Validation of Medium Resolution Imaging Spectrometer (MERIS) data products in selected European coastal basins

CIMA members are executing in collaboration with OCPortugal partners a set of investigations supporting the validation of MEdium Resolution Imaging Spectrometer (MERIS) data products in selected European coastal basins. Work developments are addressed in the future to the analysis of OLCI/Sentinel-3 deliverables.

 

Specific actions supporting the validation of MERIS data products include:

 

1) Development of bio-optical algorithms The objective is to implement Multi Layer Perceptron (MLP) neural networks for the retrieval of OC data products. Reference field measurements were produced within the framework of the Bio-Optical Mapping of Marine Properties (BiOMaP) led by the Joint Research Centre (JRC) of EC, Ispra, Italy.

 

 

The application of MLP algorithms to MERIS remote sensing images is addressed by CITI. The analysis of results has been undertaken as joint effort between CENTRIA and CITI. Specific investigations were executed for the Adriatic Sea, the Baltic Sea and the Black Sea to address:

a. dataset selection and quality assurance;

b. assessment of the algorithm performance and applicability; and

c. comparisons with standard MERIS products.

 

2) Field data collection A field campaign was carried out in the Atlantic off the Portuguese coast in spring 2011 on board of the Almirante Gago Coutinho research vessel (Portuguese Navy) managed by the Instituto Hidrográfico (IH).

 

 

 

In situ data collection acknowledged the participation of FCT/UNL jointly with research scientists of the Joint Research Centre of EU and the Oceanographic Institute of the University of Lisbon.

 

 

 

 

The objective was to collect full sets of ocean colour field data (in- and above-water radiometric measurements and analysis of water samples) with a view to better exploitation of remote sensing in the investigated region. Planned research objectives were pursued through three steps:

a. field measurements;

b. data processing and quality assurance; and

c. development of bio-optical algorithms for the Atlantic off Portugal.

 

 

 

3) Radiative transfer modeling MOX, a Monte Carlo simulation environment for modeling light distribution in the marine waters. CENTRIA members were responsible for the implementation of MOX functionalities, whereas CITI members for its execution on large supercomputers. Three research projects were accepted for granted access to the Milipeia supercomputer at the University of Coimbra, Portugal. 

 

 

 

MOX has been designed to generate two-dimensional representations of light distribution. These representations are used to produce virtual measurements. MOX simulations offer then the unique opportunity to compare radiometric products, with and without accounting for the variability introduced by the wave and tilt perturbations [5, 6, 7, 8].

 

Selected publications

T. Kajiyama, D. D’Alimonte, and J. C. Cunha, “A high-performance computing framework for Monte Carlo ocean color simulations,” Concurrency and Computation: Practice and Experience, 2015, under revision.

D. D’Alimonte, T. Kajiyama, and A. Saptawijaya, “Ocean color remote sensing of atypical marine optical cases,” IEEE Trans. Geosc. Rem. Sens., 2015, under revision.

D. D’Alimonte, , and T. Kajiyama, “Monte carlo simulations for investigating uncertainty budgets of above-water radiometric measurements,” Optics Express, 2015, in preparation.

B. Bulgarelli and D. D’Alimonte, Optical Radiometry for Oceans Climate Measurements, ser. Experimental Methods in Physical Sciences. Elsevier, 2014, ch. Simulation of in situ visible radiometric measurements, p. 43, iSBN-10::0124170110. [Online]. Available: http://­store.elsevier.com/­Optical-Radiometry-for-Ocean-Climate-Measurements/­isbn-9780124170117/­

D. D’Alimonte, G. Zibordi, T. Kajiyama, and J.-F. Berthon, “Comparison between MERIS and regional high-level products in European seas,” Remote Sensing of the Environment, vol. 140, pp. 378–395, 2014.

T. Kajiyama, D. D’Alimonte, and G. Zibordi, “Match-up analysis of MERIS radiometric data in the Northern Adriatic Sea,” IEEE Geosci. Remote Sens. Lett., vol. 11, no. 1, pp. 19–23, 2014.

D. D’Alimonte, T. Kajiyama, and G. Zibordi, “Monte carlo simulations of the sea-surface reflectance for above-water radiometric applications,” in ACRS 2013: 34th Asian Conference on Remote Sensing, Bali, Indonesia, 2013.

A. Saptawijaya, D. D’Alimonte, and T. Kajiyama, “Seawater-type based neural networks for ocean color data inversion,” in ACRS 2013: 34th Asian Conference on Remote Sensing, Bali, Indonesia, 2013, oral presentation.

T. Kajiyama, D. D’Alimonte, and G. Zibordi, “Regional algorithms for European seas: a case study based on MERIS data,” IEEE Geosci. Remote Sens. Lett., vol. 10, no. 2, pp. 283–287, 2013.

D. D’Alimonte, G. Zibordi, J.-F. Berthon, E. Canuti, and T. Kajiyama, “Performance and applicability of bio-optical algorithms in different European seas,” Remote Sens. Environ., vol. 124, no. 0, pp. 402–412, 2012.

T. Kajiyama, D. D’Alimonte, and J. C. Cunha, “Statistical performance tuning of parallel Monte Carlo ocean color simulations,” in Parallel and Distributed Computing, Applications and Technologies 2012, Beijing, China, December 2012, pp. 761 – 766. [Online]. Available: http://­dx.doi.org/­10.1109/­PDCAT.2012.125

D. D’Alimonte, G. Zibordi, J.-F. Berthon, E. Canuti, and T. Kajiyama, “Bio-optical Algorithms for European Seas: Performance and Applicability of Neural-Net Inversion Schemes,” JRC-IES Scientific and Technical Reports, Tech. Rep. JRC66326, 2011, available: http://­dx.doi.org/­10.2788/­56321. [Online]. Available: http://­publications.jrc.ec.europa.eu/­repository/­handle/­111111111/­22406

T. Kajiyama, D. D’Alimonte, and J. C. Cunha, “Performance prediction of ocean color Monte Carlo simulations using multi-layer perceptron neural networks,” in Procedia Computer Science, vol. 4, Singapore, 2011, pp. 2186–2195.

F. Mélin, V. Vantrepotte, M. Clerici, D. D’Alimonte, G. Zibordi, J. F. Berthon, and E. Canuti, “Multi-sensor satellite time series of optical properties and chlorophyll-a concentration in the Adriatic Sea,” Prog. Oceanogr., vol. 91, no. 3, pp. 229–244, November 2011.

G. Zibordi, J.-F. Berthon, F. Mélin, and D. D’Alimonte, “Cross-site consistent in situ measurements for satellite ocean color applications: the BiOMaP radiometric dataset,” Remote Sens. Environ., vol. 115, no. 8, pp. 2104–2115, August 2011.

D. D’Alimonte, G. Zibordi, T. Kajiyama, and J. C. Cunha, “Monte Carlo code for high spatial resolution ocean color simulations,” Appl. Optics, vol. 49, no. 26, pp. 4936–4950, 2010.

T. Kajiyama, D. D’Alimonte, J. Cunha, and G. Zibordi, “High-performance ocean color Monte Carlo simulation in the Geo-info project,” in Parallel Processing and Applied Mathematics, ser. Lecture Notes in Computer Science, R. Wyrzykowski, J. Dongarra, K. Karczewski, and J. Wasniewski, Eds., vol. 6068. Springer, Wroclaw, Poland 2010, pp. 370–379. [Online]. Available: http://­dx.doi.org/­10.1007/­978-3-642-14403-5_39 

Actions: 
Authors: 

Tamito Kajiyama

Tamito Kajiyama joined the Geo-Info project of the Center of Informatics and Information Technology, New University of Lisbon, in April 2008. His current research interest focuses on the design and implementation of a high-performance computing (HPC) solutions for Monte Carlo ocean color simulation.

Partners: 

Center of Oceanography (CO)

The Centro de Oceanografia (CO) is a multidisciplinary research unit of the national R&D system, whose main host institution is the Faculty of Sciences of the University of Lisbon (FCUL), the other host institution of this Centre being the University of Évora (UE).

The main research lines of the Centro de Oceanografia are the following: 

Centre for Marine and Environmental Research (CIMA)

By nature, CIMA a multi­disciplinary unit that explores the connections between environments, processes and human populations using scientific methods coupled with state­of­the­art modelling techniques. Apart from research, CIMA places a strong emphasis on training and education, as well as providing consultation services, digital applications and scientific information to the scientific and public domains. CIMA activity is focused on 3 general aims: 1. scientific research; 2. training at Masters and PhD level and 3.

Sagremarisco - Viveiros de Marisco Lda (SGM)

Sagremarisco Lda  is located  at Sagres on the extreme SW of the Iberian  peninsular. The company operates at the interface between industry, education and scientific research.   
Industry:  technical support for the local aquaculture industry and fishing fleet; also, provides facilities for marine trials of materials under development for maritime use.
Education: provides equipment and human resources for coastal field studies by both local and foreign universities, and other educational institutions.

Faculty of Science and Technology (FCT)

Faculdade de Ciências e Tecnologia (Faculty of Science and Technology; FCT) is one of the nine academic units of Universidade Nova de Lisboa (UNL). The FCT/UNL campus is located in Monte de Caparica and covers approximately 30 ha, with additional capacity expansion to 60 ha.  Founded in 1977, FCT/UNL is one of the most prestigious Portuguese public schools of science and engineering today, with a total enrolment of ca. 7500 students, of which nearly 1400 are postgraduate students (MSc and PhD).