There are two nutrient budgets that characterise the mineral cycles in an ecosystem:

The two are interrelated, and the internal budget is ultimately dependent on limits imposed by the external one. Important as this is, well-worked- out nutrient budgets, either of the internal or external type, are still relatively few in number.

For internal nutrient budgets, it is a major task to determine the mineral content of the biotic components of an ecosystem, and to assess shifts in this content with time, let alone to trace the flow of that content through the food chain. It is for this reason that most investigators have chosen to work with the cycle of one or two nutrients; even that kind of study is extremely demanding and fraught with technological as well as interpretational difficulties. For example, J. P. Witherspoon used the radioisotope of cesium (¹³⁴Cs) to study the movement of this nutrient in white oak trees at the Oak Ridge National Laboratory. He found that over two growing seasons, the maximum concentration in the leaves occurred in early June and amounted to about 40 per cent of the total input (2 millicuries), and the remainder spread to the woody tissues in the roots, stem, and branches. Of the total leaf content, 33 per cent was lost through leaf fall, 15 per cent was leached out by rain, and the remainder was incorporated in woody tissues. By November, about 70 per cent of the rain-leached cesium was in the top four inches of the soil, 17 per cent was added to the leaf litter, supplementing that which had come through leaf fall, for a grand total of about 19 per cent of the original input.

The movement of cesium in white oak suggests that the turnover time differs for different parts of the tree. This is a generally recognized phenomenon.

With the exception of tropical forests, turnover time in the trees is shortest in the canopy (primarily leaves but also including flowers and fruits) than in the litter and in all cases is longest in wood. Cycle time tends to increase with increasing latitude, a trend which becomes more evident if the soil compartment is omitted from the total cycle time; this is a reasonable omission in that element availability patterns in the soil have a considerable affect on turnover time in the soil as can be seen in the lack of a trend in soil turnover time. A trend to increased length of intra-tree cycling time with increasing latitude would be expected because uptake (and release) is directly related to the rate of primary production and that rate decreases with increasing latitude.

This short exposition indicates the complexity of interchange of nutrients within a given ecosystem.  In addition to the dynamic interchanges of nutrients that occur within ecosystems among its atmospheric, soil, and biotic components, there is an exchange of nutrients between ecosystems resulting from geological, meteorological, and biological forces.

Geological actions such as volcanic eruptions spew materials into the atmosphere or spread lava over the terrain thereby transferring nutrients from one place to another. Meteorological actions such as rock weathering or wind which carries nutrients whipped up into dust or evaporated into the atmosphere bring about exchanges of nutrients between ecosystems. Animals that feed in one ecosystem and defecate or die in another, or trees grown in one ecosystem and burned in another are obvious examples of external nutrient exchange resulting from biological activity.

With our great capacity for movement of food and and fertilizers we are without question the most powerful biological agent affecting internal and external nutrient budgets.

Regarding modelling nutrient flows, much of this type of work has involved the  judicious choice and intensive study of small watersheds. For example, in their studies in New Hampshire, Herbert Bormann and Gene Likens and their associates have been able to circumvent several of the limiting aspects of the study of external nutrient budgets. Each of the six watersheds they selected for study is characterized by watertight bedrock and lateral boundaries that coincide with topographic divides; hence each is discretely isolated from the water- borne output of adjacent watersheds and none is subject to any deep seepage or underground circulation. Further, the isolation of the forest from sites of active agriculture minimizes the mineral contribution that wind-borne dust brings to many ecosystems, and the homogeneous nature of the bedrock further reduces variability in the system.

A further and most propitious aspect of the choice of the watersheds is that they are within the Hubbard Brook Experimental Forest, an established hydrologic laboratory which continuously monitors precipitation and runoff by standard meteorlogical procedures. Weekly analysis of hydrologic input and output for particular nutrients has characterised annual budgets for the several significant cations and anions in the forest.