New paradigms for modelling mass transfers in soils

The distinction between deterministic and stochastic models is becoming increasingly blurred, and the concept of decoherence suggests that the choice between them ought to be based on the scale at which the model is to be used. Decoherence also suggests that predictive relationships for denitrification and water and solute transport may be easier to achieve at catchment or landscape scale than at smaller scales. The original statement by Laplace that defined determinism emphasizes that the choice of model also depends on the level of information available. Transport processes in the soil and many of the models used to describe them are non-linear and probably non-autonomous as well, suggesting that chaotic behaviour is possible in both soils and models. We need therefore to consider an approach to modelling them that is based on Chaos Theory. Soils and landscapes behave in a way that suggests that the whole is more than the sum of the parts. This is a characteristic of systems described as emergent in the terminology of Complexity Theory. Emergent systems characteristically show complex behaviour that can be described by simple rules and may provide a useful paradigm for modelling soils and catchments. (C) 1998 Elsevier Science B.V. All rights reserved.