The NASA-Indian Space Research Organization (ISRO) Synthetic Aperture Radar (NISAR) Mission Team has released nine new NISAR sample data products—the first public release of NISAR Level 1 to Level 3 L-band products.
Now community members will be able to test accessing the data and metadata for each data product type and gain familiarity with the characteristics of the L-band data products (data and ancillary data layers, metadata and product specifications, noise levels, resolution, etc.). These sample data products are intended to help the user community prepare for managing NISAR’s large data volumes and to refine their processing pipelines as the archive grows.
The NISAR project is still undergoing early-stage refinement and the data products are not yet fully calibrated. Through processing of the global data, the project has revealed unique characteristics of this first-of-a-kind radar system and identified required algorithm updates. As a result, while some limitations and artifacts have been identified in the data, a number of these are expected to be improved in future releases.
Known data product limitations in this release:
- Radiometric banding across the swath. This is due to incomplete calibration, specifically:
- Inter-beam channel calibration: The on-board digital beamforming is controlled by amplitude and phase weights for each of the receive channels. These weights have not yet been updated based on diagnostic mode analysis. These adjustments are expected to be very small, with a minor improvement to signal-to-noise ratio (SNR) and phase variability. As can be seen in this release, the beamforming is working very well even without adjustments
- Antenna pattern calibration, which is still being refined
- Very low amplitude radiometric ripple aligned with the azimuth direction, only noticeable in some radar dark areas, and at a spatial scale of about 600 m and shorter. We are working to improve radiometric uniformity at this low level, but SNR may still exhibit some banding at this scale
- Polarimetric channel imbalance. Polarimetric calibration and estimation of channel imbalances in the presence of a highly active ionosphere have not yet been performed. These steps will follow the radiometric corrections described above
- Quad-polarimetric (QP) product noise layers: In the QP data products, the noiseEquivalentBackscatter layer for the horizontal transmit, horizontal receive (HH) polarimetric channel is incorrectly populated with zeros. The noiseEquivalentBackscatter layers for the horizontal transmit, vertical receive (HV) and vertical transmit, vertical receive (VV) polarimetric channels are populated with correct (uncalibrated) values
- The Geocoded Unwrapped (GUNW) interferogram data product has three known limitations:
- The wrapped interferogram layer within the GUNW data products is incorrectly georeferenced. This limitation does not apply to other layers in the GUNW product
- The boundary of the ionospheric phase layer has edge-effect artifacts. These artifacts will be minimized in a future release
- Interferogram generation does not yet use the full “rubbersheeting” algorithm to estimate local image distortions due to deformation. This capability is important for fast-moving areas in global production but is not critical for the present sample data products, which exhibit modest deformation. The algorithm used here for alignment of radar imagery (coregistration) is based on geometrical offsets (derived from imaging geometry of the NISAR acquisitions, orbit, and a digital elevation model), refined with a polynomial fit to the data-driven dense offsets computed from amplitude cross-correlation of the radar data
We look forward to sharing the upcoming release of a larger volume of global data products around the end of February 2026. Fully calibrated and algorithmically improved global data products are anticipated for release around May/June 2026.