Geophysical Research Abstracts
Vol. 12, EGU2010-14455, 2010
EGU General Assembly 2010
© Author(s) 2010
GreenView and GreenLand Applications Development on SEE-GRID
Infrastructure
Danut Mihon (1), Victor Bacu (1), Dorian Gorgan (1), Róbert Mészáros (2), Györgyi Gelybó (2), and Teodor
Stefanut (1)
(1) Technical University of Cluj-Napoca, Computer Science Department, Cluj-Napoca, Romania (vasile.mihon@cs.utcluj.ro,
dorian.gorgan@cs.utcluj.ro), (2) Eötvös Loránd University, Budapest, Hungary (fmrobi@nimbus.elte.hu,
gyoreszg@caesar.elte.hu)
The GreenView and GreenLand applications [1] have been developed through the SEE-GRID-SCI (SEE-GRID
eInfrastructure for regional eScience) FP7 project co-funded by the European Commission [2].
The development of environment applications is a challenge for Grid technologies and software development
methodologies. This presentation exemplifies the development of the GreenView and GreenLand applications
over the SEE-GRID infrastructure by the Grid Application Development Methodology [3].
Today’s environmental applications are used in vary domains of Earth Science such as meteorology, ground and
atmospheric pollution, ground metal detection or weather prediction. These applications run on satellite images
(e.g. Landsat, MERIS, MODIS, etc.) and the accuracy of output results depends mostly of the quality of these
images. The main drawback of such environmental applications regards the need of computation power and
storage power (some images are almost 1GB in size), in order to process such a large data volume. Actually, almost
applications requiring high computation resources have approached the migration onto the Grid infrastructure.
This infrastructure offers the computing power by running the atomic application components on different Grid
nodes in sequential or parallel mode.
The middleware used between the Grid infrastructure and client applications is ESIP (Environment Oriented
Satellite Image Processing Platform), which is based on gProcess platform [4]. In its current format, gProcess is
used for launching new processes on the Grid nodes, but also for monitoring the execution status of these processes.
This presentation highlights two case studies of Grid based environmental applications, GreenView and
GreenLand [5]. GreenView is used in correlation with MODIS (Moderate Resolution Imaging Spectroradiometer)
satellite images and meteorological datasets, in order to produce pseudo colored temperature and vegetation
maps for different geographical CEE (Central Eastern Europe) regions. On the other hand, GreenLand is used
for generating maps for different vegetation indexes (e.g. NDVI, EVI, SAVI, GEMI) based on Landsat satellite
images. Both applications are using interpolation and random value generation algorithms, but also specific
formulas for computing vegetation index values.
The GreenView and GreenLand applications have been experimented over the SEE-GRID infrastructure
and the performance evaluation is reported in [6]. The improvement of the execution time (obtained through
a better parallelization of jobs), the extension of geographical areas to other parts of the Earth, and new user
interaction techniques on spatial data and large set of satellite images are the goals of the future work.
References
[1] GreenView application on Wiki, http://wiki.egee-see.org/index.php/GreenView
[2] SEE-GRID-SCI Project, http://www.see-grid-sci.eu/