The light curve for Hale-Bopp shows that it has been remarkably well behaved. There are no major flares or fades, though there are a number of significant departures from the mean light curve. Peak brightness was around -0.8 and although brighter estimates were made they probably included a contribution from the bright dust tail. What was unusual about Hale-Bopp was the length of time that it remained bright and the unprecedented range of observations that were made of it.

The analysis of the comet's light curve uses 2971 observations from 92 observers, made on 488 days between 1995 July 25 and 1997 July 30. The mean of all observations made on each day, corrected for aperture using a value of 0.033 mm-1, are plotted in a light curve . The mean light curve has an equation of:

m = -0.65±0.03 + 5 log(D) + 7.49±0.05 log(r)

Departures from this mean curve are shown in the upper panel, offset to -2 mag. The light curve is relatively steady from discovery to 1996 August, when the comet's intrinsic brightness began to fade. By 1996 October it had faded by about 0.5 mag relative to the mean curve. A very slow faded continued until 1997 February, by which time the comet was 0.6 mag below the mean curve. The intrinsic brightness increased by about 1.0 mag during February and March and the comet has since remained about 0.3 mag brighter than the mean curve. There is some evidence for a significant fade of about 0.5 mag between 1996 August 25 and August 30. The comet remained brighter than 0m for 45 days between 1997 March 8 and April 22. A significant fade occurred about 1997 October 20.

This figure shows the same data, corrected for distance from the earth and plotted against distance from the sun. The upper curves show pre-perihelion behaviour, with the variation from the mean curve offset to -2 mag. The lower curves show the post-perihelion behaviour in the same way, offset by 6 mags. In the months after discovery day to day variations were quite large, with an rms deviation of around 0.8 mag. Observers commented on the changing appearance of the comet and this spread probably reflects these minor cometary outbursts. This behaviour seems to have continued until the comet was around 4 AU from the sun, when it began a slow decline in intrinsic brightness, fading by about 0.8 mag until the comet was 1.5 AU from the sun. The intrinsic brightness then increased relatively rapidly, rising by nearly 1 magnitude between 1.5 AU and perihelion at 0.91 AU from the sun. This brightening reflects increasing water ice sublimation, which becomes significant when the comet is closer to the sun than 1.5 AU.

After perihelion the behaviour was relatively constant, with the intrinsic magnitude remaining about 0.3 mag above the mean curve as far as 2.1 AU from the sun.

The observed coma diameter increased from a few minutes of arc in 1995 to reach a peak near 20' in August 1996. It decreased to around 10' by the end of the year and then increased to nearly 30' in March 1997 before declining to around 5' in July 1997, though there is a lot of scatter in the data. These changes mostly relate to the changing distance of the comet from the Earth. When this is allowed for a different picture emerges. The coma was around 0.8 million kilometres in diameter (about 0.6 that of the sun) in 1995 and began to increase in 1996 March, reaching a peak of perhaps 2 million kilometres in 1996 November. Thereafter it slowly declined, reaching 0.8 million kilometres again in July 1997. The figure shows the daily upper quartile of the coma diameter observations in arc minutes (lower panel), with the corresponding real diameter in the upper panel.

The reported tail lengths slowly grew as the comet approached the sun, and there is some evidence for a slight peak of around 2 degrees in the observed length in November 1 996, when the comet was 2.2 AU from the sun. Significant increase began early in 1997 and the tail(s) reached a peak of around 15 degrees a few days after perihelion. When the observed length is corrected for the distance from the earth and the orientation of the radius vector a more confused picture emerges. Historically most comets show maximum tail development shortly after perihelion, and such tails rarely exceed 1 AU. The observations of 15 degrees in early April 1997 translate into a real length of about 0.75 AU, consistent with the historical record. The peak in 1996 November, if real, would imply a length of around 0.5 AU. Some of the tail observations of Hale-Bopp imply real lengths in excess of 1 AU (which are not plotted), as early as mid November 1996. Given that such a length was not observed when the tail was at its normal maximum extent, suggest a degree of imagination in the observations. In order to remove some of the more fanciful observations, the graph shows the daily upper quartile of the observed tail length in degrees (lower panel) with the corresponding real length in AU in the upper panel.