2 Observations

First observations of comet S4 were done in March and April 2000 using the 250 GHz Max-Planck Millimeter Bolometer (MAMBO) array at the IRAM 30 m telescope on Pico Veleta, Spain. MAMBO is a 37-element bolometer array with an effective frequency of 250 GHz and a bandwidth of about 80 GHz. Under good weather conditions each bolometer has a sensitivity of about 35 mJy $\rm s^{-0.5}$ and a HPBW of 10.7''; the pixel separation is about 23 $^{\prime\prime}$, the instantaneous field of view of the array has a diameter of about 120 $^{\prime\prime}$.

The observations were done in the standard ON-OFF mode, with individual scans of divided in 12 or 16 subscans of 10 s on and off source integration, respectively. The secondary mirror was chopped by about 50 $^{\prime\prime}$ in Azimuth at 2 Hz, and the telescope was nodded by the same distance after each subscan. The data were analysed with the MOPSI package, created by Zylka (1998), which allows for efficient sky noise reduction (see Bertoldi et al. 2000). More details on the instrumentation at both telescope sites and the observing and data reduction techniques can be found in Paper I.

The results are compiled in Table 1. The unusually high non-gravitational forces on comet S4 (if not considered in the ephemerides) can result in position errors up to 36 $^{\prime\prime}$ (Marsden 2000c) and correspondingly to underestimates of the cometary flux density. Therefore the references to the orbital elements used for the observations are included in Table 1.

On March 27, 2000 our trial observations resulted in a marginal detection (Fig. 1 and Table 1), promising a strong signal at 250 GHz near perigee. The attempt to confirm this detection in early April failed probably because of unsufficient sensitivity (Table 1).

Unexpectedly then in July our MAMBO observations failed to detect the comet, either because the signal of comet S4 was dramatically weaker as predicted, or because of problems related to the non-gravitational forces. Our telescope control programs are designed to handle unperturbed orbital elements only for the observing epoch without correction for non-gravitational parameters. Responding to our need Green (2000) provided a set of unperturbed orbital elements, which helped to detect comet S4. Since position errors due to unaccounted non-gravitational parameters vanish at the epoch of the orbital elements, Marsden (2000b) provided orbital elements for 3 epochs near the perigee of comet S4. Using these elements at the nearest epoch (as unperturbed elements) leads to errors smaller 1 $^{\prime\prime}$ (Marsden, private communication). We therefore used these elements. With these elements we could estimate the position error on July 18, it was about 12 $^{\prime\prime}$; the reason, why we did not find the comet in the instantaneous field of view or in our search map, is the unexpected low signal. Indeed, the observed flux densities were about two orders of magnitude below the predictions made from the trial measurements in March. They were also too weak for halo size and position determinations.

Figure 1: Observed signals at 250 GHz of Comet C/1999 S4. The time integrated signals are plotted for all bolometer channels as function of integration time. The signal of Channel 1 (on source) is emphasized by the upper thick line. The $\pm$rms dispersion of the off-source channels is shown as dashed lines. The lower solid line shows the average signal of the off-source channels.

Table 1 summarizes the results of the MAMBO observations. Despite the adverse and variable atmospheric conditions typical for daytime observing in summer, we obtained two fully and one nearly significant detections (Table 1, Fig. 1).

Our measurements at 32 GHz (26'' FWHM beam) using the 100 m telescope are also included in Table 1. They were made with the improved new second module of the multifeed system. To reduce confusion, weather, and interference problems, multiple scans were made in the double beam mode (see Paper I).

During the time of our observations, and in fact in the midst of our second 100 m observation, the cometary nucleus disintegrated (Kidger 2000; Sekanina 2000). For the following observations we used the orbital elements of Marsden (2000b) for the appropriate epoch in the hope that they would describe the fragment's center of gravity.

 

 

Table 1: Observational results.

Date	UT	r	$\Delta$	$t_{\rm int}$	$S_\nu$		Epoch	Orb. elements	Reference
2000	[h]	[AU]	[AU]	[s]	[mJy]		2000
250 GHz MAMBO results from IRAM 30 m telescope
Mar. 27	15.8	2.175	2.999	3711		$1.56 \pm 0.45$	Aug. 4.0	MPEC 37478	Marsden (2000)
Apr. 12	12.8	1.961	2.889	1157		$0.34 \pm 0.77$	dto.	dto.	dto.
Jul. 18	6.7	0.781	0.422	3395		$-1.2 \pm 0.6$	Aug. 4.0	MPEC 2000-O02	Marsden (2000a)
Jul. 19	6.1	0.777	0.403	5326		$2.13 \pm0.55$	Aug. 4.0	priv. com.	Green (2000)
Jul. 20	5.4	0.774	0.389	1281		$12.90\pm4.80$	Jul. 15.0	MPEC 2000-O07	Marsden (2000b)
Jul. 21	7.5	0.771	0.378	4873		$5.36 \pm 1.28$	Jul. 25.0	dto.	dto.
Jul. 24	9.9	0.766	0.379	4966		$1.92\pm 0.81$	dto.	dto.	dto.
Aug. 4	15.5	0.787	0.670	2116		$-0.69\pm 1.13$	Aug. 4.0	dto.	dto.
Aug. 7	15.1	0.803	0.777	6140		$1.64\pm 0.76$	dto.	dto.	dto.
Aug. 10	13.1	0.822	0.883	3818		$-3.6\pm 1.96$	dto.	dto.	dto.
32 GHz data from Effelsberg 100 m telescope
Jul. 19	23.8	0.775	0.392	16 000		$-0.30\pm0.22$	Aug. 4.0	priv. com.	Green (2000)
Jul. 23	17.6	0.767	0.375	13 500		$-0.60\pm0.27$	Jul. 25.0	MPEC 2000-O07	Marsden (2000b)