Skip to content

Other Space Activities

Synthetic Aperture Radar (SAR)

Mar 15, 2024

Instrument Types

The development and employment of satellite remote sensing has proved invaluable for improving our understanding of the Earth and its processes. Through these systems, it has become possible to take repeated observations of the entire globe. The applications are numerous, ranging from civil to scientific to military, as are the number of satellites taking observations.

Most satellite instruments can be divided into two groups: passive and active. Passive instruments such as optical cameras collect light and other electromagnetic waves reflected from a body’s surface (e.g. Earth) from sources like the sun. Active instruments, such as a Synthetic Aperture Radar (SAR), observe targets by transmitting an electromagnetic signal and measuring the properties of the reflected signal, including intensity (termed ‘backscatter’), and the time between transmission and return. This transmitted signal is analogous to the flash on a camera, or the ping sent out by a sonar. Having exact knowledge of both the transmitted and reflected signals can yield information on the properties of the observed area (e.g., texture, presence of water) and the distance between the sensor and the target. 1)

One of the key advantages of satellite SAR instruments is that radar signals aren’t impacted by lighting conditions and can penetrate weather. This enables day/night observations that are largely unaffected by cloud, rain, or other weather conditions. SAR makes use of the movement of the satellite in orbit, taking repeated observations as the instrument moves along its track, synthesising a much larger antenna and providing detailed, high-resolution images of the ground track. Further information on the observed environment can be obtained through the use of different wavelengths with different penetrative properties. 1) 2) 3) 4)

Most SAR instruments are limited to only sensing in one wavelength at any time. Five wavelength bands are used in current and upcoming radar land surface imaging satellite systems: P-band (~69.0 cm), L-band (~23.5 cm), S-band (~9.4 cm), C-band (~5.6 cm), and X-band (~3.1 cm) (Table 1; Figure 1). Typically, these signals will mainly reflect off of objects of similar size or larger than their wavelengths allowing for different properties to be observed by different wavelengths. Depending on the observation target, different bands have different strengths and weaknesses. For example, longer bands, such as L- and P-bands, are capable of providing information on the below canopy structure of forests, but are less sensitive to grasses and low vegetation. Shorter bands, on the other hand, are more useful for imagery of grasslands, savannahs, croplands, and pasturelands. In combination, imagery in these bands can provide substantial amounts of information about the Earth and its processes. 1) 5)

Figure 1: Demonstration of the interaction between different SAR bands and a forest canopy.  (Image credit: NASA)

Another way in which SAR imagery gathers information is in the polarisation of the emitted radiation. Signals are emitted in either horizontal (H) or vertical (V) polarisation. Upon return, backscatter may match its emitted polarisation (co-polarisation or like-polarisation), or flip to the other (cross-polarisation). These instances are denoted as HH, HV, VV, or VH. The first letter of these notations refers to the initial polarisation of the emitted signal, while the second letter refers to the polarisation of the returned signal. For example, in VH imagery, the emitted signal was vertically polarised, and the received signal was horizontally polarised. The nature of the ground surface determines whether a signal is backscattered in cross- or co-polarised state, and so often a SAR imager will receive both cross- and co-polarised images. The differences between these images can provide useful information about the landscape. 1)

The way in which backscatter occurs is of great importance to interpreting information about SAR images, and there are several recognised backscatter mechanisms: direct, forward, diffuse, double-bounce, and volume backscatter.

Direct backscatter refers to the direct reflection of radar signals back towards the sensors (Figure 2A&B). This can occur on flat, reflective surfaces that are facing the satellite, such as rocky outcrops or leaves (for short bands), and produces a strong co-polarisation, appearing bright in SAR imagery.

Forward scattering occurs when a flat, reflective surface is not directly facing the satellite, and the signal bounces away (Figure 2C).

Diffuse scattering occurs on rough surfaces, such as ploughed fields or rough water, where the radar signals are scattered in all different directions (Figure 2D&E).

Double-bounce scattering occurs when the radar signal bounces off of one surface, travels to a perpendicular surface, and then bounces back again towards the radar (Figure 2F). This can be on a rough surface, termed double-bounce diffuse type (Figure 2G), or on a smooth surface, called double-bounce specular type (Figure 2H). In double-bounce scattering, the image appears upside down.

Finally, volume scattering occurs when a radar signal is reflected multiple times within a 3-dimensional object (Figure 2I). For example, a signal may enter a tree canopy and bounce off of several leaves and twigs before ultimately bouncing back towards the satellite. This results in a mix of co- and cross-polarisation. All of these backscatter mechanisms often work simultaneously, and should all be taken into account when analysing SAR imagery. 1)

Figure 2: A) Direct backscatter from a rocky outcrop; B) Direct backscatter from a tree canopy; C) Forward scattering on a smooth surface; D) Diffuse scattering on a ploughed field; E) Diffuse scattering on rough water; F) Double-bounce scattering in an urban environment from a building to a road; G) Double-bounce diffuse type scattering from a tree to a ploughed field; H) Double-bounce specular type scattering from a tree to a smooth water surface; and I) Volume scattering within the canopy of a forest.

Example Products

Product Designation

SAR data products are designated based on their acquisition mode, product type, and processing level. Acquisition modes are the patterns with which the satellites observe the ground and can include, for example: stripmap mode, where the antenna beam moves with the satellite at a fixed angle, generating long-swath images (Figure 3a); sliding spotlight mode, where the antenna is moved mechanically to allow a long observation of a specific area (Figure 3b); or interferometric wide swath, where the SAR beam is steered from backward to forward in the azimuth direction for each burst, avoiding scalloping and resulting in homogenous image quality throughout the swath (Figure 4). Acquisition modes are often denoted in shorthand. For example, for Sentinel-1, stripmap mode is denoted as SM, interferometric wide swath IW, extra wide swath EW, and wave mode WV. 5) 6)

Figure 3: a) Stripmap mode of a SAR satellite. The angle of observation is fixed, and the observation beam moves with the satellite, generating long swath images. Yellow represents the SAR beam, blue represents the observed area (Image credit: Synspective inc.).. b) Sliding spotlight mode of a SAR satellite. The angle of the SAR beam changes as the satellite moves through its orbit, allowing continuous observation of a certain area whilst it is within range of the satellite (Image credit: Synspective inc.). c) TOPSAR sub-swath acquisition showing the back and forth burst movement of the SAR antenna (Image credit: ESA)

Product types have three classifications depending on how much processing has been applied to the data: level-0, level-1, and level-2. Level-0 is compressed and unfocused SAR raw data, and all higher levels are produced from it. Raw level-0 SAR data contains all of the information captured by the satellite instrument, and must be processed in order to be used in any meaningful way. Level-0 data is transformed into level-1 products via the application of various algorithms. First is the pre-processing, which involves raw data analysis and internal calibration. Then comes doppler centroid estimation, where absolute DC estimates are made and polynomials are fitted. The third step involves focusing the data, where range and azimuth processing are applied. Finally, post-processing occurs, whereby the final outputs are produced. These level-1 outputs are important criteria for the designation of SAR products and include Single Look Complex (SLC) and Ground Range Detected (GRD) images. 5) 6)

SLC images are close to the original raw data that was collected, making them the highest fidelity SAR images. There is no projection applied and all of the original sensor measurements remain. Use of these images is complex, and so is generally restricted to those with much skill and experience using complex-image exploitation software. SLC images are used typically for automatic processing or for the application of personalised processing chains.

Figure 4: Binary representation of SLC images. On the left and right diagrams, green lines represent slant range and purple lines represent ground range (Image credit: ICEYE)

GRD images consist of focused SAR data that has been detected, multi-looked, and projected to ground range using the Earth ellipsoid model WGS84. Pixel values represent detected amplitude, phase information is lost, and the resulting product has approximately square resolution pixels and square pixel spacing with reduced speckle at a cost of reduced spatial resolution. 5)

Level-2 products are geolocated geophysical products derived from level-1 products. They typically have a specific approach in mind. For example, Level-2 Ocean (OCN) products of Sentinel-1 for wind, wave, and currents applications may contain the following geophysical components derived from the SAR data: Ocean Wind field (OWI), Ocean Swell Spectra (OSW), and Surface Radial Velocity (RVL). 5)

Figure 5: Surface Wind measurement (OWI) from Copernicus Sentinel-1 data demonstrating funnelling effect of wind over the strait of Gibraltar (Image credit: ESA)

Backscatter Polarisation

SAR beams are emitted in one of two polarisations (H or V) and return also in one of these two polarisations. Whether a SAR instrument images in HH, HV, VV, or VH, as well as which waveband the beam is in provides different information about the observed area. For example, HV or VH polarisations are generally dominated by volume backscatter, which indicates the type of land being observed. These can be beneficial in different areas. For example, in areas with tropical forest conversion to secondary growth and other land use areas, longer wavelengths such as L-band are better able to penetrate the forest canopy and return more useful information about forest structure. This allows for clear distinctions between land-use types. Within L-band, HV channels have greater sensitivity to bare soil areas and display a wider radiometric dynamic range than the HH channels. In this environment, shorter wavelengths such as C-band do not penetrate the forest canopy so easily and reflect back off the upper canopy similarly between forest types. Another example would be that of pastureland. As longer wavelengths penetrate through most vegetation, they do not easily detect grasses and crops. Shorter wavelengths do, and with the different structures and growth stages of vegetation also providing different responses at VV and VH polarisation, much information can be derived here.

Figure 6: L-band HH/HV (from ALOS-2) and C-band VV/VH (from Sentinel-1) imagery of a) dense tropical forest and b) Brazilian rangeland and pastureland demonstrating the benefits and disadvantages of using different wavelength bands for SAR imagery, as well as the differential information provided by co-polarised and cross-polarised imagery (Image credit: CEOS)

Related Missions

BIOMASS (Biomass Monitoring Mission for Carbon Assessment)

ESA’s BIOMASS will be the first P-band SAR instrument flown in space. The Antenna Feed Subsystem will provide quantifiable data on the global carbon cycle, enabling observations of topography beneath thick forest canopies and of features in deserts and ice sheets. The data will be used to provide scientific support for international treaties and agreements, improve predictions of landscape-scale carbon dynamics, provide observations to initialise and test the land element of Earth system models, reduce uncertainties in carbon flux, and provide key information for forest resources management. 7)

Read more

ALOS-2 (Advanced Land Observing Satellite-2) / Daichi-2

Launched 24 May, 2014, the JAXA (Japanese Aerospace Exploration Agency) satellite ALOS-2 (nicknamed Daichi) uses an imaging microwave radar called PALSAR-2 (Phased Array type L-band Synthetic Aperture Radar-2) to obtain L-band observation data of the Earth. PALSAR-2 operates at a minimum spatial resolution of 3 m in stripmap mode and 1 m x 3m in spotlight mode. Its main mission goals are for cartography, regional observations, disaster monitoring, resource surveys, and global forest monitoring. 8)

Read more

ALOS-4 (Advanced Land Observing Satellite-4)

ALOS-4 is the follow on mission for ALOS-2, and carries the improved PALSAR-3 instrument to provide higher resolution monitoring of infrastructure development, disasters, forests, and sea-ice. The satellite will also be capable of monitoring vessels in the ocean and heavy marine traffic areas. Launch is planned for 2024. 9)

Read more

SAOCOM (SAR Observation & Communications Satellite / Satélite Argentino de Observación COn Microondas)

SAOCOM is a constellation of two L-band satellites (SAOCOM-1A and SAOCOM-1B) owned and operated by CONAE (Comision Nacional de Actividades Espaciales). The mission was launched 7 October 2018 with the goal of providing effective Earth Observation and Disaster Monitoring. Identical instrument payloads are carried on board the twin satellites, consisting of the high-resolution, multi-purpose L-SAR. The obtained imagery is applicable to agriculture, fishery, forestry, weather, hydrology, oceanography, emergencies, ocean and land natural resources, urban mapping, and cartography. 10)

Read more

NISAR (NASA-ISRO Synthetic Aperture Radar)

A cooperative development between NASA and ISRO (Indian Space Research Organisation), NISAR is expected to launch early in 2024, and will carry two SAR instruments. One SAR will be in L-band and the other in S-band, making this the first satellite to use two different radar frequencies. The L-SAR will observe landscape topography and heavily forested areas while the S-SAR will monitor soil moisture, particularly in polar regions as S-band frequencies are less perturbed by the ionosphere. 11)

Read more


NovaSAR-1 was launched 16 September 2018 with the intention of testing the capabilities of a new low cost S-band SAR platform. The spacecraft and its payload were developed collaboratively by SSTL (Survey Satellite Technology Ltd.) and Airbus Defence and Space Ltd.. 12)

Read more


Launched 3 April 2014, Sentinel-1 is a constellation of two C-SAR imaging satellites operated by ESA. The spacecraft is used for monitoring of sea and land ice, surveillance of oil spills and ships, monitoring of marine wind and waves, monitoring of land surface motion risks, and the mapping of forest, water, and soil management. 13)

Read more


RADARSAT-2 is a C-band commercial radar imaging satellite operated by MDA (MacDonald Dettwiler Associates Ltd. of Richmond, BC) and CSA (Canadian Space Agency). Launched 14 December 2007, the satellite’s primary mission applications are to monitor and observe glacier cover, motion and topography, ice sheet and land surface topography, and sea-ice cover and type. 14)

Read more

RADARSAT Constellation Mission (RCM)

Developed and operated by CSA with private sector support, primarily from MDA, RCM is a three satellite constellation Earth observation mission launched 12 June 2019. Each satellite carries the specialised C-band SAR-RCM (Synthetic Aperture Radar for RADARSAT Constellation Mission) sensor module. The primary mission objectives of RCM are the monitoring of glacier cover, motion, and topography; ice sheet and land surface topography; and sea-ice cover and type. 15)

Read more

TSX (TerraSAR-X) and TDX (TanDEM-X)

TSX and TDX, operated by DLR (German Aerospace Centre) and supported by BMBF (German Ministry of Education and Science), provides high-resolution X-band data for use in climatology, environmental and disaster monitoring, hydrology, geology, oceanography, and cartography. TSX was launched on 15 June 2007, while TDX was launched on 21 June 2010.  16) 17)

Read more - TSX

Read more - TDX

COSMO-SkyMed (Constellation of Small Satellites for Mediterranean basin Observation)

COSMO-SkyMed is a four spacecraft constellation developed by ASI (Agenzia Spaziale Italiana) and funded by MUR (Italian Ministry of Research) and MoD (Italian Ministry of Defence) for the purposes of both military and civil (research and commercial) data use. Each satellite’s SAR-2000 instrument observes in the X-band under all weather and visibility conditions at high resolution and in real time. COSMO-SkyMed-1, -2, -3, and -4 were launched 8 June 2007, 9 December 2007, 25 October 2008, and 6 November 2010, respectively. 18)

Read more

CSG (COSMO-SkyMed - Second Generation)

A 4 satellite constellation aimed at enhancing the quality and capability of imaging of the original 4-spacecraft COSMO-SkyMed Constellation. Developed by ASI and funded by MUR and MoD, the first two satellites were launched in 2019 and 2022, with the remaining two due to be launched in 2024 and 2025. Each spacecraft carries a COSMO-SkyMed Second Generation Synthetic Aperture Radar (CSG-SAR), that provides X-band imagery. 19)

Read more

RISAT-2 (Radar Imaging Satellite-2)

Launched 20 April 2009, RISAT-2 is an X-band SAR reconnaissance satellite of ISRO, making it India’s first satellite with a SAR instrument. The satellite possesses 24-hour, all-weather monitoring capability and its data has significantly enhanced ISRO’s capability for Earth observation, particularly for the management of disasters and environmental and urban mapping. The X-SAR data is used for oil slick detection and flood, landslide, agriculture, forestry, and urban mapping, but it is also used specifically for management of natural disasters such as cyclone assessments. 20)

Read more 


1) “A Layman’s Interpretation Guide to L-band and C-band Synthetic Aperture Radar Data,” CEOS & GFOI (Global Forest Observations Initiative), May 2023, URL:

2) “What is Synthetic Aperture Radar (SAR)?” Pathfinder Radar ISR & SAR Systems, Scandia National Laboratories, 2023, URL:

3) “What is Synthetic Aperture Radar?” NASA, URL:

4) “SAR 101: An Introduction to Synthetic Aperture Radar,” Capella Space, February 2020, URL:

5) “Product Types and Processing Levels,” ESA, URL:

6) “Synspective SAR DATA PRODUCT GUIDE,” Synspective Inc., August 2023, URL:

7) “Biomass (Biomass Monitoring Mission for Carbon Assessment),” ESA, November 2017, URL:

8) “ALOS-2 (Advanced Land Observing Satellite-2) / Daichi-2,” ESA, May 2012, URL:

9) “ALOS-4 (Advanced Land Observing Satellite-4), ESA, December 2022, URL:

10) “SAOCOM (SAR Observation & Communications Satellite),” ESA, August 2018, URL:

11) “NISAR (NASA-ISRO Synthetic Aperture Radar),” ESA, June 2018, URL:

12) “NovaSAR-1,” ESA, December 2012, URL:

13) “Copernicus: Sentinel-1,” ESA, September 2023, URL:

14) “RADARSAT-2,” ESA, June 2012, URL:

15) “RADARSAT Constellation,” ESA, June 2012, URL:

16) “TSX (TerraSAR-X),” ESA, June 2012, URL:

17) “TDX (TanDEM-X),” ESA, September 2016, URL:

18) “COSMO-SkyMed,” ESA, May 2012, URL:

19) “COSMO-SkyMed - Second Generation,” ESA, August 2014, URL: 

20) “RISAT-2 (Radar Imaging Satellite-2),” ESA, August 2023, URL: