National Institute for Environmental eScience. Environmental eScience: Metadata for remote sensing

Metadata for Remote Sensing, 8–9 January 2004

General information

The aim of this workshop is to provide a forum to discuss potential solutions to the particular challenges of working with remote sensing data within the context of the GRID, especially those relating to metadata and data management. Whilst many of the issues facing remote sensing are also being considered by other groups, e.g. the marine metadata initiative and groups working on numerical modelling in the GRID, the combination of factors create specific problems and there are issues that relate to Remote Sensing alone.The workshop will attempt to address the following range of issues:

The Nature of the Data: Remote Sensing produces large volumes of data, continuously, with diverse sampling characteristics.
Data Sources: Remote Sensed data are available from a variety of sources. This can be true for a single sensor and even an individual scene, and data from these disparate sources may not be identical, for reasons of local reception etc.
Data Access: Access to data may depend not only on the status of data but also on the status of the researcher and even on the project for which the data are to be used, depending on which data source is being accessed.
Data formats: Remote Sensed data formats are highly diverse – from instrument specific to generic standards that are rarely optimal for remote sensing data, e.g. linear altimeter data and orbital swath sensor data do not fit well into gridded format such as NetCDF.
Processing level – the audit trail: the same Remote Sensed data can exist in a variety of processing levels – level 0 (raw) data up to level 3 (derived parameters) are common. Each of these levels may receive different processing procedures dependent on the data source and reprocessing with updated or improved algorithms is commonplace. This makes the inclusion of an “audit trail” in metadata an essential element.
Post Processing: where post-processing is required, the large data volumes required may make server-side processing more efficient than local processing. A development of this is where the user requires a combination of data, which may include disparate formats and geographical and/or temporal sampling characteristics.
Accessing Data in Near Real Time: many Remote Sensed data are used in NRT, at (very) remote locations. This acts to exacerbate issues of data security, data integrity and the ability to rapidly transfer appropriate volumes of data or information.

Programme

The programme coordinator is Helen Snaith of Southampton Oceanography Centre. The programme arranged to date is

Thursday 8th January

10:00	Registration and coffee
10:30	Introduction. Martin Dove (NIEeS)
10:45	Workshop aims. Helen Snaith (SOC)
11:00	NERC data management strategy. Mark Thorley (NERC)
11:30	NERC DataGrid perspective on remote sensing metadata. Bryan Lawrence (BADC)
12:00	Satellite and airborne remote sensing data at the NEODC. Matt Pritchard (NEODC)
12:30	Lunch
13:30	SpaceGrid. Luigi Fusco (ESA, ESRIN)
14:00	Discussion: Interoperability of data sets; Portals
15:30	Tea/Coffee
16:00	Discussion: Existing standards
19:00	Reception in Churchill College
19:30	Dinner in Churchill College

Friday 9th January

9:00	Marine XML approach to EO data. Keiran Millard (HR Wallingford)
9:30	Harmonisation of metadata for remote sensing in marine science and coastal management. Roger Longhorn (GI Projects)
10:00	Discussion: Data formats and types: linear, gridded, 3D etc
10:30	Coffee
11:00	Discussion: Near real time data access
12:30	Lunch
13:30	The experience of remote sensing data processing, analysis and modelling using Condor. Yong Xue (University of North London)
14:00	Discussion: Version tracking and audit trails

Location

The main meetings of the workshop, and lunches, will be held at in Pavillion A of the Centre for Mathematical Sciences Cambridge (main entrance shown in photograph below).

Details of the location of the meeting, including travel plans, can be obtained here.

Accommodation is being provided in Churchill College Cambridge; the conference dinner on the evening of 8th January will also be held in Churchill College (photograph below).

Follow-on

It is anticipated that the workshop may generate interest in some follow-on activities. We will discuss the sort of follow-one activities that can be arranged, such as visits to NIEeS to demonstrate some of the escience/grid technologies and to provide help with installation and usage.

Registration

There is no registration charge. Accommodation in College and meals for UK academic participants will be provided with no charge. Registration is essential and must be done using our on-line form.

Please book early to ensure that you are able to obtain a place. Deadline is 31st December – we need to enforce the deadline for participants wishing us to provide accommodation.

Contact and further information

For information about registration, and all other types of queries, please contact the NIEeS office in the first instance.

Last update: Tue, Jan 6, 2004