Livrables
The main tangible scientific achievement of the algorithm work was the emergence of a new CHL retrieval algorithm with associated quality estimation algorithm. The REVAMP algorithm is a new water quality algorithm to retrieve concentrations from reflectance spectra in turbid coastal waters. Because of the optical complexity of such waters, frequently used simplified equations were abandoned in favour of the more sophisticated Hydrolight radiative transfer code.
This code can predict the observed remote sensing reflectance under any angle as a function of absorption and scattering within the water, taking into account the angular distribution of the downwelling radiance and the transmission through the air-water interface for a given wind speed.
The inversion of this model is accomplished by a non-linear optimization scheme (Levenberg-Marquardt). The concentrations are found by fitting the modelled remote sensing reflectance spectrum to an observed reflectance spectrum, by altering the concentrations of chlorophyll-a, total suspended matter and absorption by coloured dissolved organic matter.
Approximating the Hydrolight output with high degree polynomial functions substantially reduced the execution time of the algorithm. Because Hydrolight (and the derived sets of polynomial coefficients) are based on total absorption and scattering, and not directly on concentrations, the calibration with different specific inherent optical properties is easy and completely transparent. Apart from the concentrations, the algorithm also calculates a statistical measure of the so-called goodness-of-fit and standard errors in the retrieved concentrations.
The REVAMP database contains data on Sea Surface Reflectance, concentrations of Chlorophyll-a, suspended matter and CDOM and the following optical properties: Phytoplankton absorption, particle absorption and particle backscattering and CDOM absorption. The data were collected in various parts of the North Sea, the Celtic Sea and the English Channel.
There are three batches of data:
- Historic data (also from other programmes such as COASTLOOC, COLORS and MULTICOLOR) have been collected in a common format
- REVAMP data from institute-managed campaigns collected in 2002 and
- REVAMP data collected in 2003.
All data in the database are in principle in the MERIS band passes and all data was inputted and quality controlled using predefined standards. The database is quite large (for historical data in total for 15 single parameters 30.000 values were compiled;
In total 15 single parameters were catalogued, of which 30.000 measurement values were compiled. The database now contains, for 11 types of spectra, a total of 3.300 measured spectra. These data, primarily intended for algorithm selection and validation, now represents a valuable resource for increasing data mining.
Current status:
The REVAMP database is basically a relational database system. Some additions are specifically made for REVAMP as a subset of tables and a special user interface allowing data-retrieval in different formats. It is accessible via Internet by means of standard browsers with a java implementation. The address is:
http://www.mumm.ac.be/datacentre/Databases/REVAMP/
Dissemination and use potential:
The documentation is open to the public. For data retrieval, a user-ID/password combination has been provided to projects-partners.
Already a number of publications have used the information in the database and there are still some major publications in preparation or already submitted.
Two years after the closing date of the project this database will open up for the public, until then it is only accessible for REVAMP researchers in order to protect IP-rights.
After opening up for public access the database will provide a wealth of information to case-2 water researchers in Europe and worldwide.
Key innovative features:
The database is unique in the world since it contains data from almost 10 years of research in coastal waters, with parameters that are difficult to measure because of the techniques and logistics involved.
Expected benefits:
Commercial exploitation of the database is not expected. Scientific exploitation possibilities are large.
The REVAMP MERIS Regional CHL-Products Processor (MEP) is a software program that enables to implement algorithms (for the processing of MERIS images) and to validate and compare these. The concept is implemented in such a way that the MEP is a complete software environment in itself, which facilitates to plug-in algorithms in a very simple way. This has the advantage to researchers that they only have to develop Java code for the algorithm itself (or even simpler) to develop a DLL file containing the algorithm.
The advantages of this approach are that there is a clear and strict separation within the MEP of processor kernel and algorithm classes. There are two versions of the MEP, namely a scientific processor and a production processor. Both processors were build using the same building blocks and software concepts. The philosophy was that a software environment should be created to import MERIS images, flags and auxiliary data, and to output REVAMP CHL maps, flags and error estimation products. This environment was first used as a test bed for various candidate algorithms (scientific processor) and subsequently to implement the preferred algorithm for mass map production (production processor).
The MEP environment was quality controlled and harmonised to ensure robust software. The Processor was implemented in Java using the BEAM Application Programming Interface (API) (BEAM = Basic ENVISAT AATSR and MERIS Toolbox). The BEAM software can be used to analyse and visualise standard MERIS products and is available free of charge from ESA. It is a widely used standard for working with MERIS data. Further standardisation of the product format resulted in the choice for the BEAM DIMAP format for the MEP which is readily available through the API. The MEP can operate on specific files containing MERIS type of spectra (to allow testing of algorithms on close range SSR-observations) and on MERIS images. The program can be operated through a GUI and in batch mode.
Dissemination and use potential:
The MEP scientific version will be made available to a wide audience of scientists who can use the software to experiment with algorithms to extract information from MERIS images. The MEP production processor (exclusive the REVAMP algorithm which is described as a separate result) can be used by REVAMP partners for regional, local and operationalisation studies. The Consortium Agreement regulates use within the consortium and external use.
Key innovative features of the result:
The MEP is unique in its plug-in design, which allows developing algorithms for a certain sensor (MERIS in this case) without having to know any details of file formats, IO-formats etc. The production processor including the REVAMP preferred algorithm is a unique and innovative tool to rapidly produce hundreds of MERIS images into Chlorophyll-a maps.
Current status
Both versions of the MEP are residing with all the project partners.
Use of the result and its expected benefits.
It is expected that the MEP operational processor will play a role in Chlorophyll-a mapping in the settings of GMES and GSE projects. The MEP will be used by project partners for further research and to produce local products based on an insight in the local optical properties, hence improving the regional REVAMP products.
Recherche de données OpenAIRE...
Une erreur s’est produite lors de la recherche de données OpenAIRE
Aucun résultat disponible