English   español  
Por favor, use este identificador para citar o enlazar a este item: http://hdl.handle.net/10261/140925
Compartir / Impacto:
Estadísticas
Add this article to your Mendeley library MendeleyBASE
Visualizar otros formatos: MARC | Dublin Core | RDF | ORE | MODS | METS | DIDL
Título

Addressing High Resolution modelling over different computing infrastructures (HPC, HTC, Cloud)

AutorAguilar, Fernando
Fecha de publicación2015
CitaciónDelft Software Days 2015
ResumenEutrophication in water reservoirs, resulting in algae bloom, is an increasing serious problem in many water reservoirs in Europe and in the whole world due to the increase of anthropogenic pressure (human activities, including also farming) and climate change (warmer summers favour algae bloom). The framework of this presentation is the collaboration of IFCA (center of Spanish National Research Council) with the SME Ecohydros. The problem addressed is the monitoring, modelling and prediction of algae bloom in a water reservoir. The prototype has been developed by ECOHYDROS, a consulting company working in the environmental area, led by a biologist. It has been implemented in the Cuerda del Pozo water reservoir, near Soria in Spain. The development is supported by the River Authorities (Confederacion Hidrografica del Duero), by a LIFE+ project (ROEM+). IFCA-CSIC has been collaborating in this initiative since 2008, along the FP7 project DORII. One of the goals of the project is alert the authorities when the water quality level is decreasing not only in real time, but also forecast the reservoir behavior using Hydrodinamic and Water Quality models provided by Deltares, in Delft3D software suite. To feed the model, a number of instruments and sensors have been deployed across the whole reservoir. Instrumentation is a key facto. The current setup includes a central platform that is installed in the middle of the water, and instrumented with meteorological sensors (wind, temperature, solar radiation, rain, etc), and water quality sensors (conductivity, temperature, dissolved oxygen, turbidity, pH, etc.). The water quality sensor probe is placed in a cage connected to a wincher system allowing vertical profiling (range 1-30 m. in depth) that is critical to monitor the evolution of the water stratification, clearly reflected in the thermocline curves). More complex instrumentation, including radiometers, spectrometers and absorbance sensors are also included to monitor the abundance of green and cyano-green algae directly, through the correlation with their luminescence. The objective is perform a high resolution model for a whole hydrologic year (5-10 m horizontally, 0.5-1m vertically). The main problem of that is high resolution modelling using Delft3D requires powerful computer capacities and storage. In order to address this problem, different computing infrastructures have been tested like Altamira Supercomputer and European Grid Infrastructure (EGI.eu, both Grid and Cloud). This case study have been used also as a use case in different initiatives like INDIGO (H2020) and EGI-Lifewatch Competence Centre. The first step for performing the model of the reservoir was the configuration of the hydrodynamic module. It requires several parameters like bathymetry of the reservoir and other physical parameters like temperature, weather, flow from rivers, etc. all of them taken from different instruments placed at the reservoir. After trying different configurations and different hydrological years, the model has been validated and the results taken in terms of water level and temperature were really good. So that, the hydrodynamic model has been used for feeding the water quality model. The hydrodynamic model has been configured in medium and high resolution, from 100 to 5 meters horizontally and between 0.5 and 1 meter vertically. The temporal step was set between 30 minutes and 6 hours. It requires up to 72 hours of computing in Supercomputers (Altamira Supercomputer at IFCA). It also has been tested in EGI FedCloud infrastructure, with similar results. For collaborative working, Virtual Machines based on Cloud have been launched with all the needed software installed, including different modules and Graphical User Interface. That way configuration and output checking can be done online.
DescripciónTrabajo presentado al Delft Software Days 2015: Delft3D Users Meeting, celebrado el 4 de noviembre de 2015.
URIhttp://hdl.handle.net/10261/140925
Aparece en las colecciones: (IFCA) Comunicaciones congresos
Ficheros en este ítem:
Fichero Descripción Tamaño Formato  
HPCHTCCloud.pdf2,8 MBUnknownVisualizar/Abrir
Mostrar el registro completo
 


NOTA: Los ítems de Digital.CSIC están protegidos por copyright, con todos los derechos reservados, a menos que se indique lo contrario.