Forum Springer Astron. Astrophys.
Forum Whats New Search Orders

Astron. Astrophys. 343, L65-L69 (1999)

Previous Section Next Section Title Page Table of Contents

2. Observations and data reduction

2.1. Observations

The observations reported here were allocated one full ISO revolution (science window of 16 hours) of ESA guaranteed time. We used the best observational strategy for detection of faint sources, based upon in-flight experience (Altieri et al. 1998). In particular, the field was observed redundantly in 4 consecutive revolutions in 4 blocks of 4 hours. The allocated time was equally divided between the 2 broad-band ISOCAM filters LW2 (5-8.5µm), with reference wavelength 6.75µm, and LW3 (12-18µm), with reference wavelength 15 µm.

We rastered the 32[FORMULA]32-ISOCAM detector array in microscanning mode. The pixel-field-of-view (PFOV) of [FORMULA] per pixel-field-of-view (PFOV) was chosen to obtain high spatial resolution, giving a good sampling of the PSF, crucial for source cross-identification. The raster step size was [FORMULA] (2.33 PFOV), the minimum value to step out of the PSF FWHM in two consecutive pointings. The size of rasters was 10[FORMULA]10, so that, in the central part of each single raster, 100 different detector pixels sampled each sky pixel. Such redundancy was a key factor in ISOCAM deep observations. The final maps cover a field size of [FORMULA][FORMULA][FORMULA]. These new observations supercede in area and depth the previous ISOCAM observations of this cluster (Lémonon et al. 1998).

2.2. Data reduction and calibration

The Abell 2390 raw data from this programme and the Lémonon et al. (1998) data were processed together following the steps:

Dark subtraction using a time-dependent dark correction (Biviano et al. 1998).
Deglitching, flat-fielding and sky subtraction of each raster was performed using the PRETI Multi-resolution Median Transform techniques adapted for ISOCAM data analysis (Starck et al. 1997), a multi-scale analysis that decomposes temporally the different parts of the signals.
Field distortion correction (Aussel et al. 1998) for sky projection, with a final pixel size of [FORMULA].
The 5 rasters were then stacked together using the drizzling technique described by Fruchter & Hook 1998. The rasters were aligned using the centroids of the 4 to 5 brightest sources in each waveband.

The calibration into µJy was done using the refined in-flight calibration values from Blommaert et al. (1998).

We took into account the transient behaviour of the ISOCAM pixel signals, which do not stabilize at faint fluxes, but respond to a change in illumination level with transient drifts that depend both on the background level and the source intensity. We calibrated the impact of these responsive transients for our observational parameters by applying the IAS transient correction model of Abergel et al. (1996) to our data. This showed that our LW2 and LW3 responsivities were 60% and 80% of nominal, respectively.

The central parts of the stacked 7 and 15 µm images are shown respectively in Fig. 1.

[FIGURE] Fig. 1. MIR image of the A2390 cluster: (left ) LW2 image, (middle ) LW3 image. For these 2 images, the beam size (defined as 80% of encircled energy) is indicated in the lower-righthand corner; objects detected with the SExtractor software are indicated by ellipses; contours show the iso-exposure-time of 5ks, 10ks, 15ks, 20ks and 25ks. (right ) LW3 significance image (weighting the image by the square root of the effective integration time per pixel) overlaid on a ground-based Gunn R image

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 1999

Online publication: March 1, 1999