BEAM and the Sentinel Exploitation Tools
Fomferra, Norman1; Danne, Olaf1; Embacher, Sabine1; Fincke, Tonio1; Peters, Marco1; Quast, Ralf1; Veci, Luis2; Zühlke, Marco1; Storm, Thomas1; Brockmann, Carsten1
1Brockmann Consult GmbH, GERMANY; 2Array Systems Computing Inc, CANADA

The ESA BEAM project was kicked-off 10 years ago. Its initial objective was to provide a set of open source tools and application programming interfaces for the exploitation of remotely sensed data products of Envisat’s MERIS, AATSR and ASAR instruments. After its first release, the BEAM toolbox was very well received by data users and also software developers. While being further developed over the years, 24 versions have been released so far, and the former Envisat toolbox turned into a generic Development Platform for Earth Observation data visualisation, analysis and processing. Due to its open and modular architecture and its extendibility through plug-ins BEAM comprises a large number of tools supporting a variety of Earth Observation sensors and data formats. Today, BEAM counts a few thousand users and is actively developed and maintained by Brockmann Consult. External collaborators contribute to the BEAM source code and external organisations provide additional plug-ins.

The main components of BEAM are:

  • VISAT - An intuitive desktop application used for EO data visualisation, analysing and processing.
  • A set of scientific data processors running either from the command-line or invoked by VISAT.
  • A common command-line interface (gpt: graph processing tool) is used to execute processing graphs made up of operators nodes developed using the BEAM GPF (see below).
  • A data product converter tool allowing a user to convert raw data products to the BEAM-DIMAP standard format, to NetCDF, HDF, GeoTIFF, or RGB images.
  • The Development Platform, a rich Java API which provides ready-to-use components for remote sensing related application development and plug-in points for new BEAM extension modules.

    The most important extension points provided by the BEAM Development Platform are:

  • Product I/O API - allows to extend BEAM for new data product readers and writers;
  • Graph Processing Framework (GPF) API – allows to rapidly develop new data processors (processing "Operators") and to combine them to processing graphs;
  • VISAT RCP - A rich client platform used to build efficient EO imaging applications for the desktop.

    The following features are provided by the BEAM/VISAT desktop application:

  • Very fast image display and navigation even of giga-pixel images;
  • Advanced layer management allows adding and manipulation of new overlays such as images of other bands, images from WMS servers or ESRI shapefiles;
  • Rich region-of-interest definitions for statistics and various plot types;
  • Easy bitmask definition and overlay;
  • Flexible band arithmetic using arbitrary mathematical expressions;
  • Accurate reprojection and ortho-rectification to all common map projections, geo-coding and rectification using ground control points
  • Store and restore the current session including all opened files, views and layers;
  • Mosaicing, Collocation and Level-3 Binning of satellite raster data;
  • EM and K-Means Clustering, Linear Spectral Unmixing, PCA, and a rich set of sensor dependent scientific data processors.

    The supported sensors are MERIS, AATSR, ASAR of Envisat, ATSR and SAR of ERS, ALOS, AVNIR-2 of PRISM, MODIS of Aqua and Terra, CHRIS of PROBA, AVHRR of NOAA-KLM and MetOp, VGT of SPOT, TM of Landsat and many more available as plug-ins. Furthermore BEAM supports opening files of a number of formats commonly used in the modelling and remote sensing domain such as GeoTIFF, NetCDF CF and HDF-EOS. The BEAM development platform is the basis of a number of very efficient EO data applications. Among those, the most prominent application is NEST, a specialisation of BEAM comprising many new readers, processors and GUI extensions dedicated to the exploitation of SAR data. NEST is developed by the Canadian company Array Systems Computing.

    BEAM has lately been enhanced by a number of efficient and user-friendly tools supporting validation activities. A number of new features are resulting from a fruitful collaboration with the NASA Ocean Biology Processing Group (OBPG). The OBPG group is developing SeaDAS, a comprehensive image analysis package for the processing, display, analysis, and quality control of ocean color data. The OBPG group is currently replacing the SeaDAS 6 frontend using BEAM’s VISAT RCP. The new SeaDAS 7 frontend will also offer a number of great new tools for SeaDAS users. All SeaDAS command-line processing tools (l2gen, l3gen, etc) will be callable from modern and intuitive user interfaces. The new SeaDAS 7 frontend will of course support all data products offered through the OPBG, namely MODIS, SeaWiFS, CZCS, OCTS and VIIRS.

    Recently, Brockmann Consult and Array have developed the first version of the Sentinel Exploitation Tools. The main objective of the development of Sentinel Exploitation Tools is to provide users a means to work with all standard data products generated by the sensors on board of the new Sentinel satellites which will be launched 2013-2015. This includes fast visualisation, comprehensive analysis and effective processing of the data. The requested way to achieve this is to extend the existing ESA toolboxes BEAM and NEST by dedicated reader plug-ins for the Sentinel user products. This way, the majority of the existing NEST (for the Sentinel 1 data products) and BEAM (for the Sentinel 2 and 3 data products) features are applicable to these data products. The Sentinel product readers for BEAM and NEST have been developed within the ESA SUHET Project (Sentinel Product Handbook and Exploitation Tools).

    In addition to a number of dedicated readers, the Sentinel 2 Atmospheric Correction Processor for Sentinel 2 MSI L1 data will be integrated into the BEAM toolbox. Developing efficient tools for the Sentinel data products is a challenging task and is due to the greatly increased data volumes. Single Sentinel data products are magnitudes larger than their Envisat counterparts, and they do provide much more information, including increased spatial and spectral resolution coupled with the provision of per-pixel annotations such as geo-location, sensing time, measurement uncertainties and other quality information. The Sentinel Exploitation Tools will make available the following data products for a maximum of existing BEAM and NEST visualisation, analysis and processing tools:

  • Sentinel 1: SAR L0, L1 SLC, L1 GRD, L2 OCN;
  • Sentinel 2: MSI L1B, L1C, L2A;
  • Sentinel 3: OLCI L1B, L2W, L2L; SLSTR L1B, L2W, L2L; SYN L2A; VGT P, S1, S10. The Sentinel Exploitation Tools will be released with BEAM v5 in autumn 2013.