The May/June 2004 CBA campaign on the rare superoutburst of LL And

 
The dwarf nova LL And was discovered in 1979 by Wild. At the moment of its discovery, the object was merely studied, and the variable star community had to wait till 1993, before LL And again was found in outburst (visual detection by myself on December 7.096 UT, 1993). Time-resolved CCD photometry during that outburst at Kyoto University revealed the existence of superhumps with a likely period of 0.05697 d (PASJ paper, 2004). In addition, Patterson et al. (PASP paper, 2003) reported the detection of a photometric quiescence periodicity of 0.055053 d, which might be attributed to the orbital period of the system.
 

The May 2004 outburst of LL And

On May 25th, 2004, Patrick Schmeer announced his detection of a rare outburst of LL And, using the 0.37-m Rigel Telescope of the University of Iowa, reporting an unfiltered CCD magnitude of 12.6 on a CCD image acquired on May 22.464 UT (JD +147.5). The CBA network immediately announced an intensive monitoring campaign on LL And, realising very well that the unfavourable location of LL And in the morning sky (rising only close to the start of twilight) would make the campaign extremely difficult.

At present, the CBA has accumulated a total of 2800 time-series CCD observations of LL And, covering the time interval between JD 2453151.8 and JD 2453158.0 (>6.1 days). Participating observers are : Donn Starkey (CBA Indiana), Tonny Vanmunster (CBA Belgium) and Jonathan Kemp / Sarah Tuttle (MDM Observatory, 1.3-m telescope). 

 
Photometric light curve from JD +151.8 till JD +155.1

The resulting photometric light curve of LL And can be split in two sections : a first section covers the nights from JD +151.8 till JD +155.1 (figure 1 below). The second section covers the nights of JD +155.93 till JD +158.0 (figure 5). 

The initial observation sets (up to JD +153.7, colored blue-red-black-purple in figure 1 below) show no evidence of any periodic modulation. The fluctuations in this part of the light curve are best explained by atmospheric extinction effects, caused by the low altitude of the object and the unfiltered observing. Starting with the observation set around JD +153.9 (pink color) and JD +154.9 (brown color), a first hint of a weak modulation starts to appear (see figure 2). 
 

Fig. 1. LL And overall photometric light curve covering JD +151.8 till +155.1. 
Observations have been zero-averaged (all figures are Peranso screenshots)

We attempted a period analysis on this first section of the LL And light curve, using periodic orthogonal polynomials to fit the observations, and using the analysis of variance (ANOVA) statistic to evaluate the quality of the fit. This method for period analysis was recently proposed by Schwarzenberg-Czerny (AJ, 460, L107-110, 1996). It strongly improves peak detection sensitivity and damps alias periods. The next release (1.10) of Peranso will include an implementation of this method (which we refer to as ANOVA). 
 

Fig. 2.  LL And light curve around JD +154.9 showing a first hint of a weak modulation. 
Open circles represent observations that are excluded from period analysis calculations.

Using the ANOVA method of Peranso 1.10, we find a weak signal (mean amplitude about 0.06 mag) at 0.0546 +/- 0.0003 d, which could be significant (further analysis to be done). Folding the observations on this period yields the phase diagram shown below (Fig 3).

Fig. 3.  LL And phase diagram with observations of section 1 of the light curve 
folded over a weak period of 0.0546 d. The red curve is a spline-fitted curve.

 
 
Photometric light curve from JD +155.93 till JD +158.0

 
The second section of the LL And CBA light curve covers the nights of JD +155.93 till JD +158.0 (figure 5), which clearly show full-developed superhumps. They emerged about 8 days after the start of the present LL And outburst.

Fig. 4. LL And light curve on JD +155.9, showing the first superhump development in the present outburst.

 
Figure 5 below shows the LL And light curve covering the superhumping stage. Using the ANOVA method, we derive a Psh value of 0.05654 +/- 0.00012 d. The corresponding period- and phase diagrams are shown in figures 6 and 7. The amplitude of the superhump modulation is about 0.26 mag.

Our superhump period value is shorter than the value of 0.05697 d derived at Kyoto University on the basis of CCD observations from 1993, and is likely more accurate (we use 7x more observations, covering a longer time span too). Assuming an orbital period of 0.055053 d, the Kyoto Univ. fractional superhump excess value was 3.5%, which is unusually large for a short-period system. It triggered a discussion on the location of LL And on the evolutionary track of binary systems (see http://arxiv.org/abs/astro-ph/0310626). However, our superhump period value now yields a fractional superhump excess value of 2.7%, which is quite common for an UGSU-type dwarf nova.
 

Fig. 5.  LL And light curve covering the superhump stage (starting from JD +155.9)

Fig. 6. LL And period window (ANOVA method) for superhump stage, yielding Psh = 0.05654 d

 
Fig. 7. Phase diagram for P = 0.05654 d 
 

Evidently, we will continue to monitor LL And over the coming nights (weather permitting).  If the present superoutburst turns out to last for another week or so, the object might even classify as a possible UGWZ-type dwarf nova. Time will tell.