Beetle Assemblage Responses to Rainforest Deforestation and Reforestation in North-Eastern Australia

Abstract

Insects are well recognised as being the major contributor to global biodiversity, and for their critical involvement in many biotic interactions. Most of the insect diversity is found in tropical rainforests. However, these forests are threatened by high rates of clearing and the subsequent fragmentation of remaining habitat. The effects on biota, particularly insects, are poorly understood as are the mechanisms mediating faunal changes. Reforestation could potentially alleviate some of the deleterious effects of forest loss and fragmentation. However, because reforestation is a relatively new endeavour, it is little known just how much insect biodiversity can be supported by reforestation and what factors influence insect colonisation. These issues were investigated in the Atherton Tablelands of north-eastern Australia, a landscape whose rainforest has been heavily cleared and fragmented over the last 80 years, but is also the focus of reforestation efforts. To quantify the effects of rainforest loss and fragmentation, pasture sites were compared with small rainforest fragments, and with the edges and interiors of large rainforest fragments (24 sites in total). Sites with replanted rainforest (planted with a high diversity of plants) were also included. These varied in their age (2-17 yrs) and their distance (0-4.5 km) from existing rainforest (24 sites in total). Another set of reforested sites was also studied but these differed in their planting style (number of plant species, spacing etc). This second set of sites was located in two regions: the Atherton Tablelands (50 sites), and in the subtropics of eastern Australia (54 sites). At each site, beetle assemblages were surveyed using methods that sample beetles near the ground (four ground-based flight interception traps in the first set of sites and ten pitfall traps in the second set of sites), and then the assemblages among site-types were compared. Over 32,000 beetles were caught and identified to the level of family, and of these, 15,206 were identified further to the level of species. Very few beetle species were present in pasture, suggesting that converting rainforest into pasture has a very strong negative effect on beetle diversity and species composition. Irrespective of rainforest fragment size, beetle species composition in drier rainforest habitats was different from that of moister rainforest. Beetle species composition also differed between small remnants and interior rainforest: drier-associated species were more abundant in small remnants, whereas wetter-associated species were more abundant in interiors. This pattern can be best attributed to a fragmentation effect mediated by differences in microclimate. With the exception of differences between rainforest and pasture, these results were generally not observed among beetle assemblages identified to coarser taxonomic groups (family, feeding guild, and body size). Among replanted rainforest, older sites and those adjacent to rainforest had a more rainforest-like beetle species composition. However, even the closer and older sites had a substantially lower abundance and richness of rainforest-associated beetles than did rainforest. Age effects were generally stronger than distance effects. Beetle assemblage similarity to rainforest was more strongly correlated with structural similarity to rainforest than with site age or distance from rainforest. Thus the use of revegetation techniques which lead to more rainforest-like structural conditions appears to be of over-riding importance in catalysing the rapid acquisition of rainforest beetle assemblages in the initial stages of restoration. Nevertheless, not all beetle species were equally affected by the factors tested. Large-bodied beetle species (>5 mm) were more strongly influenced by distance than small-bodied species (<5 mm), suggesting that small-bodied species are better dispersers, and thus are amongst the first to colonise new habitats. Spatial ubiquity in rainforest was not a good predictor of a species' dispersal ability. Interestingly, fewer of the broader groups (family, feeding guild, and body size) showed the response to distance evident at the species level although they showed differences between reforested sites differing in age, and between reforested and reference site-types. Therefore, these results and those from the fragmentation study suggest that information at the species level is more sensitive to environmental change than data identified to a coarser level of taxonomy or grouped according to feeding ecology or body size. For the pitfall-trapped beetles in the second reforested site network, beetle assemblages in all styles of reforestation were intermediate in species composition between pasture and rainforest. The similarity of beetle assemblages to intact rainforest increased with the age and structural complexity of reforested sites, although again structural complexity appeared to be of overriding importance. This study has shown that even small patches of rainforest and reforested areas can support diverse rainforest-dependent beetle assemblages. A range of factors influence the development of beetle assemblages in reforested sites although not all species are equally affected. However, even structurally complex reforested sites cannot provide a short- or medium-term substitute for the retention of intact rainforest.