RillGrow Soil Erosion Models: background

Emergence and erosion. The eroding hillslope may be conceptualized as a self-organizing system. Antecedent microtopography determines preferential paths for overland flow and hence the spatial pattern of surface lowering; this in turn influences the path of subsequent flow.

This cycle of events constitutes a positive feedback loop operating at a 'local' scale which, if repeated a sufficient number of times, generates a 'global' response from the system: the formation of a rill network.

Viewed from this perspective, the rill network is an 'emergent' whole-system result which arises from a great many local interactions.

Modelling rill erosion as a self-organizing system. Erosion models embodying such a self-organizing systems approach do not suffer from an unrealistic requirement of earlier models: the need to pre-specify rill characteristics for an initially unrilled surface.

Additionally, it appears that the local interactions between the soil's surface, flow, and transported sediment need only be described in a relatively simple manner. This leads to models with modest data requirements yet strong generality and predictive power.

A disadvantage, however, is the considerable computational requirement of such models.