Maliau pitfall project

From BiodivBorneo09

Jump to: navigation, search

Terrestrial Arthropod Order Richness and Composition in Primary and Secondary Forest and Gap of Maliau Basin

Biofagri A.R., Molly Rooney, Serena Zhao



The debate in Borneo over land use for secondary forest is complicated by conflicting information on richness and composition of biodiversity within these sites. Within all forest types, Arthropods play vital roles, occupying various niches and serving as indicators. We conducted a study using pitfall traps to compare the order-level diversity of arthropods in primary and secondary forests in Maliau Basin. Pitfall trap samples from primary forest had greater order richness than did secondary sites, and all traps were roughly the same in evenness of order abundances. Meanwhile, diversity of gap traps were indistinguishable from diversity of forest traps, indicating that diversity differences result from differences in forest communities rather than abiotic factors. As succession progresses after logging, arthropod diversity will remain stunted as pioneer vegetation dominates. We recommend preservation of primary forest


Recently in Borneo there has been controversy surrounding land use of secondary forest. Specifically, there is a debate over whether secondary forest should be used for oil palm plantations or whether it is worth preserving or harvesting sustainably because of its biodiversity. Much of this argument hinges on the species diversity of these secondary forests relative to primary forests.

Although this information is key to settling this debate, the species richness and composition of secondary forests compared to primary forests is not fully understood. Recent research has suggested that secondary forest may be important for large animals. For instance, Mark Ancrenaz of the Kinabatangan Orang-Utan Conservation Project believes that orangutans may actually be more successful in secondary forests due to the greater number of fruiting trees in secondary forests (lecture by Ancrenaz 2009). Although it is becoming clear that secondary forests may well support mammalian life, the composition and richness of other important taxa in secondary forests relative to primary forests remains unclear.

Arthropods are highly abundant in forest communities and occupy many niches within forest food chains. Besides being major engineers and potential regulators of ecosystem conditions (Schowalter, 2000), the rapid response of arthropods to environmental change makes them useful indicators as well (Peltonen et al., 1997). As environmental indicators, changes in arthropod communities resulting from environmental change or other disturbances reflect a forest community as a whole.

One study conducted in Costa Rica found that arthropod assemblages in epiphyte mats "were thinner and less structurally diverse in secondary forest" (Yanoviak et al., 2007). Schonberg's study looking at arthropod canopy communities tropical montane landscape in the neotroprics (2004) found that there was no difference in species richness between secondary and primary canopy arthropod communities, whilst there was lower arthropod density in secondary forests. Another study conducted on moths in Mount Kinabalu National Park found that there was no distinguishable difference between moth species composition of disturbed primary, old-growth primary, and undisturbed primary forest (Fiedler et al., 2004). These studies present conflicting information on the effect of disturbance on arthropod communities. Although differing results have been found for terrestrial and aerial arthropods, Stork and Brendell (1990) reported that 70% of all arthropods inhabit the soil and leaf litter in the rainforest ecosystem in Southeast Asia. Therefore we believe that studying terrestrial arthropods will be most indicative of forest ecology.

To see what effect the forest stage and cover has on the richness and composition of terrestrial arthropod, we conducted a study in Maliau Basin, home to one of the few vestiges of virgin forest in Borneo. While small patches in the structure have been logged, the interior and edges of the basin remain untouched, providing a prime example of analogous primary and secondary forests. We aim to see whether there is a difference in richness and composition of terrestrial arthropods in closed forest and gap environments of primary and secondary forest. We predict that terrestrial arthropod richness and composition will vary significantly between locations, and that closed canopy primary forest patch will have the highest terrestrial arthropod richness. We also expect to see significant differences in order composition between site types. This prediction is based on the motility and preferences of arthropods. Because many insects live in decomposing litter, drier litter in open forest patch (which resulted from more sunlight exposure) will attract fewer insects (Iskandar, 2004). Other study stated that canopy gaps have the lowest mite density and diversity, and are more sensitive to year-to-year climatic changes than closed canopies (Marra and Edmonds, 2005). This sensitivity of gaps could lead to more arthropod extinction in gaps and therefore lower diversity. Also, Goehring et al. (2002) stated that more disturbances decrease arthropod morphospecies richness and the number of unique species in all forest types.


We sampled terrestrial arthropods in primary forest and secondary forest(twelve years of growth after logging) close to the Maliau Basin Research Centre and near Maliau River in the Maliau Basin. We set 20 pitfall traps to measure terrestrial arthropod presence. We set five pit falls in each: primary forest, primary gap, secondary forest, and secondary gap. At each site we placed three plastic cups half-filled with detergent water and covered with cardboard rain awnings in an equilateral triangle. Trap locations were chosen using a random walk method along the separate trials in the secondary and primary forest. The pit falls were in place approximately 36 hours before collection.

To assess the arthropod order richness and composition we used the following method on each of the 20 samples:

  • We shook contents of the pitfall in tray for 10 seconds.
  • We separated the sample into four subsample areas within the tray in a consistent diamond arrangement using a circular plastic barrier.
  • We counted presence or absence of each order within the subsamples. The presences of orders within the subsamples were then added to show abundance of orders in each sample. Absolute counts of individuals of each order were overlooked in favour of presence/absence tallies because we were not able to account for individuals that had degraded beyond classification.

The composition and order richness were compared between primary forest, primary gap, secondary forest, and secondary gap. We conducted an analysis of variance on the arthropod samples and checked for correlations between any of the locations.


To analyze the pitfall data, we first compared the order richness between all site types using an ANOVA test. The null hypothesis is there is no difference in richness between sites. As stated in the introduction, we predict however that we will see greater order richness in primary forests. We also specifically compared the difference in richness between primary and secondary forest using a t-test with the same null hypothesis.

To check for a difference between gap and forest order richness we used t-tests(the total number of orders was normally distributed within each site. We compared order richness between forest and gap, secondary forest and gap, and primary forest and gap. The null hypothesis is there is no difference in order richness between forest and gap, overall or specifically within primary or secondary.

We then compared abundance of orders, evenness, between site types using the diversity function in the vegan package. The null hypothesis is there is no difference in evenness between site types.

To investigate composition differences, we used an ANOVA test comparing the abundances of each order between secondary and primary forest. For each order, the null hypothesis is there is no difference in abundance of that specific order between site types. In our introduction we predict we will see a difference in composition between sites so we expect to see a difference in abundance between site types for some orders.


We found that the orders richness of primary traps and secondary traps is significantly different (p = 0.01), with an average of 8.7 orders per pit in the primary sites and 5.7 orders per pit in the secondary sites. Thus, primary forests have a greater richness of orders than secondary forests.

Based on our ANOVA analysis, we found that the only three orders varied significantly in abundance between secondary and primary forests: hymenoptera (p=.082), hemiptera (p=.063), and orthoptera (p=.027). There were, in total,16 orders represented.

All traps were distributed with evenness values between .85 and .95. There is no significant difference in sample evenness between traps grouped by the four sites. Therefore, primary and secondary forests have similar patterns of rank-abundance relationships. The subsampling method for counting order presence and absence decreases the representation of the most abundant orders while increasing the representation of rarer orders. However, while order evenness is distorted at the trap level, all of the traps are similarly distorted, allowing for comparison across the sites.

Overall, forest and gap do not differ significantly in order richness (p=0.5784). Primary gap and forest species richness is not significantly different (p=0.3330). Similarly, secondary gap and forest richness is not significantly different (p=0.8726).


Since all of the traps have roughly the same evenness in abundance, no order disproportionately dominates the community at each trap. Each order has roughly equal representation in accounting for diversity. Therefore, the primary traps, with a greater average number of orders per trap, show greater diversity than the secondary traps.

The three orders that showed differences in abundance between primary and secondary sites may reflect the organisms' adaptations to highly specific environments: Orthoptera, which inhabit grass and airy spaces, were found to be more abundant in secondary traps. Further investigation into life histories of Hymenoptera and Hemiptera, which were less significantly different between forest types, would show which factors in the habitat caused increased or decreased populations. However, these three orders do not contribute to a sizable difference in the sampling of the sixteen orders.

Within primary sites, within secondary sites, and across both types, order diversity does not differ significantly between forest and gap traps. This implies that abiotic factors caused by gaps, such as increased light and heat, are not sufficient to change arthropod composition within a forest type. Therefore, the differences in arthropod communities result from the different organisms in each habitat, rather than merely physical conditions. Even thickly regrown secondary forests that are no longer sparse and light-filled contain plant communities different from those of primary forests. Since arthropod diversity relies on the other organisms in the ecosystem, low diversity will remain in secondary forests even as they fill in.

Once forest has been logged, even as vegetation fills in again, the arthropod community will remain in its perturbed state, for a minimum of twelve years, the length of time the Maliau secondary forest has been recovering. Because arthropods occupy so many diverse niches, their prolonged disturbance indicates a lasting impact on logged forest communities. We do not yet know the duration of successional change sufficient to restore forest arthropod communities to pre-logging compositions. Land-use management, therefore, cannot rely on logged secondary forest to regenerate to primary forest conditions on an economic timescale, and the long-term resources forgone by logging primary forest can be considered lost fr the purposes of one generation. However, while an individual human may not live to see the full fruition of successional processes on secondary forest, this does not diminish the need to also protect secondary forest from clearing. With arthropod order abundance evenness indistinguishable from that of primary forests, secondary forests still host diverse communities, with some organisms (such as Orthoptera) present in greater abundance in secondary forests. Therefore, we advocate the preservation of primary forest to prevent potentially irreversible damage, while preventing wanton degradation of differently-rich secondary forests.


Fig. 1 We found that the order richness of primary forest(PF) and secondary forest(SF) is significantly different (p-.01).

Works Cited

  • Angulo-Sandoval, P., Fernandez-Marin, H., Zimmerman, J. & Aide, T. Changes in patterns of understory leaf phenology and herbivory following hurricane damage. BIOTROPICA. 2004, Vol. 36(1), pp. 60-67
  • Ramirez, A. & Hernandez-Cruz. L. Aquatic insect assemblages in shrimp-dominated tropical streams, Puerto Rico. BIOTROPICA. 2004, Vol. 36(2), pp. 259-266
  • Schonberg, L., Longino, J., Nadkarni, N., Yanoviak, S. & Gering, J. Arboreal ant species richness in primary forest, secondary forest, and pasture habitats of a tropical montane landscape. BIOTROPICA. 2004, Vol. 36(3), pp. 402-409
  • Tsvuura, Z., Griffiths, M.E. & Lawes, M.J. The effect of herbaceous understory cover on fruit removal and seedling survival in coastal dune forest trees in South Africa. BIOTROPICA. 2007, Vol. 39(3), pp. 428-432
  • Pollination Ecology and the Rain Forest: Sarawak Studies. Roubik, D., Sakai, S. & Hamid, A. A. (ed.) Springer, 2005
  • Goehring, D.M., Daily, G.C., Sqekerçiog˘lu, Ç. H. 2002. Distribution of ground-dwelling arthropods in tropical countryside habitats. Journal of Insect Conservation 6: 83–91
  • Iskandar, D.T. 2004. The Amphibians and Reptiles of Malinau Region, Bulungan Research Forest, East Kalimantan: Annotated checklist with notes on ecological preferences of the species and local utilization. Center of International Forestry Research, Jakarta.
  • Marra, J.L. & Edmonds, R.L. 2005. Soil Arthropod Responses to Different Patch Types in a Mixed-Conifer Forest of the Sierra Nevada. Forest Science 51(3)
  • Stork, N.E. & Brendell, M.J.D. 1990. Variation in the insect fauna of Sulawesi trees with season, altitude and forest type. – In: Knight, W.J. & Holloway, J.D. (eds.), Insects and the rain forests of South East Asia (Wallacea): 173-190. Royal Entomological Society of London, London.
  • Schowalter, T.D. 2000. Insect ecology: an ecosystem approach. Academic Press, San Diego.
  • Peltonen, M., Heliövaara, K., Väisänen, R. 1997. Forest insects and environmental variation in stand edges. Silva Fennica 31:129-141