Soil structure plays a key role in the ability of soil to fulfil essential soil functions and services in relation to e.g. root growth, gas and water transport and organic matter turnover. However, soils are not a very easy object to study as they are highly complex and opaque to the human eye. Traditionally, they have been studied using invasive or destructive techniques. The advantage of using X-ray computed tomography (CT) in soil morphology is that it enables non-destructive quantification of soil structure in three dimensions (3D). The prime objective of the present study was to characterize soil aggregate properties such as volume, surface area and sphericity based on 3D images. We tested the methods on aggregates from different treatments and quantified changes over time. A total of 32 collections of aggregates, enclosed in mesocosms, were incubated in soil to follow the structural changes over time for different treatments. The aggregates had different origins (tillage and no-till), and the mesocosms were incubated in soil grown with and without plants. The aggregates were not segmented into single aggregates, but considered as an aggregate cluster. To describe the aggregate cluster shape changes, several morphometric parameters were quantified such as aggregate cluster volume, sphericity, and the number of inter-aggregate pores. These parameters were measured from 3D images produced non-destructively by an X-ray CT scanner at three different times: (a) the initial state before incubation, (b) after summer incubation, and (c) after summer and winter incubation. The macroporosity of the aggregate clusters decreased after incubating the samples during summer and during summer and winter. The plant treatment curbed the decrease in porosity over time. The volume and surface area of the aggregate clusters increased with time irrespective of tillage and plant treatments. The sphericity decreased with time. The structure model index (SMI) was not sensitive to effects of time and treatments. This means that with time the aggregate clusters became less round and more elongated, but they kept a rough surface. The 3D CT scanner image analysis based on quantification of morphometric parameters has the potential to provide new fundamental insight into soil aggregate formation and the effect of different treatments despite the limitations of its spatial resolution. In our study, time was the most important factor affecting the changes in aggregate shape and structure. Contrary to what we had expected, there was no significant effect of seven years of different tillage treatments on the measured parameters at any time of measurement. The segmentation method used to separate aggregate from air is of primary importance, particularly when it comes to the quantification of aggregate morphometric parameters.