WFPC2 Associations

Table of Contents

Introduction

A bit of history
The first associations (type A)
The new associations (type B)
Comparing type A and type B (Reason to go "B")

Related Documents

FAQ, Warnings, User Interface, Products, Publications, Browser

People

Introduction

Up to February 1998,: archive users had to retrieve WFPC2 observations on a single exposure basis. Previously, lacking higher level of abstraction, there was no simple way of tracking back which observing strategy (e.g., cosmic ray rejection, dithering) a Principal Investigator decided to employ.
In February 1998,: ST-ECF introduced the first generation of WFPC2 Associations, known as type A associations.
In November 2001,: CADC and ST-ECF released the second generation of WFPC2 Associations, known as type B associations.

Since the introduction of associations,

an archive user can browse through the WFPC2 associations, to immediately see which exposures are aligned on the sky (CR-SPLIT), or which exposures are dithered and by what extent (offsets).

Sometimes, despite of what the PI asked for, exposures which were meant to be well aligned are instead shifted by some pixel fraction. The associations describe what actually happened during the observing run. Therefore,

an automatic pipeline has been put in place to allow WFPC2 archive researchers to retrieve from the ST-ECF archive not only On-The-Fly (OTF) re-calibrated exposures, but also other new products, like co-added Cosmic Ray free (CR) images.

Type A Associations

To construct type A associations, the telescope pointing information (jitter) was used. When available such information provides a reliable way for identifying CR-SPLIT observations, that is, observations split into two or more exposures to later remove the cosmic rays. Unfortunately, only about 57% of the WFPC2 observations have accompaining good jitter information; the rest divides between missing jitter (telemetry drops) or not accurate pointing information (no FINELOCK), or the telescope was slewing.

Type B Associations

To provide more complete associations a new type scheme has been implemented. Type B associations contains all exposures following the same creterion of association of type A, but REGARDLESS of the availability or the accuracy of the jitter information. The offsets between the various members of an association are measured using a cross-correlation technique. When the cross-correlation fails (not enough signal, typically in the blue filters), the jitter information is used. In case the jitter information is missing or cannot provide reliable offsets, the science header World Coordinate System is used.

Comparing Type A and Type B

Why yet another type of associations ?

Excluding calibration observations (constituting ~30% of the whole archive, 21% if expressed in observing time), the percentage of observations that end up into associations is:

Type A	Type B
75%	93%

The percentage of observations that not only end up into associations, but for which we also know reliably the offests is:

Type A	Type B
57%	93%

which expressed in exposure time reads:

Type A	Type B
29.6E+06 seconds	42.1E+06 seconds
66.4% of total scientific time	94.4% of total scientific time

This means that the Science Archive Facility can provide automatically stacked (co-added and cosmic-ray free) products covering almost the entire number of photons collected by WFPC2.

People involved in the project

CADC	D. Schade, D. Durand, S. Gaudet
ST-ECF	A. Micol, P.D. Bristow, M. Dolensky, B. Pirenne

Comments to: Alberto Micol (FirstName.LastName@eso.org)