Identifying correct statistical test for comparison of two groups related to time


I'm conducting research into woodlice grouping behaviour.
I have tested three conditions that are full darkness, full light, and a half-dark, half-light group with each having a total of three replicates.
For the full darkness and full light conditions, I measured the frequency of aggregates, and the number of individuals within each aggregate at 0, 1, 2, 3, 4 and 5 minutes.
For the condition that had half-light and half dark, woodlice could choose between a sheltered or unsheltered area of the dish they were in. Consequently, I measured the proportion of individuals present in each side, plus the total number of aggregates on each side and the number of individuals within each aggregate, again recorded against time.
I have done some preliminary tests of difference comparing the end results of aggregate numbers and produced scatter graphs showing aggregate frequency and size plotted against time. I want to statistically compare how aggregate frequency and size changes with time between the light and dark groups as well as the sheltered and unsheltered groups but I can't figure out the best way to do this. Do I need to use an ANOVA? And as I have three replicates do I need to combine these into averages and compare the average of each condition or keep the replicates seperate?

Any advice would be greatly appreciated and I can clarify anything if needed.