Ecosystem Modelling: Vegetation and Disturbance

Vegetation is continuously changing in time and space. The changes may be autogenic, driven by the properties of the component plant species, or allogenic, driven by external factors. Allogenic change may be a result of gradual changes in environment, but more often results from disturbance events, and the way vegetation responds will be very dependent upon the disturbance history (Miles, 1979). The term disturbance is used in ecology to refer to the partial or complete reduction of biomass due to some external factors and does not include natural mortality or decay processes. Typical causes of disturbance are grazing, cutting, fire, and frost. Disturbance can also be defined as an event that causes a significant change from the normal pattern in an ecological system (Forman and Godron, 1986), but the problem with this relative definition of disturbance stems from the difficulty of defining the “normal” range of conditions for an ecosystem (White and Harrod, 1997). This is why an absolute definition requiring measures of real change is more appropriate for a mechanistic modelling approach. This was the option made by Waring and Running (1998), for example, who defined disturbance in a forest as any factor that brings about a significant reduction in the overstory leaf area index for a period of more than one year. Grubb (1985) reviewed the concept of disturbance and distinguished vegetation resilience 'in situ' and 'by migration' (Grubb and Hopkins, 1986), referring to the regeneration capacity of surviving individuals in the disturbed area and recolonisation from neighbouring areas respectively. In the simplest case of one type of disturbance on a homogeneous site the disturbance regime can be generally described by its frequency and its intensity. However, if intensity is associated with biomass accumulated, as in the case of fire, there is an inverse relationship between intensity and frequency. These considerations were very well established for fire regimes in North America (see for instance Heinselman 1978) and in Europe (e.g. Angelstam 1997). Therefore the simplest way to describe a fire regime is to determine the average time between disturbances (years) or conversely, the mean frequency of the disturbance (expressed in year –1). This paper focuses only on fire and grazing. Henderson and Keit (2002) performed a canonical correspondence analysis on the sources of variation in shrub composition of eucalypt forests which showed that fire and grazing accounted for more variation than the combination of all other environmental and spatial variables. Although they are perhaps the two most important sources of disturbance, they differ very much in character; the former causes an abrupt discontinuity in the system while the latter is characterised by a gradual and continuous process of selective biomass removal.