Impacts of intensive logging on the trophic organisation of ant communities in a biodiversity hotspot

Abstract

Trophic organisation defines the flow of energy through ecosystems and is a key component of community structure. Widespread and intensifying anthropogenic disturbance threatens to disrupt trophic organisation by altering species composition and relative abundances and by driving shifts in the trophic ecology of species that persist in disturbed ecosystems. We examined how intensive disturbance caused by selective logging affects trophic organisation in the biodiversity hotspot of Sabah, Borneo. Using stable nitrogen isotopes, we quantified the positions in the food web of 159 leaf-litter ant species in unlogged and logged rainforest and tested four predictions: (i) there is a negative relationship between the trophic position of a species in unlogged forest and its change in abundance following logging, (ii) the trophic positions of species are altered by logging, (iii) disturbance alters the frequency distribution of trophic positions within the ant assemblage, and (iv) disturbance reduces food chain length. We found that ant abundance was 30% lower in logged forest than in unlogged forest but changes in abundance of individual species were not related to trophic position, providing no support for prediction (i). However, trophic positions of individual species were significantly higher in logged forest, supporting prediction (ii). Consequently, the frequency distribution of trophic positions differed significantly between unlogged and logged forest, supporting prediction (iii), and food chains were 0.2 trophic levels longer in logged forest, the opposite of prediction (iv). Our results demonstrate that disturbance can alter trophic organisation even without trophically-biased changes in community composition. Nonetheless, the absence of any reduction in food chain length in logged forest suggests that species-rich arthropod food webs do not experience trophic downgrading or a related collapse in trophic organisation despite the disturbance caused by logging. These food webs appear able to bend without breaking in the face of some forms of anthropogenic disturbance.

Figure 1. Trophic organisation of ant communities in unlogged and logged forest.

Trophic positions of (a) colonies and (b) species in unlogged and logged forest are grouped into trophic categories of 0.5 trophic levels (<2.25, 2.25–2.75, 3.25–3.75 etc.). Bubble sizes represent the proportion of colonies or species in each trophic category for each type of forest (i.e. expressed as a percentage of the total number of colonies [a] or species richness [b] for each forest type).

Figure 2. Causes of differences in trophic organisation of ant communities between unlogged and logged forest.

Mean trophic positions±S.E. are shown for species found in both unlogged (UL) and logged (L) forest, for species found only in logged forest, and for species found only in logged forest. Means are based on all samples and presented for all species, and for the two commonest ant subfamilies (Myrmicinae and Ponerinae).

Figure 3. Trophic position of species commonly sampled in unlogged and logged forest.

Mean trophic position±S.E. are shown for all species sampled at least 10 times in both types of forest. p values are: *≤0.05, **≤0.01, ***≤0.001. From left to right, species are: Lophomyrmex bedoti, Hypoponera sp1, Ponera sp4, Anochetus graeffei, Pachycondyla sp3, Pheidole rabo, Tetramorium sp2, Oligomyrmex sp1, Hypoponera sp7, Pheidole tjibodana, Paratrechina sp2, Strumigenys sp1, Strumigenys fuarda, Eurhopalothrix jennya.