AstroGrid-2 Science Scope

Contents:

Introduction

This document outlines the science drivers and outputs for the AstroGrid-2 (AG2) project. This cobers the period Jan-2004, when the AstroGrid-1 project (AG1) finishes, until Dec-2007.

After extensive peer review the AG2 Proposal has recieved funding primarily to cover hardening and extensions to the AstroGrid infrastructure. The key aim is to allow the pervasive deployment of the AstroGrid (1) system.

A design study was conducted during Feb/May 2004 by the Project Manager and Project Scientist into the desired scope of AG2. This document is the result of work by the Project Scientist as to the Science drivers and key outputs for the project.

The technical scope is given in TechnicalScope

Executive Summary

We are building an infrastructure that will improve and accelerate astronomical research; make comparison and integration of data seamless and transparent; and make manipulation and analysis of datasets as easy and powerful as possible. AstroGrid-2 will go further, for example by providing improved workflow capabilities, e.g. in the development of ontological processes which involve directed workflows for common sets of tasks. All these goals are quite general, but in order to build a system that meets astronomers needs we have analysed concrete example science cases. We describe below how this has been done, and how the process will be repeated and improved for AstroGrid-2, with a special new focus on personalisation and capturing direct user requirements for new tools that work with the infrastructure. As well as focusing on specific science cases, we also have in mind preparing UK astronomers for specific future missions and facilities that are of key importance. Finally, this section also notes how the science drivers for the AstroGrid-2 project are aligned with the development of the Euro-VO initiative (of which AstroGrid is a primary member).

The Science Requirements Analysis for AstroGrid

The current AstroGrid project recognised that the creation of a supporting infrastructure to enable astronomical scientific endeavour must be guided and informed by the scientific priorities of the community. This ensures that the capabilities that AstroGrid create are able to meet the demands of the astronomy end users.

To this end, AstroGrid, in its Phase-A period, undertook a rigorous science requirements capture process, led by the AstroGrid Project Scientist.

• Scientific use cases were taken from astronomers associated with the consortium, and from the community at large.
• These were prioritised, and a key set of ten requirements chosen to be representative of both topical science, and processing and analysis needs. (These key AstroGrid science drivers are fully described at http://wiki.astrogrid.org/bin/view/Astrogrid/ScienceProblems).
• The key science drivers were decomposed into use cases. From these the functionality required is being developed within the context of AstroGrid.

Enhancing support and access to current datasets

AstroGrid-2 will enable continued and enhanced opportunities in using existing and legacy data sources. The vital VO-enablement of existing archival datasets held in the UK will provide continued and enhanced value for the UK astronomical community, especially in the era of the new projects described below. In the near future it is clear that the VO-enablement of existing data collections is likely to be a very high priority for the UK community. Following on from the initial pilot VO-enablement activities in the current AstroGrid, new science areas here will include study of galaxy properties via access to multi-object spectral surveys such as those of 2dF and the SDSS. Such studies have enormous potential for astrophysics research leading to both new understanding from multi-wavelength statistical samples and to serendipitous discoveries that can play a pivotal role in leading to new paradigms. All of this depends on the timely VO-enablement of existing data sets.

New Science Priorities

A major initial driver for the AstroGrid project was to support the effective scientific exploitation of major projects emerging in the 2003-2005 time frame, with particular emphasis on the VISTA project (http://www.vista.ac.uk). When it comes into operation in 2006, VISTA is set to provide the UK and European communities with the deepest large area survey of the sky in the Infra-Red. Its precursor, WFCAM on UKIRT, will provide a powerful complementary Infra-Red survey of the Northern sky. The current AstroGrid project will provide the capabilities to deliver the processed data WFCAM data as provided by the VISTA Data Flow System pipeline project, to the community.

The capabilities now planned for AstroGrid-2 aim to meet the demands of other projects identified by PPARC's Science Committee 2003 Strategy Paper and PPARC Road Map as being of high scientific importance to the UK community. These include, in approximate time order:

• exploitation of the UK's investment in European Southern Observatory (ESO) (now)
• the two Gemini telescopes (now)
• XMM-Newton (now)
• SWIFT (late 2003)
• WFCAM (2004)
• VISTA (2006)
• e-MERLIN(2007)
• Herschel (2007)
• Planck (2007)
• ALMA (2008-10)
• GAIA (2010-12)
• JWST (>2010)
• Xeus (>2012)

The above projects will all produce high value scientific data, injected into the VO by pipeline processing. AstroGrid-2 will provide the broad capabilities needed to maximise the utilisation of these data, with emphasis on interoperability, ease of handling large data flow, access irrespective of end user location and so forth.

By way of illustration of the key new demands made by science programmes exploiting data from these projects, typical science from two of the above is given:

Science Drivers from VISTA

Science Drivers from ALMA

Science Drivers from GAIA

Science Drivers from Herschel

Herschel, with its suite of far-IR to sub-mm instrumentation will enable new insights in a number of areas:

• Galaxy formation and evolution in early universe and the nature of active galaxies
• Star forming regions and interstellar medium physics in Milky Way and external galaxies
• Molecular chemistry of cometary, planetary and satellite atmospheres in solar system

Science areas like these will require the use of the capabilities of VOs and AstroGrid-2 in particular to enable access to complementary data sets at all wavelengths from radio to X-ray and beyond. Science priorities such as probing protoplanetary discs and star-forming regions, or investigating Ultra Luminous IR Galaxies and testing cosmological theories against the properties of high-redshift objects, involve regions which are heavily obscured by dust or which are most securely and rapidly identified using the broadest spectral energy distribution.

Towards the Euor-VO: Delivering the AVO Science Reference Mission.

Solar System Research

AstroGrid-2 is aimed at facilitating astronomy. However, new missions will create a new community of astronomers and solar physicists with common interests. AstroGrid-2 will support these science areas from an astronomical perspective, offering solutions developed complementary to those developed in the Solar community.

With the incorporation of the UKSSVO project into the AstroGrid project a number of Solar and STP prioriteis will be addressed. These will include: investigating basic plasma processes such as magnetic reconnection on multiple scales: from solar flares, flaring in active galactic nuclei, to magnetospheric reconnection.

Facilitating New Ways of Working

We expect that new tools will encourage new ways of working, and that the focus of the current AstroGrid on seamless querying and data integration will be an example. AstroGrid-2 will develop further innovative concepts and capabilities :

• Topic specific workspaces, giving access to all data and tools relevant to a certain astrophysical problem.
• Aided work-flows, where the user constructs a personalised data pipeline, using VO components, and in a manner where the system provides sophisticated guidance.
• Methods whereby the outputs of data manipulations can be automatically fed back into the operations of telescopes, both for real-time and ordinary observational programmes.
• Advanced ways of visualising and exploring multi-channel data.

AstroGrid-2 Science Requirements Process

As for the initial project it is vitally important that the deliverables from AstroGrid-2 are aligned with the scientific needs of the community. To this end a system is required to allow the capture and prioritisation of the user science needs.

It is anticipated that the systems and procedures put into place by AstroGrid will be adopted by AstroGrid-2 and adapted where necessary. The project will initially seek out a wide and demanding set of science drivers which will shape the virtual observatory. It is anticipated that the large collection of use cases gathered during the current AstroGrid project will be used as a basis upon which to build. This activity will led by the AstroGrid Project Scientist, and will be completed in the first few months of the new project.

AstroGrid, during its Phase-B development activities, has constituted the AstroGrid Science Advisory Group to advise the project, through the Project Scientist, on a number of issues:

• scientific utilisation of the AstroGrid Virtual Observatory
• advising the project on ranking of scientific priorities
• technical issues as they affect end users (e.g. user interfaces, functionality provision, access to data products etc).
• deployment of the AstroGrid Virtual Observatory and take up through e.g. the beta tester programme

It is proposed that this group will continue for AstroGrid-2. Adjustments to the composition of the group will reflect the updated priorities for AstroGrid-2. For example members representative of the other major funded projects (potential projects described in Section-8) will supplement the AG-SAG, to ensure that the scientific demands of these projects (which will rely on the AstroGrid-2 infrastructure to a greater or lesser extent) are met. The fit of the AG-SAG within the overall project management structure is shown in Section-9.1.

Personalisation in AstroGrid-2

AstroGrid-2 recognises that new requirements will emerge through the project as the user community develop new ideas whilst working in the new virtual observatory environment. A key concept of AstroGrid is to allow user customisation - for instance, configuration of the AstroGrid portal to reflect their priorities, and their own working areas with MySpace. AstroGrid-2 will take this a step further by developing specific applications and tools to meet a specific user demand (as prioritised via a call, and independent peer review by the AstroGrid Science Advisory Group). This will be managed within the project iteration cycle and will allow development of tools in a 3 or 6 month period. Thus this would allow for instance the grid-enablement of a user code or application to perform an advanced analysis step (e.g. a classifier of objects, SN, extreme Pop III stars, etc) based on a combination of indicators from various data sources. This concept is outlined in Section 4.3, 'Personalising the Virtual Observatory'.

---

-- NicholasWalton - 25 Apr 2004