Home Page
FAQ
Publications
Documentation
Software
Movies
Image Gallery
Image Data
Data Search
Derived Products
Related Links
Contact Us
SCP file site

Scatterometer Climate Record Pathfinder
ERS Enhanced Resolution Image Product Users Notes


This page is designed as a dynamic repository of relevant information to aid users of ERS-1/2 enhanced resolution image products.

Enhanced resolution images made from ERS-1/2 data use the Scatterometer Image Reconstruction (SIR) algorithm. This version of the algorithm does not incoporate a median filter and, for ERS-1/2, uses a 2-d Hamming window as the spatial response function for each beam. In the processing, a linear model relating sigma-0 and incidence angle is assumed, i.e. sigma-0(db) = A + B (theta - 40) where A is the "incidence angle normalized sigma-0" at 40 deg incidence in dB, B is the effective incidence slope of sigma-0 versus incidence angle in dB/deg, and theta is the incidence angle of the observation. This simple linear model is used in place of the gamma=sigma-0/cos(theta) as it more accurately represents the sigma-0 versus backscatter response over a wider range of surface and volume scattering conditions. The SIR algorithm makes images of A and B on an 8.9 km pixel grid. The effective resolution is estimated to be 20-30 km resolution, depending on region and sampling conditions. Raw ERS measurements have a quoted nominal resolution of 50 km on a 25 km sampling grid.

In producing ERS-1/2 SIR images, multiple passes of the spacecraft are combined to produce a higher spatial resolution (at a cost of reduced temporal resolution). The overlapped processing enables true resolution enhancement of the images with any "excess sampling" contributing to an improve SNR. Because the AMI SAR and scatterometer modes are mutually exclusive, the resulting coverage gaps, produce reduced quality images when the SAR mode is operated over the study area. In combining the multiple passes, sigma-0 is assumed to be independent of azimuth angle. While true for most areas, some azimuth dependence in sigma-0 has been observed in Antarctic firn, presumably due to sastrugi or snow dunes.

ERS-1/2 data is obtained from fast-delivery products sent to JPL. No recalibration has been applied. However, several times throughout the mission dataset, the calibration parameters were changed prior to in the tapes delivered to JPL. These can be observed as "steps" or "jumps" in the A or B time series. In order to make a more uniform data set, calibration corrections for these steps are being developed and will be available soon.

Images are produced in the BYU .SIR file format using the standard naming scheme. The files are gzipped to minimize storage and transfer requirements. The standard images are designed for land and ice observation and so are landmasked. However, .SIR format land mask files (containing 0 for ocean and 1 for land) are available for each standard region. .SIR format images containing "images" of the latitude and longitude of each pixel for each region are also available.