Radiocarbon (14C) is commonly used as a tracer of the carbon cycle to determine how fast carbon moves between different reservoirs such as plants, soils, rivers or oceans. However such studies mostly emphasize the mean value (as Δ14C) of an unknown probability distribution. We introduce a novel algorithm to compute Δ14C distributions from knowledge of the age distribution of carbon in compartmental systems at equilibrium. Our results demonstrate that the shape of the distributions might differ according to the speed of cycling of ecosystem compartments and their connectivity within the system, and are mostly non-normal. The distributions are also sensitive to the variations of Δ14C in the atmosphere over time, as influenced by the counteracting anthropogenic effects of fossil-fuel emissions (14C-free) and nuclear weapons testing (bomb 14C). Lastly, we discuss insights that such distributions can offer for sampling and design of experiments aiming to capture the precise variability of Δ14C values in ecosystems.