Observations of the February/March 2004 outburst of BZ UMa

February and March typically are not very good months for variable star observing in Belgium, because of the highly unstable weather during that period. 2004 was no exception to this rule. We had 2 clear nights during the February/March 2004 (normal) outburst of BZ UMa. 

February 26/27, 2004

Below, I present my observations, obtained on Feb 26/27, 2004 using a 0.35-m f/6.3 telescope and an unfiltered ST-7XME CCD camera. In total, I collected 491 observations over a period of 7.30h, under very good sky conditions.

Overall, BZ UMa faded an impressive 0.4 mag over the duration of the observations (corresponding to 1.3 mag per day). See figure 1 below. Using a Beta version of the new period analysis software package Peranso, I first detrended the observations. This clearly revealed the presence of oscillations with an amplitude of about 0.04 mag (figure 2). All period analysis techniques available in Peranso (I used PDM, a few Fourier methods and a few string methods) essentially indicated the non-periodicity of the oscillations.

Fig 1 - overall light curve of BZ UMa, with superimposed trendline (Peranso screenshot)

Fig 2 - lightcurve after detrending, revealing oscillations (Peranso screenshot)

February 29/March 01, 2004

The light curve of Feb 29 / Mar 01, 2004 showed a totally different BZ UMa outburst stage. The object had faded by about 2.85 mag compared to my previous observation night.  I collected 397 observations over a period of 7.9h, under mediocre sky conditions (haze, strong moonlight).

Figure 3 - light curve of BZ UMa on 2004, Feb 29 / Mar 01 (Peranso screenshot)

The most dominant feature in the light curve this time are high-amplitude modulations (average amplitude about 0.4 mag). See figure 3 above. Using a Beta version of the period analysis software Peranso, I combined these observations with the ones I obtained on Feb 26/27 (yielding a total of 888 observations), and started looking for significant periods using a number of techniques (PDM, Lomb-Scargle, Bloomfield). All analyses pointed to a dominant period at 0.0691 +/- 0.0010 d, which is pretty close to the orbital period value of 0.06799 d, derived by Jurcevic et al. (Jurcevic, J.S., Honeycutt, R.K., Schlegel, E.M., and Webbink, R.F. 1994, PASP, 106, 481).

Figure 4 - Lomb-Scargle period determination based on all available BZ UMa observations (Peranso screenshot)

I finally left out the observations of Feb 26/27, 2004 an again used the Lomb-Scargle method in Peranso to make a period analysis, resulting in figure 5 below. Clearly, the same period is found, but this time without the "noise" induced by the Feb 26/27 observations.

Figure 5 - Lomb-Scargle period determination based on Feb 29/Mar 01 observations (Peranso screenshot)





Copyright © 2004 - Tonny Vanmunster.