Lambert and Collar (2002) recently provided a list of Sundaic bird species that are most likely to be affected by the processes of logging, fragmentation, and hunting. Terrestrial and understorey insectivorous species were particularly vulnerable to the effects of logging, and to a lesser extent some insectivores, particularly sallying species, that inhabit the lower and mid-levels of the forest. Birds typical of the canopy appeared to be much more resilient, and, with the exception of the highly specialized Green Broadbill (Calyptomena viridis), frugivorous and nectarivorous species were rarely suspected of declining in logged forests. The groups that are most affected by logging comprise: (1) some extreme lowland specialist species, because logging in these forests is most intense (Black Hornbill Anthracoceros malayanus, Crestless Fireback Lophura erythrophthalma), (2) nomadic species, or species requiring large areas (hornbills, raptors), (3) primary forest species, intolerant of logging disturbance (Great Argus Argusianus argus, some trogons, some woodpeckers, some babblers, and some flycatchers), and (4) species that require large cavities for nesting in.
The most seriously affected taxa are those species of the primary forest interior. Terrestrial insectivores and low- to mid-understorey flycatchers are consistently vulnerable to logging practices (Johns 1989; Lambert 1992a, b; Thiollay 1992). Among avian understorey insectivores, both the number of species present and their proportional representation within the population sample changes following logging at moderate intensities. Lambert (1992a) identified, among others, trogons Harpactes spp., woodpeckers (Picidae), wren babblers (Kenopia striata and Napothera spp.), and flycatchers (Cyornis spp. and Ficudela spp.) as prone to decline in logged forests. Declines are believed to reflect the loss of understorey vegetation, foraging substrata and the associated cryptic insect prey that understorey insectivores specialize on (Robinson 1969). However, physiological conditions (i.e., high temperature and water stress) appear more significant than local food abundance in determining the ranging of some understorey species in Panamanian forests (Karr & Freemark 1983). Microclimatic conditions are altered by loss in canopy cover and understorey species are often reluctant to cross open spaces or dense second growth that separates remaining patches of undisturbed forest. Forest interior insectivores may also be displaced by flycatcher species more characteristic of edge habitat, or by the invasion of bulbuls (Pycnonotidae). These transient frugivorous species are associated with rapidly growing pioneer species, but also include insects in their diet. The fruits found in regenerating logged forests may support higher abundances of insectivore-frugivores, such as bulbuls, and result in increased competition with specialized insectivores, but evidence is lacking.
Frogs and toads
Frogs and toads (= anurans) display a large diversity of reproductive modes with complex life cycles that include a larval form (tadpoles) requiring different habitats and resources than the adult forms (e.g., Inger & Voris 2001). Some Bornean species of amphibians carry out almost their entire life cycles in leaf litter. Other species inhabit the lower canopy only, using small vines and other water holding plants and substrates for egg deposition and tadpole development. Anurans may therefore be sensitive indicators of habitat quality because in order to complete their life cycles they may require a greater variety of specialized resources in a specified order.
Logging may directly and indirectly affect anurans through its impact upon ponds and streams in the immediate vicinity of logging operations, on watersheds throughout the area, and through impacts on microhabitat conditions in the forest leaf litter in non-riparian habitats. Streams and ponds provide microhabitats for frogs and most frogs require aquatic habitats for reproduction. Frog species that rely on water held in plants and treeholes will be heavily impacted by the direct loss of sites to deposit eggs. During development and metamorphosis, larval frogs are sensitive to changes in water chemistry and physical characterisitics of aquatic habitats. Research in Sabah showed that destruction of the lower vegetation and destruction of the floor litter leads to extreme levels of erosion during rains with enormous increase in sedimentation levels in streams, while also affecting water chemistry (Douglas et al. 1992; 1993). Sedimentation has a clear, detrimental effect on reproduction in several species of stream-breeding frogs, effectively preventing survival of tadpoles (R. Inger, in litt., 3 October 2003). Changes in nutrient composition, pH levels, and temperature may lead to malformations and death among larval anurans (Feder & Burggren 1992; Ouellet et al. 1997).
Virtually all leaf litter anurans depend upon the structural but constant diversity of the leaf litter microhabitat for predator escape, nesting, and upon the diversity of leaf litter arthropods for food (Scott 1976; Vitt & Caldwell 1994, 2001). Logging, primarily through the creation of gaps, drastically affects the ground level environment of tropical forests through changes in incident light, temperature, humidity, and rates of litter accumulation. The removal or change in quality of the leaf litter directly and indirectly affect anurans through loss of microhabitats that provide refuges from predators and dessication and through reduction in food resources (Zou et al. 1995); the greatest diversity of leaf litter arthropods was in the mature forest. The effect of variation in leaf litter arthropods – the primary food of leaf litter anurans – on growth and reproduction is virtually unstudied in the tropics (see Vitt & Caldwell 1994).
Evidence for amphibian declines following logging was provided by Iskandar (1999a, b), who studied this topic in North Sumatra. He found that logged forests contained only about 20% of the individuals found in an unlogged forest of the same area. Abundance was strongly correlated with the amount of forest litter. As opposed to these findings, Iskandar (2003) reported that in areas in Malinau where logging was performed in 1995–1996, the relative abundance of amphibians and reptiles was about 2.1 specimens per plot (n = 21), but in unlogged forest the relative abundance was only about 0.95 specimen per plot (n = 51). At present, there is no explanation as to why the unlogged area has fewer specimens compared to other areas affected by logging, although the species composition in these areas is more or less the same. One possible explanation is that both the reduced impact logging and conventional logging plots in Malinau still had more or less closed canopy forest, and that the difference between these plots and unlogged forest was small. In the North Sumatran study mentioned above, every large tree had been cut in the logged areas.
In conclusion, the wide range of ecological specializations in Bornean frogs makes it hard to generalize on the effect of logging on species richness and abundance, but it appears that at least some species are little affected by logging. Clearly, species such as Fejervarya cancrivora and Rana nicobariensis cope well with logging as these are commensals of man and not rain forest species. Whether any forest species are truly unaffected by logging remains unclear.