All Figures

	Figure 1: Example of one matching pair of a F160W image ( left, image n4k6j1a4q) and the corresponding G141 grism image ( right, image n4k6j1zyq) from HLA dataset N4K6IZZCQ.
Open with DEXTER


In the text

	Figure 2: Distribution of NICMOS G141 grism images on the sky.
Open with DEXTER


In the text

	Figure 3: Cumulative distribution of number of astrometric calibration stars used in each field.
Open with DEXTER


In the text

	Figure 4: Difference between objects coordinates computed from the WCS in the image header and coordinates listed in astrometric reference catalogues. The crosses mark the mean of the distributions before and after the installation of the NCS, and the circles make the 1 scatter around the mean. The systematic error in the WCS coordinates is significant larger for the pre-NCS data.
Open with DEXTER


In the text

	Figure 5: Cumulative distribution of estimated astrometric accuracies of the object coordinates before and after the astrometric correction.
Open with DEXTER


In the text

	Figure 6: The zeroth order (wavelength independent) plane of the post-NCS flat field cube ( left) and the first order (wavelength dependent) plane ( right).
Open with DEXTER


In the text

	Figure 7: NICMOS G141 spectrum of the planetary nebula Hb 12 showing the emission line identifications. The upper axis shows the offset in pixels from the position of the direct image.
Open with DEXTER


In the text

	Figure 8: NICMOS sensitivity curves as a function of wavelength for the G141 grism for pre-NCS and post-NCS data sets.
Open with DEXTER


In the text

	Figure 9: Distribution of the number of extracted spectra from HLA datasets.
Open with DEXTER


In the text

	Figure 10: Magnitude distribution of the targets with spectra in the HLA release. The blue histogram are the extended sources.
Open with DEXTER


In the text

	Figure 11: Measured signal-to-noise of the HLA spectra as a function of magnitude.
Open with DEXTER


In the text

	Figure 12: Top: direct image F160W magnitude, as a function of the 2MASS H magnitude. Bottom: difference of the two magnitudes as a function of the 2MASS magnitude.
Open with DEXTER


In the text

	Figure 13: Total throughput of the F160W filter for the NICMOS/NIC3 camera.
Open with DEXTER


In the text

Figure 14: Top: the ``spectroscopic'' magnitude, i.e. the magnitude derived from the spectrum of the object, is shown as a function of the direct image F160W magnitude. Bottom: the difference of the two magnitudes is plotted as a function of the direct image magnitude. For both plots, the colour of the points indicates the level of contamination, coded according to the difference  in the magnitude of the target and that of the contaminator. The colour codes are as follows: green: , yellow: , orange: and red: .

Open with DEXTER


In the text

Figure 15: Examples of catastrophic discrepancies selected from the outliers of Fig. 14. Each panel shows the direct image associated to the source for which an inaccurate spectroscopic magnitude was obtained (marked in red); other objects in the field used for the contamination calculation are marked with blue ellipses. The corresponding cutouts from the grism image are shown below the undispersed images. Left: source in a crowded field, close to an undetected source to the right. Middle: source close to bright star close to the edge of the image. Right: source contaminated by a bright star outside the field (note the spike on the centre-left of the image).

Open with DEXTER


In the text

	Figure 16: Noise measured in extracted spectra using the DER_SNR algorithm as a function of the noise as predicted by error propagation from the image error array.
Open with DEXTER


In the text

	Figure 17: Comparison of HLA spectra with extractions using the NICMOSlook program. All three spectra are H emitting galaxies discovered by McCarthy et al. (1999). The blue curves are NICMOSlook extractions, whereas the red curves are the HLA extractions. The blue labels are the McCarthy et al.'s names.
Open with DEXTER


In the text

	Figure 18: Comparison of an HLA spectrum with a Gemini spectrum of the same source. The blue curve is the heavily smoothed spectrum of SDSS J083643.85+005453.3 from Stern et al. (2003), the red curve is HNG_J083643.82+005453.4_N6LE01ULQ.
Open with DEXTER


In the text

	Figure 19: Comparison of an HLA spectrum with a TNG NICS spectrum of the same source. The blue curve is the spectrum of SDSS 104433.04+012502.2 from Maiolino et al. (2004), the red curve is HNG_J104433.08-012501.6_N6LE03C7Q.
Open with DEXTER


In the text

	Figure 20: Comparison of an HLA spectra of brown dwarf ASR 24 (red curves and error bars) with the spectrum published by Greissl et al. (2007) (blue curve). The latter has been scaled to match the mean flux level of the HLA data.
Open with DEXTER


In the text

	Figure A.1: Schematic drawing of a slitless spectrum measurement.
Open with DEXTER


In the text