Predicting the overall effect of woody borders on insect assemblages in agricultural landscapes

Predicting the overall effect of woody borders on insect assemblages in agricultural landscapes is difficult because the effects that they have been hypothesized to have are numerous, and often conflicting. Management decisions are further complicated because the aims of insect conservation and agricultural production may often conflict. Results of four studies carried out on insects in alfalfa fields using different methods are discussed. The results of these studies suggest that woody borders can increase insect richness without a cost to agricultural production. ___________________________________________________________________________ Introduction Fencerows in North America have typically been formed from areas being cleared for agriculture with linear remnants of earlier continuous forest left between them, or by trees growing up along fences, stone piles and other features between fields. In this way their formation and structure is different from the hedgerows of Europe, which are frequently planted, and often tended. Much more research has been done on woody borders in Europe than in North America. Despite these differences however, the literature shows that much research has focused on the same questions in Europe and North America. For the purposes of this paper we use the term woody border to refer to both hedgerows and fencerows, and the term field margin to refer more generally to margins with or without woody vegetation. Fencerows have been suggested to be important in the conservation of many different taxa, including insects, small mammals, and birds. The fate of fencerows however, will probably be determined by the effect that they have on agriculture through insects. In this paper we review some of the ways in which woody borders may have an impact on the assemblages of insects found in agricultural areas. We look at the results of four studies carried out by members of the Landscape Ecology laboratory at Carlton University examining insect assemblages in alfalfa (Medicago sativa L.) fields and discuss what the results show about the impact of woody borders. Finally we discuss these results in light of previous work, and what this tells us about the use of woody borders for insect conservation and pest control. Insect conservation through fencerows Woody borders increase the structural and vegetation diversity in agricultural areas. For insects in these areas, field margins can be a source of increased habitat diversity (Morris & Webb, 1987). Thomas and Marshall (1999) found a correlation between arthropod diversity and floral diversity of field margins. Several studies have found greater insect diversity in Holland, J.D., and L. Fahrig. 2001 Landscape woody border increases insect diversity in alfalfa fields. In C. Barr and S. Petit (eds.) Hedgerows of the world: their ecological functions in different landscapes, Proceedings of the 2001 Annual IALE(UK) Conference, University of Birmingham, UK. Pp. 167-176. Hedgerows of the World, IALE(UK) 2001 168 hedgerows than in agricultural fields (Lewis, 1969; Bowden & Dean, 1977). This increase in diversity has been found to be consistent across a range of scales beyond the woody border itself. Dennis and Fry (1992) found increased insect richness in the part of the field adjacent to a hedgerow. Holland and Fahrig (2000) found an increase in insect family richness in fields situated in landscapes with more woody borders within 1km of the field than in fields with less woody border in the landscape. The consistency of these results is a strong argument for the conservation potential of woody borders. Insect conservation must be concerned with more than simply maximizing species richness. If we wish to preserve native habitats and species it may be important to see to what extent woody borders can aid in preserving species which inhabited the forest habitat that has largely been cleared for agriculture. This is not to suggest that woody borders can replace forest patches as refuges for these species. Studies have shown that forests and fencerows differ in structure and vegetation species (Fritz & Merriam, 1994, 1996), and in carabid beetle assemblages (Fournier & Loreau, 2001). Woody borders may however be important in providing dispersal opportunities for some forest species (Burel & Baudry, 1995). This would help maintain the species in the landscape through rescue effect (Brown & Kodric-Brown, 1977) and recolonization of local extinctions. The ability of field margins to contribute to such processes through dispersal depends on limiting the disturbances, such as grazing, within these margins (Charrier et al., 1997; Petit & Usher, 1998). The presence of woody vegetation could be important in providing protection from such disturbances if, for example, they limit access to the margin by grazing mammals. Entomophagous insects and woody borders The enhancement of insect predator species’ populations is one of the most commonly invoked reasons for the preservation of woody borders around crop fields (ex: Dennis and Fry, 1992; Dennis & Wratten, 1991; Hart et al., 1994; Sustek, 1992; Wratten, 1988). These studies have found that field margin habitats contribute to both the density and diversity of predacious species (Coombs & Sotherton, 1986; Kromp, 1999; Sotherton, 1985). Field margins are considered important in providing complementary habitat in the rich carabid beetle literature that shows these species over-wintering in sheltered field borders and foraging in adjacent crop fields (Wratten & Thomas, 1990; Thomas et al., 1998). Therefore the margins increase biodiversity and help to increase predator insects. However, there has been recent concern expressed regarding the lack of empirical evidence showing an actual decrease in herbivores due to the predators (Kromp, 1999). As well, the importance of trees and shrubs in these studies is not consistent. Some studies show these are important in the habitat of predators, while others show that the grass-herbacious layer is by far the most important (Dennis & Fry, 1992). Nicholls et al. (2001) however, recently found that a shrubby corridor through a vineyard increased predator densities by providing an alternate food to the variably available field herbivores, and that this did in fact lead to a decrease in herbivore density. As discussed above however, even if trees and shrubs are not important themselves in the habitat of some predators, they may act to preserve the lower grass/herbacious strata that are important. In a simulation study, Topping and Sunderland (1994) found that stable and diverse margins were best for maintaining high populations of predators. Woody vegetation could help to stabilize the field margin habitats by offering protection from some disturbances such as grazing, and increase structural diversity. Herbivorous insects and fencerows Examining the effects that woody borders may have on herbivorous insects presents additional challenges because the aims of insect conservation and agriculture may not result Hedgerows of the World, IALE(UK) 2001 169 in similar recommendations. Conservation requires that woody borders aid in maintaining viable populations of as many native insect species as possible. Agriculture demands that woody borders not lead to increased herbivory in crops, and preferably that they help to limit populations of crop pests. There are many different ways in which woody borders may have an effect on assemblages of herbivorous insects in crop fields. Field margins may offer complementary habitat (sensu Dunning et al., 1992; Morris & Webb, 1987) for some insect species for foraging (Bowden & Dean, 1977; Hawkes, 1973), over-wintering (Dennis & Fry, 1992), summer aestivation (Manglitz, 1958), and mating (Hawkes, 1973). Woody vegetation has been suggested to increase the value of the field margin for the various insect activities in several of these studies. Having complementary habitats in the landscape should lead to an increase in insect diversity because some species in crop fields will require the woody borders to complete some part of their life cycle. In addition to the possible role as complementary habitat, woody borders may function as the sole habitat for some herbivore species. Regardless of the mechanisms involved, the increase in habitat heterogeneity resulting from woody borders should lead to an increase in the diversity of herbivorous insects in the landscape. If woody borders are complementary habitat for some species and are exempt from disturbances within the adjacent field as mentioned above, they could act as refuges during disturbances such as pesticide application (Dyer & Landis, 1997; Marc et al., 1999; Powell, 1986; Reichart & Lockley, 1984), harvesting, and livestock grazing. This could increase density of species that utilize such refugia, and increase richness if some species rely on the woody borders to maintain viable populations. The effects that woody borders have on herbivore species in crop fields could be largely determined by the effect that they have on the dispersal of these species. Studies have shown that linear barriers (Mader et al., 1990), field margins (Jepson, 1994; Thomas et al., 1998), and woody borders in particular (Bowden & Dean, 1977; Frampton et al., 1995; Lewis, 1969; Mauremooto et al., 1995) can reduce insect dispersal. This could lead to two main effects on the movement of herbivores. The permeability of patch boundaries can be an important factor in determining emigration from the patch (Stamps et al., 1987). If woody borders act as movement barriers to some herbivore species then some dispersing individuals will remain in the crop field, and a population increase may result because some individuals will turn back into the crop field upon encountering the woody borders at the field edge. Bach (1988) found such an increase in a leaf beetle (Coleoptera: Chrysomelidae) in host plant plots surrounded by non-host vegetation. Some field margins may also make it harder for herbivores to find suitable crop fields (Fry 1994). These effects are not mutually exclusive. Although a reduction in movement could be a benefit to agriculture if it leads to reduced colonization of fields by herbivores, this must be balanced against a reduction in insect pollination, and a possible loss of diversity if some species are unable to maintain viable regional populations. Because agricultural areas are a dynamic mosaic of patches (Duelli et al., 1990) viable populations must continually repopulate local extinctions as crops in different fields are removed (den Boer, 1981; Duelli et al., 1990; Fahrig and Merriam, 1994; Hanski, 1994). There are several different effects that woody borders may have on herbivorous insects in agricultural fields, and associated with each is a predicted effect on both the insect richness and density. This makes it very difficult to predict the overall effect of woody borders on insect assemblages. Different agricultural disturbance regimes undoubtedly add an additional element of complexity to this problem. Therefore it is difficult to suggest ways in which to maximize insect conservation while minimizing insect damage to crops. We suggest that the best way to look at the overall effects of woody borders on insect assemblages is through Hedgerows of the World, IALE(UK) 2001 170 field studies on broad taxonomic groups. Such studies should be informed by studies on individual species, and well-parameterized simulation studies. Four studies in alfalfa fields Members of the Landscape Ecology lab at Carleton University have recently carried out four studies on insects in alfalfa fields using intensive sampling of fields within one landscape (Fahrig & Jonsen, 1998), a computer simulation (Bhar & Fahrig, 1998), and focal patch approaches (Jonsen & Fahrig, 1997; Holland & Fahrig, 2000). Focal patch studies consider the effect of predictor variables measured within a landscape (amount of woody border, for example) on the response variable, which is measured within a patch embedded in the larger landscape (Brennan et al., in press). The individual landscape (and corresponding focal patch) are the unit of replication in this type of study (Brennan et al., in press; figure 1).