What is NISAR GCOV data?¶
Explain to me like a scientist who is new to SAR
GCOV is one of NISAR’s data science products. GCOV takes radar signals collected from NISAR and process them into images that are:
Geocoded, meaning they are assigned to their corresponding place on the Earth.
Terrain corrected, meaning mountains, hills, and other slopes will not distort the radar data.
GCOV data includes several different of radar layers. The diagonal covariance layers provide real-valued backscatter. Backscatter is the portion of a radar signal that is reflected back toward the satellite after interacting with the Earth’s surface. The strength of the backscatter depends on surface roughness, moisture, and the type of material being observed. Backscatter information is commonly used to describe surface roughness for the application of interest.
GCOV data also contains layers called off-diagonal covariance channels, which describe how different radar channels interact with each other, rather than how strong the received signals are. These layers can provide useful information about surface structure and orientation.
GCOV can contain radar data acquired in different polarization modes, including horizontal / vertical (H/V, available in L-band) and, in some cases, circular (RH/RV, available in S-band only). Different modes can be utilized to help capture different aspects of how the Earth reflects radar signals. This cookbook provides examples for L-band data only.
Explain it to me like an experienced SAR scientist
GCOV is one of NISAR’s main Level-2 products. GCOV data are radiometrically terrain-corrected RTC and geocoded (Data is placed onto a map using latitude/longitude, or another coordinate system, so each pixel matches a real location on Earth). Through RTC and a fixed map projection, GCOV removes topographic radiometric distortions and places all layers onto a consistent geographic grid.
GCOV uses a polarimetric covariance framework. The The diagonal covariance layers provide real-valued backscatter measurements, while the off-diagonal covariance channels contain more complex inter-channel correlations. These channels enable polarimetric decomposition, scattering characterization, and target structure analysis.
GCOV supports linear polarization. (H/V, available in L-band) and, at times, circular (RH/RV, available in S-band only). The covariance matrix representation varies depending on the acquisition mode, with full, hybrid, or reduced sets of covariance terms stored in the frequencyA/frequencyB groups. This cookbook provides examples for L-band data only.
What are the applications for GCOV data?¶
Explain to me like a scientist who is new to SAR
GCOV is useful because it removes many of the complexities that create challenges in a radar analysis. Given that GCOV data is already corrected for terrain and mapped to a coordinate system, one can immediately analyze patterns on the ground without needing to correct viewing angles, slopes, or satellite geometry. It gives you consistent, ready-to-use data across space and time.
In addition to its terrain-corrected imagery, GCOV includes extra radar layers that help scientists tell different types of surfaces apart. These additional layers, called off-diagonal covariance channels, can help identify whether something is rough or smooth, wet or dry, or whether the signal is coming from the surface, vegetation, or inside snow. GCOV may also include measurements collected with different polarization modes, which allow the imagery to highlight various ground features that interact with the ground in unique ways.
Some of its many uses include:
Tracking snow, ice, and freeze–thaw patterns
Monitoring forests, vegetation, and land cover change
Studying wetlands and surface water
Mapping agricultural fields and soil moisture
Detecting flooding, landslides, or burn severity
General Earth surface monitoring where backscatter is needed
GCOV provides reliable, consistent, analysis-ready imagery for science, mapping, and environmental monitoring.
Explain it to me like an experienced SAR scientist
GCOV data are terrain corrected and geocoded, so you can immediately analyze patterns on the ground without needing to correct viewing angles, slopes, or satellite geometry. It gives you consistent, ready-to-use data across space and time.
The presence of off-diagonal covariance channels allows GCOV data to aid polarimetric applications. These channels allow users to:
Evaluate scattering mechanisms
Assess vegetation structure and orientation
Distinguish surface vs. volume scattering
Analyze anisotropic or mixed scattering responses
Similarly, GCOV’s support for multiple polarization modes (linear and, at times, compact-pol) enables analyses and classifications that rely on polarization diversity.
Some of GCOV’s many uses include:
Tracking snow, ice, and freeze–thaw patterns
Monitoring forests, vegetation, and land cover change
Studying wetlands and surface water
Mapping agricultural fields and soil moisture
Detecting flooding, landslides, or burn severity
General Earth surface monitoring where backscatter is needed
GCOV offers an analysis-ready representation of surface scattering behavior that avoids the complexities of RSLC-level geometry while supporting basic mapping and advanced polarimetric techniques.
What data layers are included with a GCOV product?¶
Explain to me like a scientist who is new to SAR
A GCOV file contains:
Radar images that have already been corrected for hills, slopes, and terrain
Location information that tells you exactly where each pixel is on the map
Basic details about the data, such as when it was collected and which radar settings were used at the time of collection
Different frequency sections, depending on how the satellite was operating when the data was collected
Various polarization modes, commonly horizontal / vertical (H/V) and, depending on location and operation settings, circular polarization (RH/RV, S-band only)
All of this is stored in an HDF5 file, which is similar to a folder that contains organized, labeled pieces of data.
Explain it to me like an experienced SAR scientist
A GCOV granule includes:
Terrain corrected polarimetric covariance layers derived from RSLC data.
Geoloaction datasets (projection definition, coordinate vectors, pixel spacing)
Metadata describing acquisition timing, frequencies, polarization combinations, and processing parameters
Frequency-specific groups, (frequencyA and/or frequencyB, depending on acquisition mode)
Linear (H/V, available in L-band) or circular (RH/RV available in S-band only) polarization modes, depending on radar settings and location
Everything is organized hierarchically within an HDF5 file structure with groups, datasets, and attributes.