Objetivo
THE HELP QUANTIFY THE ROLE OF OCEAN CIRCULATION IN THE CO2-BUDGET.
A method for classifying remotely sensed data involving a segmentation based training system has been developed.
A methodology has been developed, involving a segmentation based training programme, which overcomes some of the problems involved in image classification of remotely sensed data. By using this system it has been demonstrated that significantly better performance can be achieved than with existing maximum likelyhood classification techniques. This methodology is embodied in a suite of FORTRAN based computer programs which have been developed in a video home system (VHS) environment on a VAX 11/750 computer.
DEEP-SEA SEDIMENT CORES FROM THE EAST AND WEST PACIFIC OCEAN WILL BE SELECTED FOR STUDY BY STABLE ISOTOPE ANALYSES. INITIALLY THE UPPER PART OF SEVERAL ALREADY IDENTIFIED CORES (V19-25, V24-110, ODP677) WILL BE SAMPLED TO SELECT THE MOST SUITABLE FOR FURTHER WORK.
SEDIMENT SAMPLES WILL BE SIEVED TO FACILITATE EXTRACTION OF THE FORAMINIFERAL SHELLS FOR ANALYSIS; FORAMINEFERA WILL BE EXTRACTED USING STANDARD MICROSCOPE METHODS; THE SELECTED SPECIMENS WILL BE ANALYSED MASS SPECTROMETRICALLY FOR THEIR OXYGEN- AND CARBON- ISOTOPIC COMPOSITION.
SPECIMENS WILL ALSO BE EXTRACTED FOR RADIOCARBON ANALYSIS.
BY THE END OF THE FIRST YEAR TWO ADDITIONAL PACIFIC DATA SETS WILL HAVE BEEN GENERATED, AND CORED FROM THE TROPICAL ATLANTIC (AND PERHAPS INDIAN) OCEAN WILL HAVE BEEN SELECTED.
ALSO DURING THE FIRST YEAR EXTENSIVE DATA EXCHANGE BETWEEN GIF-SUR-YVETTE AND CARMBRIDGE WILL TAKE PLACE SO THAT THE TWO GROUPS CONVERGE ON THE BEST TIME-SCALE AND CORRELATION METHODS TO GENERATE THE NECESSARY TIME-SERIES FOR MODELLING THE HISTORY OF OCEAN CIRCULATION AND PALAEOCHEMISTRY CHANGE.ALSO DURING THE FIRST YEAR THE MODELLING GROUP AT HAMBURG WILL BE VISITED TO ENSURE THAT THE MODELLING EFFORT THERE IS OPTIMALLY LINKED TO THOSE PARAMETERS THAT CAN BE RECONSTRUCTED FROM THE GEOLOGICAL DATA.
DURING THE SECOND YEAR CORES FROM THE TROPICAL ATLANTIC OCEAN WILL BE ANALYSED. BY THIS STAGE THE MODELLING WILL HAVE PROGRESSED FAR ENOUGH SO THAT MODEL PREDICTIONS MAY INFLUENCE THE CHOICE OF CORES FOR ANALYSIS.
Ámbito científico (EuroSciVoc)
CORDIS clasifica los proyectos con EuroSciVoc, una taxonomía plurilingüe de ámbitos científicos, mediante un proceso semiautomático basado en técnicas de procesamiento del lenguaje natural.
CORDIS clasifica los proyectos con EuroSciVoc, una taxonomía plurilingüe de ámbitos científicos, mediante un proceso semiautomático basado en técnicas de procesamiento del lenguaje natural.
- humanidadeshistoria y arqueologíahistoria
- ciencias naturalesinformática y ciencias de la informacióninteligencia artificialvisión artificialreconocimiento de imágenes
- ciencias naturalesciencias físicasópticamicroscopía
- ciencias naturalesinformática y ciencias de la informaciónciencia de datosintercambio de datos
Para utilizar esta función, debe iniciar sesión o registrarse
Tema(s)
Data not availableConvocatoria de propuestas
Data not availableRégimen de financiación
CSC - Cost-sharing contractsCoordinador
CB2 1TN CAMBRIDGE
Reino Unido