A Review of Damage by Mammals in North Temperate Forests: 1. Deer

A Review of Damage by Mammals in North Temperate Forests: 1. Deer Forestry Commission, Alice Holt Lodge, Wrecclesham, Farnham, Surrey GU10 4LH, England SUMMARY The causes of browsing, bark stripping, and fraying damage by deer are examined by reviewing the available literature. Trees species differ in vulnerability and each form of damage occurs within certain age and size classes. Stem morphology has an important influence on bark stripping; lower branches and bark thickening tend to deter stripping in any one species. Site related factors such as hiding cover, snow and soil fertility also influence damage. The data relating deer population density to damage are imprecise and there is a need for improved density estimation methods to demonstrate the benefit of culling in different habitats. Vegetation affects both habitat and diet selection in deer, and can create both positive and negative relationships with damage. Computer models are proposed as an aid to damage prediction and forest protection decision making. INTRODUCTION The problems of deer are familiar enough to most foresters in Britain. Deer cause damage by browsing, stripping bark and fraying trees with antlers. Although methods of tree protection and deer control are well established to deal with the problem, there remains a need for prediction of possible losses, so that appropriate measures of protection and control can be taken. Effective prediction of losses is only possible if the causes of deer damage are well enough understood. The purpose of this paper is to review the factors such as deer population density, tree characteristics and habitat that influence damage and to identify future research needs as well as to suggest improvements in methods of damage prediction. Altogether five species of deer occur in Britain in sufficient numbers to cause damage (red deer Cervus elaphus, sika C. nippon, roe Capreolus capreolus, muntjac Muntiacus reevesi and fallow Dama dama). All, with the possible exception of fallow deer, have been increasing in range or numbers and are likely to continue to do so (Taylor, 1981; Ratcliffe, 1987, 1989; Gill, 1990). The review addresses the problem of damage in British forests but draws heavily on investigations conducted in all north temperate forest ecosystems. Techniques of tree protection and deer control are not discussed. Collection of relevant published literature ceased after September 1991. Forestry, Vol. 65, No. 2,1992 R. M. A. GILL 146 Forestry Two papers, covering other aspects of mammal damage, are due to be published; part two will review the causes of small mammal damage and part three will cover the responses of trees to damage, including the effects of browsing on tree regeneration, herbivore resistance and production losses due to both browsing and bark stripping. The following definitions apply to this review: BROWSING The characteristics of browsing damage Description The term 'browsing', in the context of forest damage, refers to all forms of feeding damage other than bark stripping and therefore can involve the removal of twigs, shoots, leaves, needles, buds or flowers, from either young trees or coppice stools. Small seedlings can be uprooted. Deer are selective and the parts taken will depend very much on the species of tree and time of year. Some reports state that only the current year's growth is removed and a browsed shoot is unlikely to be re-browsed until new growth has formed (Holloway, 1967a; Severinghaus and Severinghaus, 1982), but this is likely to depend on browsing pressure. Conifers are usually browsed in winter, often with increasing intensity as winter progresses, whereas broadleaves are more usually damaged in summer (Holloway, 1967a; Konig, 1976; Miller etal., 1982; Cummins and Miller, 1982; Klein etal., 1989; Maizeret and Ballon, 1990). Exceptions to this general pattern however do occur, for example a peak in browsing on springflush growth has been reported on Douglas fir (Pseudotsuga menziesii) and Sitka spruce (Picea sitchensis) in spring or early summer (Browning and Lauppe, 1964; Welch etal., 1988a). Furthermore larch (Larix sp.) has been reported to be browsed more in summer than winter (Holloway, 1967a) and twigs of some broadleaved trees, particularly willow (Salix sp.) can form a significant component of the diet of red and roe deer in winter (Szmidt, 1975; Jamrozy, 1980). Damage: Injury to trees in the form of tissue removal (leaves, bark, flowers, shoots, buds etc.). It does not necessarily imply economic loss. Selection: Choice of a particular tree crop, individual tree or part of a tree for damage by an animal from those available. Vulnerability or susceptibility: Likelihood or frequency of damage to a tree or tree crop of particular characteristics. It can only be recorded by observing selection by an animal and is therefore prone to all the factors that influence selection. Incidence: Percentage of trees damaged in an area. Intensity: Severity of damage to an individual tree, such as the proportion of shoots browsed, or the number or size of bark wounds. Deer: Any member of the Cervidae. Damage by Deer 147 Differences between tree species and variety Deer show marked preferences for particular tree species. These are most apparent in mixed species stands, where the level of damage on each species is usually distinctly different (e.g. Horton, 1964). The susceptibility of each of the common tree species to red and roe deer browsing in Europe have been summarized in Table 1. In general, willows, aspen (Populus tremula) and silver fir (Abies alba) are most preferred whereas Sitka spruce, Scots pine (Pinus sylvestris) and Corsican pine (P. nigra) are usually least preferred. There are also however many discrepancies: lodgepole pine (P. contorta) for example has been reported to have both high and low relative preference. As discussed later, preferences can depend on the vegetation composition, so they cannot be expected to be consistent. In common with between species differences in palatability, there are also reports of within species variation. The susceptibility of Douglas fir to blacktailed deer (Odocoileus hemionus) browsing has been found to be a heritable characteristic, passed from parent to known progeny trees (Dimock etal., 1976; Silen and Dimock 1978). In Finland moose (Alces alces) have been reported to prefer browsing some clones of Scots pine more than others (Haukiojaefa/., 1983; Danell et al., 1990), and also show a preference for varieties of southern rather than northern origin (Niemela et al., 1989). It is not yet clear however whether these origin differences are more related to the levels of secondary compounds or to phenological development. The influence of size and age Most browsing by deer usually occurs at an intermediate level between ground and full reach resulting in smaller and larger trees being relatively protected (Holloway, 1967a; Loyttyniemi and Piisila, 1983). In Sitk (...truncated)