Pastures, Pests and Productivity: Simple Grazing Models with Two Herbivores

Simple grazing models with two herbivores are used to assess the effects of pasture pests on stability and productivity of continously-growing pastures. Algebraic and graphical methods are also presented for estimating losses from pasture pests at different stocking rates directly, from data on productivity/stocking rate relationships. Pests are considered as competing grazing herbivores and denuders of pasture area. Denuding pests have no effect on stability but grazing pests increase the likelihood of discontinuous stability. The forms of the damage function and its dependence on stocking rate are described for both types of pest. Estimating losses from pasture pests by an equivalent reduction in stocking rate can give useful 'best bet' results even when the true effect is to reduce per capita productivity at fixed stocking rates. However, the first will overestimate or underestimate the second in particular situations, depending on the level of the fixed stocking rate relative to the economic optimum. Actual economic losses for rabbits and porina caterpillars, representing grazing and denuding pests, are estimated as $2.1/rabbit and $0.08/porina m-2/stock unit carried, based on a reduction in stocking rate. At a fixed optimum stocking rate losses are 300/0 greater, and at a fixed 75% optimum stocking rate losses are 60% less. Given the sensitivity of losses to stocking rate, field trials to estimate pest or weed effects should assess their impact on stocking rate rather than their effect on productivity at a fIXed stocking rate whose relationship to the theoretical optimum is unknown. The equilibrium model is shown to apply to seasonal pastures, but regular variations in growth rate reduce productivity and increase stability, converting some discontinuously stable systems into continuously stable ones. The likelihood of discontinuous stability appears in any case remote. The 'laissez-faire' herbivore/vegetation model (the 'extensive' equivalent of the one described in this paper, with herbivore numbers varying) can not be applied to more than one herbivore.