Information, tools and data related to the China-Brazil Earth Resources Satellite (CBERS) PDS on AWS
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Type Name Latest commit message Commit time
Failed to load latest commit information.


This repository describes the organization of China-Brazil Earth Resources Satellite (CBERS) data available on AWS and provide additional data and tools for its use.

Information about the CBERS program is available from INPE's site (in Portuguese).

Accessing CBERS data on AWS

Entry in Open Data on AWS is

All CBERS-4 MUX scenes are available from the beginning of the CBERS-4 mission.

Scenes from AWFI, PAN5 and PAN10 acquired from June/2018 are ingested daily. The complete archive for these cameras will be ingested no later than Sep/2018.

The images are pulled daily from INPE's catalog at and are converted to Cloud Optimized Geotiff format.

The data is organized in two s3 buckets, both located in the us-east-1 region.

cbers-pds bucket

The data are organized using a directory structure based on each scene’s path, row and acquisition date. For instance, the files for CBERS-4 MUX scene CBERS_4_MUX_20160518_163_128_L4 are available in the following location: s3://cbers-pds/CBERS4/MUX/163/128/CBERS_4_MUX_20160518_163_128_L4/

The naming convention used for the files is CBERS_{N}_{CAM}_{YYYYMMDD}_{PPP}_{RRR}_L{L}_BAND{B}.{tif|xml}, where:

  • N: mission.
  • CAM: camera identification: MUX, AWFI, PAN10M or PAN5M.
  • YYYYMMDD: acquisition date.
  • PPP: path, see more information on reference grid below.
  • RRR: row, see more information on reference grid below.
  • L: processing level:
    • 2: system corrected image, expect some translation error.
    • 4: orthorectified with ground control points.
  • B: band number
  • tif|xml: tif for Geotiff files, xml for metadata files.

cbers-pds is a Requester Pays bucket, which means that you can access it freely within the us-east region, but you will incur charges if you download it elsewhere

cbers-meta-pds bucket

The data is organized using a directory structure identical to the one shown above, the difference is that this bucket stores only quicklooks for each scene. Each directory contains a large and a small quicklook, for instance:

  • Large quicklook: s3://cbers-meta-pds/CBERS4/MUX/157/133/CBERS_4_MUX_20150607_157_133_L4/CBERS_4_MUX_20150607_157_133.jpg
  • Small quicklook: s3://cbers-meta-pds/CBERS4/MUX/157/133/CBERS_4_MUX_20150607_157_133_L4/CBERS_4_MUX_20150607_157_133_small.jpeg

This bucket also provides manifest files describing all scenes at its root level, in gzipped and plain format:

The plain csv file does not include a first line with the column ids.

SNS Topics

SNS messages are sent to topics when a new quicklook is available in the cbers-pds bucket. The quicklook generation is thelast step in the ingestion procedure.

The SNS topics are:

  • arn:aws:sns:us-east-1:769537946825:NewCB4MUXQuicklook for MUX quicklooks
  • arn:aws:sns:us-east-1:769537946825:NewCB4AWFIQuicklook for AWFI quicklooks
  • arn:aws:sns:us-east-1:769537946825:NewCB4PAN10MQuicklook for PAN10M quicklooks
  • arn:aws:sns:us-east-1:769537946825:NewCB4PAN5MQuicklook for PAN5M quicklooks