Microbes play an important role in facilitating organic matter decomposition in soils, which is a major component of the global carbon cycle. Microbial dynamics are intimately coupled to environmental transport processes, which control access to labile organic matter and other nutrients that are needed for the growth and maintenance of microorganisms. Transport of soluble nutrients in the soil system is arguably most strongly impacted by preferential flow pathways in the soil. Since the physical structure of soils can be characterized as being formed from constituent micro-aggregates which contain internal porosity, one pressing question is the partitioning of the flow among the "inter-aggregate" and "intra-aggregate" pores and how this may impact overall solute transport within heterogeneous soil structures. The answer to this question is particularly important in evaluating assumptions to be used in developing upscaled simulations based on highly resolved mechanistic models. In our synthetic model of soils, firstly we statistically generated a number of micro-aggregates containing internal pores. Then we constructed a group of diverse multi-aggregate structures with different packing ratios by stacking those micro-aggregates and varying the size and shape of inter-aggregate pore spacing between them. We then performed pore-scale flow simulations using computational fluid dynamics methods to determine the flow patterns in these aggregate-of-aggregates structures and computed the partitioning of the flow through intra- and inter-aggregate pores as a function of the spacing between the aggregates. The results of these numerical experiments demonstrate that soluble nutrients are largely transported via flows through inter-aggregate pores. Although this result is consistent with intuition, we have also been able to quantify the relative flow capacity of the two domains under various conditions. For example, in our simulations, the flow capacity through the aggregates (intra-aggregate flow) was less than 2 % of the total flow when the spacing between the aggregates was larger than . Inter-aggregate pores continued to be the dominant flow pathways even at much smaller spacing; intra-aggregate flow was less than 10 % of the total flow when the inter- and intra-aggregate pore sizes were comparable. Although the results may not be exactly the same as those obtained from actual soil systems, such studies are making it possible to identify which model upscaling assumptions are realistic and what computational methods are required for detailed numerical investigation of hydrodynamics and microbial carbon cycling dynamics in soil systems.