Документ взят из кэша поисковой машины. Адрес оригинального документа : http://www.stsci.edu/hst/wfc3/documents/ISRs/2002/WFC3-2002-02.pdf Дата изменения: Fri Sep 26 03:24:55 2008 Дата индексирования: Thu Jan 15 04:55:36 2009 Кодировка:

Instrument Science Report WFC3 2002-02

Expected WFC3-IR count rates from zodiacal background
M. Stiavelli January 18, 2001

ABSTRACT This report captures the present estimates on the count rate from zodiacal background seen by the WFC3 instrument. The instrument as currently designed allows for background limited observations in the J band. In the H band the combined detector, instrument, and OTA noise exceeds by 38 per cent the zodiacal background.

Introduction
The near-IR channel of WFC3 was initially designed to be zodiacal background limited for observations with the J and H broad band filters. This was accomplished by estimating the count rate from zodiacal background and using it to place a constraint of 0.4 electrons/s on the combined detector noise and instrument thermal background. Since the time this constraint was derived the instrument designed has evolved significantly and both the J and H filter have been reduced in width. Now that the design and filter specifications have stabilized it is a good time to revisit the issue and confirm that WFC3 is indeed zodiacal background limited in broad band observations. The count rate from zodiacal background is estimated from a a simple model of the IRAS background as a function of the angles. The model has been updated in order to account for the new COBE data. This model is verified by using different models or direct measurements. In section 2 I will introduce the zodiacal background model and the consistency checks that were carried out and in Section 3 I will apply them to derive expected count rate from the instrument as presently designed. Section 4 is devoted to discussion of the results and conclusions.


Instrument Science Report WFC3 2002-02

Zodiacal background model
The model that was used is based on an IDL program obtained from Pierre Bely and built by combining J. Good's model based on IRAS data (NASA IRAS sky survey suppl. May 94) with a model of the scattered light by Angel and Woolf (see also Wheelock et al. 94). I changed the dust distribution and dust albedo of the original model to improve the agreement with observed DIRBE data points. I will refer in the following to this model as a DIRBE-improved IRAS-based zodiacal background model. Table 1. Values of the zodiacal background from the IRAS-based model.

Elongation Inclination F(1.2µm) F(1.6µm) (degrees) 0 0 0 60 60 60 90 90 90 135 135 135 180 (degrees) 45 60 90 0 45 60 0 45 60 0 45 60 45 (MJy/sr) 0.25 0.18 0.13 0.44 0.21 0.17 0.20 0.15 0.14 0.13 0.11 0.12 0.10 (MJy/sr) 0.20 0.14 0.10 0.35 0.17 0.13 0.17 0.12 0.11 0.10 0.09 0.09 0.08

On the basis of the results given in Table 1 I will consider the values for (180,45) as the minimum background from the model, the values at (60,0) as the maximum and those at (90,45) as the typical background. I have also carried out an average along the elongation for a 45 degrees inclination. This yields an average zodiacal background of 0.17 MJy/sr at 1.2 µm and 0.14 MJy/sr at 1.6 µm. The zodiacal light model presented in the NICMOS Handbook (v2 1997) supposedly providing mean values for an inclination of 45 degrees over the ecliptic, was used to verify

2


Instrument Science Report WFC3 2002-02 these values. The model yields 0.33 MJy/sr at 1.2 µm and 0.32 MJy/sr at 1.6 µm. These numbers are intermediate between the typical and the maximum values and are listed in Table 2 as NICMOS. It should be noted that the current version of the NICMOS ETC is using a zodical background lower than that predicted by the model. As an additional sanity check I verified the background values with COBE measurements. At 1.25 µm Kelsall et al. (1998) give 0.15 MJy/sr in the Lockman hole observed at solar elongation 90 degrees, and 0.35 MJy/sr on the ecliptic plane. These numbers are well within the range found above. In order to estimate a count rate the above numbers need to be converted to more practical units. In Table 2 I list the minimum, typical, and maximum zodiacal background in erg cm-2 s-1 е-1 arcsec-2 and in photons (HST area)-1 s-1 е-1 arcsec-2. For the second estimate I have considered the total collecting area of a 2.4 m primary ignoring any obstruction. Table 2. Zodiacal backgrounds in Flux units and in counts

Location

erg cm-2 s-1 е-1 arcsec

-2

photons (HST area)-1 s-1 е-1 arcsec-2

1.2µm

1.6µm
-19 -19 -19 -19 -19

1.2µm 0.013 0.020 0.023 0.059 0.044

1.6µm 0.008 0.012 0.014 0.035 0.032

Minimum 4.90 10-19 2.20 10 Typical Average 7.35 10-19 3.31 10 8.32 10-19 3.86 10

Maximum 2.15 10-18 9.64 10 NICMOS 1.60 10-18 8.74 10

In the following, I will use the average background estimate which has also been highlighted in Table 3.

The expected count rate
The count rates per HST area can be converted directly into detected count rates after multiplication by the total instrument throughput, the HST area, and the filter width. I have adopted for the J band a filter width of 3000 A and a throughput of 0.30 and for the H band a filter width of 3500 A and a throughput of 0.32. The derived values are listed under the column "Throughput" in Table 3. Alternatively, one can use the WFC3 ETC to derive the count rates for the given input fluxes. These results are listed under the column "ETC" in Table 3.

3


Instrument Science Report WFC3 2002-02 Table 3. Estimated zodiacal background count rates for WFC3/IR

Location

Throughput 1.2µm 1.6µm 0.18 0.26 0.31 0.77 0.70 1.2µm 0.24 0.36 0.40 1.04 0.77

ETC 1.6µm 0.16 0.25 0.29 0.72 0.66

Minimum Typical Average Maximum NICMOS

0.23 0.34 0.39 1.00 0.74

As it can be seen from Table 3 there is overall agreement between the estimates through the ETC and those based on the instrument throughput at the filter central wavelength.

Discussion and Conclusions
From the numbers summarized in Table 3 and by considering the WFC3/IR channel as described in the instrument CEI specs one can draw the following conclusions: 1. 2. J band observations with WFC3/IR will typically be background limited. For H band observations the combination of detector, instrument, and OTA noise will exceed the typical zodiacal background by 38 per cent. This is partly due to a shorter wavelength cutoff than expected and to a lower detector peak QE at 1.6 µm compared to what had initially been assumed. Assuming that the whole difference is due to a total throughput in the H band lower by 75 per cent (combined filter and QE effect) and considering that the noise is also decreased (fewer counts from the zodiacal background) one finds that for faint sources the exposure times in the H band will have to be increased by about 38 per cent.

3.

4. These considerations ignore the earthshine which would add to the zodiacal background and potentially make all observations background limited. However, observation with the NICMOS instrument seem to indicate that earthshine was not a major contributor to the background (less than 20-30 per cent of the total, Calzetti private communication.) Finally, one should note that at this time the flight detector has not been selected. Once the detailed characteristics of the WFC3 flight detector will be known it may be advisable to revisit this calculation.

4


Instrument Science Report WFC3 2002-02

Acknowledgements
I wish to thank Chris Hanley, Carey Lisse, and Massimo Robberto for providing me with independent consistency checks and Mauro Giavalisco for providing useful comments.

References
Good, J., in IRAS sky survey atlas: Explanatory supplement. Kelsall, T., et al., 1998, ApJ 508, 44. MacKenty, J.W., et al., 1997, NICMOS Handbook, p. 33 Wheelock, S.L., et al., 1994, IRAS sky survey atlas: Explanatory supplement.

5