (94.01%)
• 46 in geostationary orbit (4.37%)
• 1 in medium earth orbit (0.09%)
• 16 in elliptical orbit (1.52%)

Satellite imagery is regularly used for agriculture, providing data to enable economic and environmental benefits to farm management processes. Satellite imagery can improve revenue generation for agricultural applications, providing information related to:

• Crop health monitoring and management
• Crop type
• Crop insurance damage assessment
• Production management practices
• Fertilizer application requirements
• Yield estimates
• Regrowth monitoring
• Illicit crop monitoring
• Pest and invasive species monitoring
• Irrigation requirements and application
• Field boundary management
• Field-scale mapping
• Monitoring agri-environmental measures (e.g., acreage) to inform subsidy allocations
• Assessing storm damage

EO resolution imagery provides detailed forestry information to:

• Obtain information on forest acreage, stand density (the quantity of trees per unit area) and monitor stand development (stand—a group of forest trees of relatively uniform species composition, age, and condition to be considered as a single unit for management purposes)
• Survey, evaluate, and monitor forest health
• Update forest management plans: tree felling, delimitation and monitoring of parcels, biomass estimation, plant health, and plantation monitoring
• Estimate fire, storm, and other extreme weather damage
• Plan and protect conservation areas
• Perform fuel analysis and identification of areas where the danger of fire is high
• Map deforestation
• Monitor forest regrowth and conservation activities

EO allows for improved risk management in a number of fields:

• Continental-scale mapping
• Urban growth
• Soil sealing
• Land cover classification

During disasters, EO satellites offer fast data to help rapid relief efforts. EO imagery can also be used to improve extreme weather predictions and categorize areas at risk.

Satellites allow for the monitoring of the Earth's oceans, including the following:

• Ship tracking
• Bathymetric data—used for measuring beach erosion, subsidence, sea levels, construction of harbors, and for creating nautical charts
• Detection of iceberg threats on shipping routes
• Marine environmental protection
• Oil spill monitoring
• Illegal oil discharge detection
• Detection of unlicensed fishing vessels
• Port monitoring
• Maritime piracy—detection of incoming inhospitable objects

Satellite imagery of the Earth helps countries looking to police their borders, coastlines, or overseas assets. Defense and security EO applications include:

• Mission planning and situational awareness through up-to-date mapping of sites of interest and the surrounding environment
• Analysis and monitoring of key sites and installations of interest e.g., weapon storage, airfields and hangers, ports
• Detection, recognition, and identification of military vehicles, particularly aircraft and naval vessels, including the identification of aircraft, detection of radar and surface-to-air missile emplacements, differentiation of tracked and wheeled vehicle types, and the identification and analysis of maritime vehicles
• Identification of specific infrastructure, such as railways and control towers.

EO allows both commercial and government entities to oversee and manage the use of natural resources, including the following:

• Water resource management
• Energy infrastructure optimization
• Oil and gas
• Mining

A number of companies and public entities are active in the EO industry, including these top commercial companies operating in the sector:

• US companies Planet, Spire Global, and BlackSky Global
• Chinese company Chang Guang
• Argentinean company Satellogic
• Finnish company ICEYE Ltd

The following are other EO companies:

• Premise
• Descartes Labs
• Slingshot Aerospace
• Capella Space
• Wyvern
• Azavea Inc
• Blackshark.ai
• Rendered.ai
• Rocket Lab
• Dotphoton SA
• Skyrora
• Orbital Insight
• Satellite Vu
• ExoAnalytic Solutions, Inc
• Thales
• Spacesense
• Alba Orbital Ltd
• Astra Research Center
• Enview
• Axelspace

ESA's EO science for society has compiled a list of EO industrial partners here. The European Union Agency for the Space Programme (EUSPA) estimates EO revenues for data and services to be €2.8 billion in 2021. Over half of the global revenue was generated by the top five segments: Urban Development and Cultural Heritage, Agriculture, Climate Services, Energy and Raw Materials, and Infrastructure.

EUSPA breakdown of EO revenues for data and services in 2021.