Документ взят из кэша поисковой машины. Адрес оригинального документа : http://www.iki.rssi.ru/earth/pres2006/shelestov.pdf Дата изменения: Mon Feb 19 19:12:35 2007 Дата индексирования: Tue Oct 2 12:20:06 2012 Кодировка: Поисковые слова: comet

Distributed system for water quality monitoring using remote-sensing data

Shel estov A., Kravc henko A., Kussul N. Space Res earc h Institute of Ukraine, Kiev

Mos c ow 2006


W ater Quality Service


Goal: continuous monit oring of bio-optic al properties of water mass in Dnieper estuar y Tool: ocean c olor analysis
- -



The "health" of wat er Level of biologic al activit y Monitoring harmful algae Monitoring phytoplankton levels Developing total maximum daily loads



Applic ations

Mos c ow 2006


Context


W ater qualit y ass essment problem is consider ed as one of the highest priority application within European global monit oring GMES initiative W ater monitoring is included in the work plan of internation al grou p on Earth Obser vation (GEO) f or 2007-2009 period
- -



WA-07-P2 (Global W ater Quality Monitoring) WA-07-P3 (Satellite W ater Measurements)

Mos c ow 2006


Case Study Area


Test area: view f rom MODIS, 250m red ban d Kherson St at e Hydr obiology St ation



Mos c ow 2006


Rem ote Sensing Data


Provide "s ynoptic view" of proc ess es instead of point data s ources Usually the only source of dat a Depen dence on weather c onditions (optic al) Multispectral c apabilities in optical r ange are required due t o com plex pr ocess of atmos pheric c orrection (only 10% of light at sensor c omes f rom ocean) Use optic al data f rom MODIS sensor
36 spectral bands in 0.4µ-14µ 250m-1km spatial res olution, 1-2 images/day at 20 m0i6ddle altitudes





Mos c ow


Service's Dataflow

Mos c ow 2006


SeaDAS Package


Seadas (SeaW iFS Data Analysis System)
-

Developed in NASA GSFC http://oceancolor.gs fc.nasa.gov/s eadas/ L1A to L1B processing, L2 product generation, L3 binning Thematic process ing (msl12)




Functionality
- -

Chlorophyll concentration, SST, SST4, ... Originally developed for SeaWiFS data 8 bands 1.1 km spatial resolution at nadir

Mos c ow 2006


m sl12 ­ Ancillar y Data


Meteorology
- -

Climatolog y (default) Near real time data


Wind (10m), pressure at MSL, Precipitable water Every 6 hours, between ­ simple interpolation global NCEP 1deg grid (not enough in coastal regions)



Ozone
-

EP TOMS, TOAST from satellite data, climat ology, NOAA OISST



SST
-
Mos c ow 2006


SeaDAS Product: Chlorophyll Concentr ation

Mos c ow 2006


SeaDAS Product: Sea Surface Tem perature

Mos c ow 2006


Num erical W eather Prediction


Provides prediction of 4-d space-time distribution of main atmos pheric par amet ers W eather Research&Forec astin g model
mesoscale/limited area NW P model Finer resolution (compared to global circulation models) Require forecast frames from global models for boundary c onditions Parametrization support Variational Dat a Assimilation support Support for distributed memory cluster arch.
Mos c ow 2006




Current W RF Configuration


Conf igured f or territor y of Ukraine Input & boundary c ondition f rom GFS 3 day f orec asts ever y 6 hours 10 km horizontal grid, 200x200 gridpoints 31 vertic al levels One run takes 3.5 hours on 2 x2 Opt eron s yst em











Mos c ow 2006


W RF Configuration for Dnieper Estuary


Mesosc ale NW P models c apable t o run with 1km horizont al res olution Using nest ed grids Domains: 10 x10km, 3.3km, 1.1km All domains 200x200 grid points approx. 3 x com p. time increas e
Mos c ow 2006










In-situ Data


Tem per ature pr of iles, chloroph yll concentration, species biodiversit y

Mos c ow 2006


Technolog y: Grid


Grid
To tak e user away from details of distributed High Performanc e Computing resources Grid system designed to integrate res ources of different administration domains How: introducing of wid e set of standard interop erabilit y protoc ols



Globus Toolkit v4 - Grid middleware, Karajan workf low description langu age, Torqu e - loc al cluster scheduler OpenMPI & Scaly - MPI stan dar d implem entations, Gan glia - monitoring t ool
Mos c ow 2006




Current Infrastructure

Mos c ow 2006


Hardware

SKIT -1
Mos c ow 2006

SKIT -2

Development cluster


W RF on Grid (1)

Mos c ow 2006


W RF on Grid (2)

Mos c ow 2006


Data Presentation


Open G eos patial Consortium (OGC) standards f or data pres entation (www.opengeos patial. org)
- -

vector and raster geospatial dat a HTTP bas ed transfer, HTTP/GET paramet er passing Visualization




Main standards
WMS (Web Map Servic e), SLD (Style Layer Descriptors), WMC (Web Map Context) WFS (Web Feature Service), WCS (Web Coverage Ser vic e)

Delivery


Mos c ow 2006


OGC Im plem entations


Commercial (ArcIMS, IONIC RedSpider) Open Source (UMN MapServer, GeoSer ver) Using open source UMN MapSer ver sof tware
- -





CGI application (simple user interfac e, OGC interfac es) as library

Mos c ow 2006


Visualization: W RF Forecasts


http://dos.ik d.kiev.ua/? option=com_wrf

Mos c ow 2006


Future Plans


Complet e validation Tune s ystem par amet ers Add dat a assimilation of sat ellit e dat a in meteorologic al model Include ecologic al water model support
-







Casc ade of models: weather, hydrolog y, hydrod ynamic

Mos c ow 2006


Thank You!

Mos c ow 2006