ESA > Join & Share > Technology Projects > MASS Project

MASS Project

Project Title   Multi Application Support Service System
Project Acronym   MASS
Contractor(s)   SPACEBEL (B), Intecs HRT (I), Telespazio (I), Kongsberg Spacetec (N), GIM (B), VITO (B)


Project Objectives
          How it Works


Objectives   Top

The MASS project will provide a demonstrator for an e-Business architecture supporting integration of end-to-end services in the Earth Observation (EO) activities. The main entry point of this architecture is an Internet Portal where users, partners, service providers and data providers will be integrated in a coherent supply chain.

At the longer term, MASS should become an ESA standard Internet interface to EO thematic applications. It will support cost effective and easy production of thematic products that integrate EO products and geodata from different providers. MASS is a collection of functionality, infrastructure components and gateways, that provides an open service-oriented and distributed environment among business users (service users and service providers). Via MASS, users and service providers can interact automatically and dynamically to exchange services.

MASS is aimed at providing an opportunity for improving the market expansion and penetration of existing or prototyped Earth Observation (EO) products and services, also into the Geographic Information Systems (GIS) world, through an enabling, open environment for Service Providers and potential users. MASS will also give to the European development and service industry the opportunity to take a leading role in the installation, maintenance and operation on request of personalised systems and services related to the future EO-related Business-to-Business (B2B) market.

The main objectives of MASS are to:

  • provide an infrastructure enabling the interactions among Service Providers and with users,
  • make basic or end-to-end Services visible via MASS,
  • allow basic or end-to-end Services to remain on the service provider infrastructure,
  • allow for easy plug-in and plug-out of Services to/from the MASS environment,
  • allow chaining of Services into more complex ones,
  • support “subscription??? type Services (e.g. fires active monitoring and alerting),
  • support the evolution and maintenance of services,
  • allow easy identification of, and access to requested services, with progress follow-up until completion.



Context   Top

The general context around the MASS system includes users and existing EO services.


We can identify several categories of MASS users:

  • Anonymous users can order MASS services which are free of charge.
  • Registered users are users whose identity is known by MASS. They can request quotations for services and order paying services.
  • Service providers are allowed to make their services available through the MASS Portal. They can also combine existing services using the MASS workflow capabilities.
  • The MASS administrator and help desk manage the MASS Portal and publish news items on the MASS Portal.


MASS places no restrictions on the granularity of a (Web) service that can be integrated. The grain size can range from small (for example a component that must be combined with others to create a complete business process), to large (for example an application). We can identify two main categories of MASS services:

  • Basic Services: are limited services running on the service providers local infrastructure. They are connected to MASS over the Internet using the SOAP over HTTP protocol.
  • Complex Services: are services consisting of a combination of Basic Services. Complex Services can be modelled by a service provider using the graphical workflow definition tools provided by the MASS system. Complex services can comprise Basic Services of multiple service providers.


Architecture   Top

MASS is a fully distributed and highly scalable system, based on open standards such as J2EE, XML, SOAP, WSDL, XSLT etc. It consists of the following main components: MASS Portal, Access Points, User Terminals and Service Provider Terminals. The MASS Toolbox facilitates the service provider's task to integrate his service with MASS.

MASS Portal

The MASS Portal is based on the BEA WebLogic 6.1 application server, uses the Apache Struts framework and JavaServer Page technology and runs on a SUN Ultra10 with the Solaris 7 operating system. Persistency is implemented using Oracle 8.1.7 RDBMS. The BEA WebLogic Integration 2.1 product is integrated to provide sophisticated workflow capabilities allowing service providers to chain services. A simplified representation of the MASS Portal design is depicted below.




User Terminals

Users need to be connected to the Internet to use MASS. They can access the MASS Portal via a regular Web browser. No special software is required. If the user wishes to use services that require input images, then he should also have an FTP server that can be accessed by MASS.

Service Provider Terminals

Service Providers can access the MASS Portal via a regular Web browser. To create service workflows, they should run the Workflow Studio application that can be downloaded from the MASS Portal. This is a Java application that can be run on any platform having a Java Virtual Machine, e.g. Windows NT, Windows 2000 or Linux. 128 MB of RAM is required as a minimum.


The MASS Toolbox is a utility that will be provided to service providers willing to integrate their service into MASS. It will be a configurable tool implementing the SOAP interface required by MASS. It will provide multiple mechanisms to facilitate connecting MASS to the service backend processes running on the service providers local systems. The toolbox will run on any platform having a Java Virtual Machine, thus a large selection of platforms will be supported. The toolbox will only contain open source components, so no expensive software licenses are needed.



Access Points

An access point is a simplified version of the MASS Portal. It provides all functionalities of the MASS Portal, except workflow capabilities and some local Functions. The access points are designed using open source software only. The main components are the JBoss J2EE application server and the MySQL RDBMS. The access points can be installed on any platform having a Java Virtual Machine, e.g. Windows NT, Windows 2000, Solaris or Linux.


How it Works   Top

Service Providers are able to build services consisting of reusable building blocks already made available on the MASS Portal by other service providers, using a graphical workflow definition tool (i.e. the Workflow Studio component of BEA Integration 2.1). The XML inputs/outputs of the different building blocks are chained using XML, XPath and possibly XSL. The XML output of a service may refer to an FTP server where the result may be retrieved, or the result may be rendered in a Web browser, using XSL stylesheets. The reusable building blocks (i.e. services) are implemented as Web services and the MASS Portal accesses them via SOAP. Each service therefore has to publish as a minimum sendRequestForQuotation and sendOrder function, which are called from inside workflow definitions. These functions have to respect an imposed XML Schema.


Output   Top

The MASS project is organised in three Phases. In each Phase, the MASS Portal is extended and upgraded, and additional services are made available. The integration of the following services is foreseen:

In Phase 1 of the project, Telespazio's FIERS fire hazard mapping service was integrated in MASS.  To maximise reusability, also in other services, the service was decomposed in three separate reusable basic services, i.e. a Risk Index Processor service, a Normalised Differential Vegetation Index (NDVI) Processor service and a Land Surface Temperature (LST) Processor service were integrated in MASS.

Service A1: NDVI


NDVI (Normalized Differential Vegetation Index) is one of the most important information that can be extracted from visible and infrared images about vegetation, its health and its phenological status. Time analysis of NDVI variations over one or more seasonal cycles can be used for vegetation species identification and growth monitoring.

NDVI maps can be produced either once a day using low resolution (1 km pixel size) sensors (such as NOAA-AVHRR or TERRA-MODIS) or less frequently (once every 15 days) but with better resolution (e.g. with Landsat data, 100 m pixel size).

The provision of this layer can be considered as a "basic service", in the sense that some users can be interested in acquiring only this single product (e.g. NDVI maps can be useful to those interested in vegetation status monitoring).

Service A2: LST Map


Land surface temperature can be measured with Thermal Infrared channels available on some satellites (NOAA/AVHRR, MODIS and Landsat with different spatial resolution and revisit time).

Long term measurement can be useful for research institutes involved in global radiation budget measurements.

Local authorities, and mainly urban areas authorities, responsible for air quality measurement and management need this information which represents an essential input datum for pollutant transport model.

LST maps can be produced once a day using low resolution (1 km pixel size) infrared sensors (such as NOAA-AVHRR or MODIS).


Service A3: Fire Risk Map


As identified at present, the end-to-end service A consists of supplying daily forest fire risk index maps for the entire national or regional territory during the summer season.

The real time forest fire risk index is generated from daily AVHRR data (Vegetation Indexes and Surface Temperature ), at 1x1 km resolution obtained by NOAA satellite.

Landsat repetition time is not comparable with the daily NOAA data, as it should be for a risk index generation in order to prevent damages. In MASS project, risk index by Landsat will be generated just to simulate the possible future service that will be possible to realize by means of the next satellites generation just as Cosmo-Skymed.

The possibility to have a look on the status of the vegetation and its hydrological stress every day or its trend, by the analysis of the last 5-10 days, provide a very useful tool to better manage the men and the means available for fires struggle.

In Phase 2, additional services from Kongsberg Spacetec, GIM, VITO and Spacebel will be integrated in the MASS system. These services are briefly presented below:

Service B: DISP Interferometric Processor

The SPACEBEL interferometry chain implements an interferometric and a differential interferometric SAR processor, also called the DISP InSAR/DInSAR processor. It will be decomposed in 3 basic services: B1: Interferometric coherence map, B2: Digital Elevation Model (DEM) single pair and B3: Combination of Ascending/Descending DEM. These services use as input, ERS-1/2 Annotated Raw Data, ERS-1/2 SLC-I Data or ENVISAT ASA-IMS-1P Data.



Service C: ICE Services

GIM's Intelligent Geodata Conversion and Extraction Services (ICE) consist of 6 different basic services. Once plugged into MASS, each of them can be used independently from the others. Any end-to-end service that is plugged in MASS (via a Remote Service Provider RSP) has the possibility of using the ICE basic services for customised data delivery to the end-user. Each of the following ICE services can also be used as a stand-alone service on MASS:
  • Selection Area of Interest (AOI)
  • Co-ordinate transformation for vector and raster data
  • Data format conversion for vector and raster data
  • Generalisation of vector data
  • Resampling of raster data
  • Clipping of Area of Interest
Service D: Ship and Oil Spill Detection

Kongsberg Spacetec's oil spill detection and monitoring is based on the use of SAR images (RADARSAT and ENVISAT). To make the system also suitable in coastal areas system starts by masking out land areas. Potential oil spills are detected as dark spots in the images, and are validated through a post processing step. The output is a list with potential oil spills with characteristics. In some cases, depending on the user’s requirements, a quick-look of the SAR image is produced as well. Overlaid on the image there is possible to add important geographic information such as oil platforms and pipelines, which can be used to identify possible sources of the oil spill.

Kongsberg Spacetec's ship detection system is based on the use of SAR images (RADARSAT and ENVISAT). To make the system also suitable in coastal areas system starts by masking out land areas. Then a search for ship targets is performed followed by wake search around the detected ship candidates. The ship detection module will be based on a operational prototype of an automatic ship detection system developed by the Norwegian Defence Research Establishment.


Service E: VITO

Six basic services from VITO will be made available through MASS. These services will be integrated using the MASS Toolbox, and experiences during the integration will lead to improvements of the Toolbox.

  • S10 Data Service: This data service will give access to products in the SPOT VEGETATION S10 archive, that are at least 3 months old (See http://free.vgt.vito.be).
  • Net Primary Productivity: Net Primary Productivity or NPP products represent repeatable estimates of the net flux of carbon between the atmosphere and terrestrial vegetation. It quantifies these carbon fluxes by combining a simplified carbon exchange model with satellite observations SPOT-VEGETATION S10 images. Some applications of NPP are linked to the Kyoto protocol, to analyse whether a vegetation area is a carbon sink or a carbon source, to analyse the vegetation damage due to climatological anomalies (late frost, long period of drought,...) and environmental pollution (ozone damage, heavy metal pollution,...).

  • Net Ecosystem Productivity: The Net Ecosystem Productivity (NEP) is based on NPP but takes also into account the respiration losses of the soil. In this way, NEP is an estimate of the carbon fluxes between the ecosystem and the atmosphere.


  • Dry Matter Productivity: These 10-daily DMP-images are very similar to the Net Primary Productivity products (NPP). They are also based on the classical Monteith-model which combines the remote sensing imagery (NDVI converted to fAPAR) with meteorological data in order to obtain estimates of the productivity of the terrestrial vegetation.
  • Desert Locust: The desert locust is a common threat to agriculture, subsistence farming and vulnerable pastures. To prevent plagues from desert locust, early detection and immediate control of locust populations are required. Satellite imagery, like SPOT VGT S10, is used to detect potential development of desert locust habitats.


  • Small Water Bodies: The Global Vegetation Monitoring Unit of JRC has developed a method to map and monitor small water bodies in arid regions using low resolution satellite imagery from the VEGETATION instrument. Although this instrument provides images of the Earth with a 1-km resolution, it was demonstrated that it is sufficient to accurately map and monitor the presence of water in natural and artificial ponds and swamps of about 1 km² in size. The accuracy obtained is in the order of 90%, i. e. 90% of the water bodies detected with the method over an area of 1 million 1 km² were found to be correct.


Contributors to this page: andreadv .

Page last modified on Wednesday 22 of December 2010 15:36:16 CET by andreadv.