We study the relationship between auroras observed by the ultraviolet imager (UVI) on the Polar spacecraft and Pi2 pulsations detected by the Circum-pan Pacific Magnetometer Network at magnetic latitudes lower than 55¿. Instead of focusing on the few minutes around auroral breakups, we treat both auroral power and pulsation parameters as continuous time series in an effort to quantify the relationship between the two phenomena. A 3-hour interval on 3 April 1996 is selected for the present study because it includes several episodes of auroral intensification. The auroral power is represented by electron precipitation energy (PUVI) that is inferred from the observed 170-nm photon emission over the 2000--0200 magnetic local time sector in 36-s time windows separated by ~5 min. Parameters characterizing Pi2 signals in the magnetometer data are computed in a moving 5-min time window using the autoregressive spectral analysis technique. The parameters include the amplitude APi2 in the Pi2 band (7--25 mHz), the frequency, the ellipticity, and azimuth of the major axis of polarization for oscillations with a well-defined spectral peak. Overall, there is a high degree of similarity between the temporal profiles of PUVI and APi2. The pulsation amplitude varied from ~0.1 nT for a PUVI change of 20 GW; although due to different decay timescales, ~10 min for APi2 and ~80 min for PUVI, the linear cross-correlation coefficient between the two parameters remained at a modest level (~0.75). In one instance, APi2 was enhanced without auroral intensification, and for this pulsation the azimuth of the major axis of polarization was oriented in the east-west direction instead of the north-south direction that is the norm for Pi2 pulsations associated with auroral intensifications. Examination of solar wind plasma data from the Geotail spacecraft suggests that this particular event was directly driven by changes in the solar wind dynamic pressure. We conclude that low-latitude magnetic pulsations are a sensitive indicator of changes of auroral intensity and that pulsations that are not associated with aurora could be distinguished by examining their polarization properties.