Modelling Uncertainty and Variability in Environmental Systems

Scientists from different disciplines now routinely work together on complex, multi-interactive, geochemical and biochemical environmental processes and it has become necessary to define a common language that can be understood by all the relevant scientific community. This in turn will aid the transfer of expertise between scientists from different disciplines. This paper attempts to relate different modelling approaches dealing with uncertainty in environmental processes. The literature review is based on hydrological, geophysical and environmental risk modelling, but the conclusions are relevant to all scientists working on environmental problems, which are almost invariably poorly defined. We argue that modelling of environmental processes should be stochastic rather than deterministic. After listing the basic definitions and possible sources of uncertainty in environmental models, we present the main approaches to the modelling of uncertainties in, and variability of, environmental processes. We propose a general methodology for dealing with uncertainty, using as an example aspects of uncertainty assessment in modelling of rainfall-flow process at a catchment scale.