Reading e-Science Centre

1
Reading e-Science Centre
e-Science for a digitised environment Environmenta
l Science Applications
Analysis, modelling and assimilation of climate
and real-time data
External Collaborations
Associated Personnel
2
Existing projects

• GODIVA (NERC pilot project)
• Grid for oceanographers to enable distributed
  analysis and visualisation of ocean data

3
GODIVA Web Portal (http//www.nerc-essc.ac.uk/godi
va)

• Allows users to interactively select data for
  download using a GUI
• Users can create movies on the fly
• cf. Live Access Server
• Useful as browse tool

4
Existing projects

• GODIVA (NERC pilot project)
• Grid for oceanographers to enable distributed
  analysis and visualisation of ocean data
• GADS (Grid Access Data Service)
• Web Service for providing access to data (1 TB
  at ESSC alone)
• Will be part of workflows (e.g. for use in
  GODIVA)
• Workflows are groups of Web Services which work
  together to form a distributed application

5
Example workflow
(from myGrid project)

• First Web Service (GADS) extracts data from a
  server
• Second Web Service finds the depth of an
  isotherm in the dataset

INPUTS
OUTPUT
6
Existing projects

• GODIVA (NERC pilot project)
• Grid for oceanographers to enable distributed
  analysis and visualisation of ocean data
• GADS (Grid Access Data Service)
• Web Service for providing access to data
• Will be part of workflows (for use in GODIVA)
• Provision of live Met Office data to community
• Thousands of users, hundreds of institutions,
  tens of countries
• Esp. for MERSEA (EU model intercomparison
  project)
• Also in future to BMT for oil spill / search and
  rescue systems

7
Real-time data access

• Met Office FOAM data
• Can easily create subsets of data

8
Intercomparison of model data
Compare results of different ocean models e.g. EU
MERSEA project
9
Existing projects (contd.)

• TRACK software for identifying weather systems
• Computationally intensive but parallel, best run
  on a distributed system
• Would like to do this on many different data sets
  (e.g. ECMWF)

10
Future projects

• Create more Web Services for benefit of community
• Collate Web Services written by other people
  (create UDDI registry)
• Use satellite / in situ data in addition to model
  data
• SEEDA
• Provision of Met Office data to BMT

11
SEEDA project

• Create prototype of system in which real-time Met
  Office data are used to predict path of oil spills

This is the OSIS software, which currently uses
tidal data only would be improved with real-time
Met Office forecasts of ocean currents
12
Future projects

• Create more Web Services for benefit of community
• Collate Web Services written by other people
• Use satellite / in situ data in addition to model
  data
• SEEDA
• Provision of Met Office data to BMT
• EurOPA
• European portal for real-time data for marine
  hazard analysis (EU)
• POPPIE
• Environment Agency / Met Office pollution
  tracking software
• Re-engineer software to improve architecture use
  Web Services
• Bristol University simulations of fragmentation
  in volcanic rocks
• Monte Carlo simulations of virtual rocks

13
Technical expertise

• Strong focus on Web Services
• Data access
• Data analysis/processing
• Data visualisation
• Creating distributed applications using workflow
  engines to orchestrate Web Services
• Also web portals (cf. Live Access Server) ESP
  community
• Future migration to Grid Services would like
  help with OGSA!
• Question How to transfer large data sets between
  Web Services?
• Use of Condor pool at ESSC

14
Summary

• ReSC is engaging with environmental scientists in
  many disciplines and sectors
• Focus on atmosphere/ocean science
• Working with universities, govt agencies and
  commercial sector
• Would like to draw on experience of other
  e-scientists
• ReSC complements, but is distinct from, NIEeS