In addition to these local effects, shelter-builders may have larger regional effects by influencing population dynamics. For example, because many species of shelter-building caterpillars are bivoltine or multivoltine, shelter-provisioning by one generation can facilitate establishment of future generations of both conspecifics and heterospecifics (Cappuccino 1993) resulting in regional differences in caterpillar densities and diversities. Such feedback loops between engineers have been proposed by Jones et al. (1997) and modeled by Gurney and Lawton (1996). We suggest that the rapid generation times of these insects make them ideal candidates for empirical study; such studies will provide much-needed parameters useful in testing and refining these models.
While multiple studies have indicated that shelter-building tends to increase arthropod species richness, the spatial scales used in sampling differ among studies, leaving open the question of how the engineering effects on species richness change with the number of habitat "patches" sampled (comparing engineered and nonengineered habitat patches). Studies have found strong effects of shelters on arthropod richness at relatively small scales (saplings or branches of larger trees with a rela tively small number of habitat patches; Martinsen et al. 2000, Lill and Marquis 2003), but at larger scales (i.e., landscapes) these engineering effects on richness might be expected to attenuate, because the proportion of obligate shelter-users is relatively small compared with non-shelter-using arthropod species in the regional fauna.
If shelter-building has a compounding effect on plant damage by attracting more herbivores, which has been shown in some systems (Marquis et al. 2002, Marquis and Lill unpublished data), and damage influences plant fitness, then shelter-builders have the potential to alter plant community structure. If plant damage is concentrated on a dominant competitor, engineers could indirectly increase plant community diversity (or decrease diversity if focused on inferior competitors). A variety of studies have documented the strong engineering effects of herbivores on plant communities (e.g., Wilby et al. 2001), but most of these have focused on soil disturbance by animals and its influence on recruitment. Few shelter-builders appear to routinely defoliate plants (outbreaking species in Table 6.1 are exceptions), but the successive use of shelters by other herbivores has the potential to compound plant damage over the season.
Was this article helpful?