Landscape Ecology vol. 10 no. 4 pp 191-196 (1995) SPB Academic Publishing bv, Amsterdam

Human-mediated vegetation switches as processes in landscape ecology J. Bastow Wilson and Warren McG. King Botany Department, University of Otago, P.O. Box 56, Dunedin, New Zealand Keywords: Switch, boundary, ecotone, landscape, mowing, trampling, human effects, positive feedback, garden design

Abstract Switches are processes in which there is positive feedback between vegetation and environment. Landscape features can be created and modified by switches. The concept has previously been used with physical factors and non-human animals as the switch mediator, i.e. the factor which the vegetation modifies and which in turn affects the vegetation. Here, the switch concept is extended to include some types of human behaviour as possible switch mediators. With this extension, the switch concept can explain the impact on the landscape of some types of human behaviour. Examples are given of the behaviour of mower drivers, mowing up to a boundary which they create and/or maintain, and of walkers trampling tracks which they create and/or maintain. Other possibilities are discussed briefly. It is concluded that the concept of a human-mediated switch can unify the study of human behaviour, vegetation processes and landscape ecology. Introduction Human impact is now the main determinant of landscape pattern over much of the globe (Hansen et al. 1988; Fuentes 1990; Lepart and Debussche 1992). The processes of disturbance and management by humans are often considered qualitatively different from the ‘natural’ processes that are studied by most ecologists. In this paper, we suggest that some types of human impact are mediated by a basic process that is in common with natural communities - the vegetation switch. Wilson and Agnew (1992) introduced the concept of the ‘switch’ in vegetation - a positive feedback process between species composition and the environment. A switch operates when a community modifies the environment in a direction that favours the maintenance of that community. The term ‘switch’ is used following analogous use by Odum (1971).

A switch can operate in time: vegetationlenvironment change can be accelerated or delayed, giving the possibility of a sharp temporal change. A switch can also operate in space: a small initial difference in biota or environment can switch between alternative stable states of vegetationlenvironment across an abrupt boundary. A spatial switch can create a mosaic in a previously nearuniform area. It can also sharpen an existing gradient, creating an abrupt ecotone. We are concerned here mainly with the latter process, which clearly has the potential to modify the landscape (Armand 1992; Wilson and Agnew 1992; Agnew et al. 1993). Wilson and Agnew (1992) gave examples of switches mediated by a range of factors, and able to create landscape features. Some mediators are abiotic, e.g. water supply, pH, light, fire and wind. A well-known example is the firellight switch, which creates the savannah/closed-forest mosaic

193

mower-driver, but the shorter community does not attract him. Where there is an underlying gradient, e.g. in slope, water supply, soil type etc., such behaviour could sharpen the gradient into a boundary. Differences in slope can be reflected in species composition (Dargie 1987), and such a vegetational gradient could be sharpened by a mowing switch. Alternatively, a switch mediated by the behaviour of a mower-driver could produce a boundary in an initially-uniform area of grassland, due to an initial decision by a mower operator. However, the initial position of the chosen boundary is likely itself to be caused by an implicit gradient. This implicit gradient might be a psychological one, such as distance from a land boundary (e.g. a fence). All these are examples of what Wilson and Agnew (1992) saw as a Sharpening (category B) outcome of a switch. The same process is seen even more markedly at a boundary between mown grassland and shrubland or woodland (Gysel 1951). Gysel suggests that the exact boundary is often the dripline of the tree, but the coincidence is not necessarily directly causal, for example the mowing line might be influenced by the dampness of the grass. A similar switch could operate when a farmer clears the scrub on an area of land, and maintains the shrubless state up to the boundary. The boundary produced by a One-sided switch is intrinsically unstable, in that the boundary can move into the area in which the switch does not operate, in this case into the mown area. This would occur if the spatial extent of mowing is repeatedly slightly short of the tall grass, so that the tall-grass zone gradually extends into the mown area. This occurs on the University of the South Pacific compounds (A.J. Watkins, pers. comm.). A gradual change could also occur in the opposite direction, with extension of the mown area, if the switch is less than fully efficient.

Track trampling

Informal pathways or tracks can be established in semi-natural vegetation, or even across mown grass where there is frequent pedestrian traffic (Fig. 2).

The first pedestrians start by walking along a certain route. This route is not necessarily the shortest distance between two points, it can be influenced by environmental variation such the topography or the softness of the ground (Bates 1950). Once a pathway is established, other pedestrians follow it, and reinforce it (Leland et al. 1986). They do this partly by following an indication of the appropriate route without any other reason for preference. However, in semi-natural vegetation there may be a genuine advantage in following an established route because shrubs are trampled and the way made easier. Once the path is established, vegetation change will occur (Williams and Lambert 1960; Page et al. 1985; Sun and Liddle 1991; Liddle 1991). Sometimes this change affects only species abundances: Bates (1935), comparing a footpath with adjacent vegetation, found that some species were present at much greater frequency on the footpath, but there were no species exclusive to the footpath. However, sometimes the vegetation changes to the extent that species enter the footpath community which are absent in adjacent untrampled vegetation. Liddle and Greig-Smith (1975) found two such species on tracks through a sand dune complex. These changes can be related to differences in species’ trampling tolerance, which in turn can sometimes be related to the species’ morphology (Sun and Liddle 1993; Bates 1935). Once such vegetation change has occurred, the path will be more clearly marked, and the vegetation will comprise prostrate species, which will further encourage pedestrian traffic, creating a positive-feedback mechanism - a switch. The two elements of the track-trampling switch are: (i) The shorter track vegetation encourages trampling. This also attracts walkers from other areas; (ii) Trampling keeps the vegetation of the tracks short and encourages trampling-resistant, prostrate species. Lack of trampling keeps other areas tall. Both effects reinforce element (i), completing the positive feedback loop that is intrinsic to a switch. The process represents a Reaction (i.e. Type 2)

195

Conclusion

Human interference is often seen as unnatural, a process (‘noogenesis’)that is not related to the laws of ecology, and which creates a system (‘noosphere’) that is replacing the natural biosphere (Vernadsky 1945). However, humans are animals as much as any others, and are subject to the same laws of nature (Odum 1971). Seeing some human behaviour as a switch mediator enables us to see at least some human activity as falling into the same patterns as traditional non-human ecology. The ‘unnatural’ character of abrupt edges has played an important role in gardedlandscape design. The Renaissance ‘Italianate’ style developed into the formal gardens of France, reaching a peak at Versailles. The style is characterised by hard edges, most dramatically in the parterres. Such gardens were seen as symbolic of human control over the environment, of safety from enemies and relief from stress (Glacken 1967; Clifford 1966). Renaissance values in garden design were overturned by moves towards a more ‘natural’ landscape, making more use of natural forms (Clifford 1966). The natural approach was espoused and executed by Kent, Bridgeman, Brown and Repton. Its primary characteristic was the elimination of straight lines, but often edges were softened too, as in Brown’s work at Stowe. Some modern garden design has also emphasised the softening of edges: ‘Nature abhors hard edges’ (Ward 1912). An alternative trend in modern garden design, the counterpart to the modernist movement in architecture, has involved a reversion to hard edges, as in the lines and grids of trees, and the square mono-specific beds of the Miller House garden, midwestern U.S.A. (Frankel and Johnson 1991). In the garden of the Wright House, Seattle, the hard straight edges of the terraces contrast with the created ‘natural’ background which has deliberately softened edges (Frankel and Johnson 1991). Ecology can interpret the hard/soft edge dichotomy in gardedlandscape design in three ways. Firstly, the action of humans on the landscape can be seen as artificial and hence undesirable; from this point of view a human switch is a vicious cycle that exacerbates the undesirable influence of hu-

mans. A second approach is to see hard edges as artificial, but to welcome this, to see the formality of, e.g., the Miller House garden as the victory of mankind and its machine over nature (Gutkind 1952). The third viewpoint is that whilst gradual change is common in nature, hard edges are also present, caused by sharp environmental changes or by natural switches (Wilson and Agnew 1992; Agnew et al. 1993); human switches are an example of these edge-sharpening processes, and the hard edges they produce are therefore not to be eschewed. Not all human influences on the landscape operate via a switch; probably only a few do. In order for a switch to operate, the crucial elements are that not only does human behaviour affect the vegetation, but the vegetation also affects human behaviour, and this feedback is in the direction that reinforces the behaviour and vegetation differences. When such a process is operating, the concept of the switch can bridge the gap between human behaviour, the study of vegetation processes and landscape ecology. Acknowledgements

We thank Chris Bycroft, Louis Leland, Stephen Roxburgh, John Steel, Susan Walker and Anni Watkins for ideas and discussion, and the Editor, Frank B. Golley, for stimulating suggestions. JBW thanks the Biology Department, Open University, UK, for facilities whilst on study leave. References Agnew, A.D.Q., Wilson, J.B. and Sykes, M.T. 1993. A vegetation switch as the cause of a forest/mire ecotone in New Zealand. J. Veg. Sci. 4: 273-278. Armand, A.D. 1992. Sharp and gradual mountain timberlines as a result of species interaction. In Landscape Boundaries: Consequences for Biotic Diversity and Ecological Flows, pp. 360-378. Edited by A.J. Hansen and F. di Castri. SpringerVerlag, New York. Bassett, P.A. 1980. Some effects of defoliation on the vegetation dynamics of two Camargue grasslands. Acta Oecol. Oecol. Plant 1: 121-135. Bates, G.H. 1935. The vegetation of footpaths, sidewalks, carttracks and gateways. J. Ecol. 23: 470-487.

196 Bates, G.H. 1950. Track making by man and domestic animals. J. Anim. Ecol. 19: 21-28. Buchanan, W.G. 1981. Selective mowing. USGA Green Section Record 19: 21-22. Bond, S.D. 1987. Field margins, a farmer’s view on management. In Field margins [British Crop Protection Council Monograph no. 351. pp. 79-83. Edited by J.M. Way and P.W. Greig-Smith. PCPC Publications, Thornton Heath, UK. Clifford, D. 1966. A history of garden design, edn. 2. Faber and Faber, London. Dargie, T.C.D. 1987. An ordination analysis of vegetation patterns on topoclimate gradients in south-east Spain. J. Biogeogr. 14: 197-211. Darlington, J.P.E.C. 1982. The underground passages and storage pits used in foraging by a nest of the termite Marcotermes michaelseniin Kajiado, Kenya. J. Zool. 198: 237-247. Frankel, F. and Johnson, J. 1991. Modern landscape architecture: redefining the garden. Abbeville Press, New York. Fuentes, E.R. 1990. Landscape change in mediterranean-type habitats of Chile: patterns and processes. In Changing Landscapes: An Ecological Perspective. pp. 165-190. Edited by I.S. Zonneveld and R.T.T. Forman. Springer-Verlag, New York. Gibson, D.J., Seastedt, T.R. and Briggs, J.M. 1993. Management practices in tallgrass prairie: large and small-scale experimental effects on species composition. J. Appl. Ecol. 30: 247-255. Glacken, C.J. 1967. Traces on the Rhodian shore. University of California Press, Berkeley. Gutkind, E.A. 1952. Our world from the air. Doubleday, New York. Gysel, L.W. 1951. Borders and openings of beech-maple woodlands in southern Michigan. J. Forest 49: 13-19. Hansen, A.J., di Castri, F. and Naiman, R.J. 1988. Ecotones: what and why? Biology International, special issue 17 [A new look at ecotones: Emerging international projects on landscape boundaries]: 9-46. Leland, L.S. Jr., Hughes, P., Halder, S. and Rowan, M. 1986.

An evaluation of the effectiveness, costs and benefits of laying a paved path to permit grass regeneration. J. Envir. Syst. 16: 1- 11. Lepart, J. and Debussche, M. 1992. Human impact on landscape patterning: Mediterranean examples. In Landscape Boundaries: Consequences for Biotic Diversity and Ecological Flows. pp. 76-106. Edited by A.J. Hansen and F. di Castri. Springer-Verlag, New York. Liddle, M.J. 1991. Recreation ecology: effects of trampling on plants and corals. Trends Ecol. Evol. 6: 13-17. Liddle, M.J. and Greig-Smith, P. 1975. A survey of tracks and paths in a sand dune ecosystem. J. Appl. Ecol. 12: 909-930. Marshall, E.J.P. 1988. The ecology and management of field margin floras in England. Outl. Agric. 17: 178-182. Odum, E.O. 1971. Fundamentals of ecology, edn. 3. Saunders, Philadelphia. Page, R.R., da Vinha, S.G. and Agnew, A.D.Q. 1985. The reaction of some sand-dune plant species to experimentally imposed environmental change: a reductionist approach to stability. Vegetatio 61: 105-1 14. Parish, R., Turkington, R. and Klein, E. 1990. The influence of mowing, fertilization, and plant removal on the botanical composition of an artificial sward. Can. J. Bot. 68: 10801085. Sun, D. and Liddle, M.J. 1991. Field occurrence, recovery, and simulated trampling resistance and recovery of two grasses. Biol. Conserv. 57: 187-203. Sun, D. and Liddle, M.J. 1993. Trampling resistance, stem flexibility and leaf strength in nine Australian grasses and herbs. Biol. Conserv. 65: 35-41. Vernadsky, W.I. 1945. The biosphere and the noosphere. Am. Scient. 33: 1-12. Ward, C. 1912. Royal Gardens. Longmans, Green, London. Williams, W.T. and Lambert, J.M. 1960. Multivariate methods in plant ecology. 11. The use of an electronic digital computer for association-analysis. J. Ecol. 48: 689-710. Wilson, J.B. and Agnew, A.D.Q. 1992. Positive-feedback switches in plant communities. Adv. Ecol. Res. 33: 263-336.