Report on Multi-Sensor Merging and Dynamic Bio-Optical Algorithm Selection for the Adriatic Sea (Rapporti progetti di ricerca)

Type
Label
  • Report on Multi-Sensor Merging and Dynamic Bio-Optical Algorithm Selection for the Adriatic Sea (Rapporti progetti di ricerca) (literal)
Anno
  • 2008-01-01T00:00:00+01:00 (literal)
Alternative label
  • Frederic Melin, Rosalia Santoleri, Simone Colella, Gianluca Volpe (2008)
    Report on Multi-Sensor Merging and Dynamic Bio-Optical Algorithm Selection for the Adriatic Sea
    (literal)
Http://www.cnr.it/ontology/cnr/pubblicazioni.owl#autori
  • Frederic Melin, Rosalia Santoleri, Simone Colella, Gianluca Volpe (literal)
Http://www.cnr.it/ontology/cnr/pubblicazioni.owl#altreInformazioni
  • In the latter case, it is recognized that a single empirical Chla algorithm is unlikely to produce a data set with a consistent level of uncertainty across the optical diversity of the Adriatic Sea. The approach, based on the novelty detection technique, is to distinguish two dominant water types and dynamically apply an algorithm developed for each class. One type is defined by a globally distributed data set, taken as representative of open water conditions (Class 1), and the other type represents conditions encountered at a coastal site in the north Adriatic (Class 2). The final blended output is the weighted average of the results of the two associated algorithms, with weights defined by the probability of the input LWN to belong to aspecific optical class. This report first describes the methodology and its application to the Adriatic Sea. The resulting satellite products are validated for a site in the north Adriatic Sea. The coverage given by the merged LWN series is then documented. The classification of the Adriatic waters shows the dominance of Class 1 in most of the central and southern Adriatic Sea, whereas Class 2 is present mostly in the northern region and along the Italian coastline. A significant part of the basin is found novel with respect to both classes and corresponds to the areas of transition between the two water types. Finally the 10-year multi-sensor Chla time series is briefly illustrated. (literal)
Http://www.cnr.it/ontology/cnr/pubblicazioni.owl#descrizioneSinteticaDelProdotto
  • The potential offered by optical remote sensing (ocean colour) could be better exploited if the uncertainties associated with satellite data in coastal regions and marginal seas could be reduced. This is particularly true for the European waters that present a wide range of optical water types. This calls for new developments to address the challenges posed by the optical diversity and complexity of the European domain. In the framework of ECOOP, the Adriatic basin serves as a test bed for such new approaches. Specifically, the availability of contemporaneous satellite missions and the recent developments of advanced remote sensing techniques and algorithms will be integrated to generate an improved multi-sensor time series. In the framework of the project, the ocean colour task specifically aims at delivering a time series of quality assessed geophysical products covering the Adriatic Sea. These products include the spectrum of normalized water leaving radiance LWN( ), the absorption coefficients due to phytoplankton and to chromophoric dissolved organic matter (CDOM) and non-pigmented particles, the backscattering coefficient due to particles, and the concentration of chlorophyll a (Chla). The first step is the creation of a data set of multi-sensor LWN( ) using an optically-based merging technique. These are used to generate maps of inherent optical properties (IOPs) and Chla, using an semi-analytical bio-optical model and empirical band-ratio algorithms, respectively. (literal)
Http://www.cnr.it/ontology/cnr/pubblicazioni.owl#supporto
  • Altro (literal)
Http://www.cnr.it/ontology/cnr/pubblicazioni.owl#affiliazioni
  • ISAc CNR UOS Roma Institute for Environment & Sustainability EC – Joint Research Centre (literal)
Titolo
  • Report on Multi-Sensor Merging and Dynamic Bio-Optical Algorithm Selection for the Adriatic Sea (literal)
Prodotto di
Autore CNR
Insieme di parole chiave

Incoming links:


Prodotto
Autore CNR di
Insieme di parole chiave di
data.CNR.it