Science Grade Data Products

Proposal Submission

Background

ESO aims to produce science grade data products (SGDPs) for all VLT instruments. SGDPs are data products which can be used as is by astronomers to carry out measurements. A more detailed discussion of the criteria for a being labeled "science grade" is given in the instructions for SGDP survey 2008 and an excerpt is given here for short reference:

Science grade data products (SGDPs) can be used as-is to extract scientific conclusions, or to carry out quantitative measurements. Typical SGDPs are fully calibrated and mosaiced images which include noise maps or flux calibrated spectra with error bars.

Any assumptions used in the creation of SGDPs (e.g. on instrument properties, environmental conditions or noise properties) are independent of the science goals. Assumptions on the scientific contents of the data, external knowledge which is not generally valid or depends on the targets, or scientific judgment related to the contents of the data are not used for the production of SGDPs. SGDPs are therefore general purpose data products and independent of specific targets. For example, photometric redshifts are not SGDPs since they depend on templates.

SGDPs are not necessarily usable as-is for all scientifc purposes. For instance a high-resolution spectrum may allow to determine abundances from narrow lines, but not to fit lines extending over several echelle orders. A calibrated image may allow to derive a colour-magnitude diagram, but not information on gravitational lensing.

It is well known that most of the currently ESO produced data products do not fully meet all criteria to label them "science grade". The reason of any shortcoming might be in the calibration data, observing strategy, used software or even the instrument hardware. Frequently, it is a combination of several of those reasons. The Data Product Department has decided to use an integrated approach to the improvement of data products, where resources a pooled to work on all aspects of a data quality issue simultaneously. This contrasts with the previous procedure to accept individual tickets addressing narrow issues e.g. in the pipeline software. The new procedure is based on SGDP proposals.

SGDP Proposals

SGDP proposals are proposals to enhance the data quality of an instrument, instrument mode or observing mode to the level "science grade". A typical SGDP proposal will propose work on pipelines, requests new calibration data and test data to demonstrate the achieved improvements. Proposals might also be related to an instrument upgrade. Every proposal must include a detailed quantitative description of the data improvement it aims to achieve, as well as plan how to achieve it.SGDP proposals can be submitted at any time.

Do SGDP Proposals Replace DFS Tickets?

The new process for SGDP proposals is independent and in addition to the DFS ticket system. Specific problems or requests with a pipeline should still be reported as a DFS ticket. However, the new process does impact on the way DFS tickets are handled. In order to provide as much man power as possible to the implementation of SGDP proposals, most pipelines will be frozen by default. No features will be added to frozen pipelines, and there are only limited resources to fix bugs. In practice this meant that DFS tickets for all non-active pipelines  will be put on-hold. Putting DFS tickets on-hold does not mean they will not be dealt with eventually, but the level of support available for such ticket is significantly lower than it used to be. Therefore, the response time for this tickets is expected to be much slower, even if a ticket is marked "high priority". If a ticket is urgent enough to require immediate attention, in addition to submitting the ticket you need to send an email to the Data Products Board explaining the reason why the ticket is urgent.

Who can submit Proposals?

SGDP proposals are ESO internal proposal, and can be submitted only by ESO staff members. Most proposals will be submitted by members of an IOT, and every proposals should be discussed within the relevant IOT(s). The PI of a proposal is the project responsible. Authors of the proposals should be committed to fully support the work if a proposal is implemented. Proposals related to instrument upgrades should be submitted by members of the upgrade team.

Who will evaluate Proposals?

Proposals will be evaluated and ranked by the Data Products Board. Members of the board are the head of the Science Data Products (SDP) group, the IOT Coordinator, the head of Paranal Science Operation (PSO) , the head of the Data Products Department (DPD) , the head of Pipeline Systems Department (PSD) , and a representative from Instrumentation Project Department (IPD).

How will accepted proposals be implemented?

The main purpose of this new procedure is to pool and concentrate resources to simultaneously work all angles of projects to improve the data quality. Proposals will only be accepted if there is a firm commitment of all necessary resources. Proposers should specify clearly the resources they will commit to the project. As global responsible for the enhancement of ESO data products, the Science Data Products group will lead the implementation phase. A project plan with goals, milestones, and a clear schedule will be produced for every project. Regular progress meetings will take place. At the conclusion of the project, the achieved improvement will be demonstrated with appropriate test data, and a final report will be produced.

Criteria for Evaluating Proposals

Setting the global goals and priorities for the improvement of data quality is the responsibility of the Data Products Board. Given the very limited resources available, the overall philosophy in accepting proposals is to do only few projects at any time, but do them fast and with full support. Specific criteria for the evaluation of proposals are: Proposals should be articulated, comprehensive, and set some specific measurable goals to improve data quality. They are different from individual tickets submitted e.g. to PSD concerned with individual features of specific instruments.

  • Proposals must demonstrate that their goals are realistic, can be achieved within a specified time frame with available resources, and provide a quantitative estimate of the scientific goals if applicable.
  • Proposals should take into account the current status of pipelines.
  • Proposals which aim to improve the data quality of two or more instruments are preferred. Such proposals should discuss in detail how the proposed solution can be applied to several instruments.
  • Most instrument upgrade projects carried out by the Instrument Division require adaptation of software and/or the calibration plan. Such work is approved via different channels and not the subject of SGDP proposals. However, upgrades also offer the opportunity to improve the data quality beyond was is achieved by the hardware improvement. Given the unique opportunity of instrument upgrades, such proposals receive preferential treatment.

How to submit a Proposal

Proposals should be submitted as a pdf document.
Templates for the proposals are available as:
Word .doc template to create .pdf proposal and LateX template to create .pdf proposal file.

Proposals should be submitted via this webform :

Proposal Submission Form

Relevant Documents