Sunspot group decay

We examine daily records of sunspot group areas (measured in millionths of a solar hemisphere or mu Hem) for the last 130 years to determine the rate of decay of sunspot group areas. We exclude observations of groups when they are more than 60 degrees in longitude from the central meridian and only include data when at least three days of observations are available following the date of maximum area for a group's disk passage. This leaves data for over 18 000 measurements of sunspot group decay. We find that the decay rate increases linearly from 28 mu Hem day(-1) to about 140 mu Hem day(-1) for groups with areas increasing from 35 mu Hem to 1000 mu Hem. The decay rate tends to level off for groups with areas larger than 1000 mu Hem. This behavior is very similar to the increase in the number of sunspots per group as the area of the group increases. Calculating the decay rate per individual sunspot gives a decay rate of about 3.65 mu Hem day(-1) with little dependence upon the area of the group. This suggests that sunspots decay by a Fickian diffusion process with a diffusion coefficient of about 10 km(2) s(-1). Although the 18 000 decay rate measurements are lognormally distributed, this can be attributed to the lognormal distribution of sunspot group areas and the linear relationship between area and decay rate for the vast majority of groups. We find weak evidence for variations in decay rates from one solar cycle to another and for different phases of each sunspot cycle. However, the strongest evidence for variations is with latitude and the variations with cycle and phase of each cycle can be attributed to this variation. High latitude spots tend to decay faster than low latitude spots.