The importance of termites in tropical ecosystems

The importance of termites in tropical ecosystems

Termites (Insecta: Isoptera) are predominantly tropical in distribution. Their species richness is highest in lowland equatorial rain forests, and generally declines with increasing latitude and altitude (Collins 1983; Eggleton et al., 1994; Jones 2000). Termite survival is limited by low temperatures and high aridity, and very few species occur beyond 45° latitude (Collins 1989) or above 1800m altitude (Collins 1980; Jones 2000). Recent studies (Eggleton 2000) indicate that the forests of West Africa have the highest termite diversity, closely followed by South America. The forests of Southeast Asia and Madagascar are considerably lower in species richness. These regional diversity anomalies are also associated with significant differences in clade and functional diversity (Eggleton 2000; Davies, 2001).

Termites are at the ecological centre of many tropical ecosystems (Wilson 1992). They can achieve very high population densities. For example, in the forests of southern Cameroon, termites are one of the most numerous of all arthropod groups (Watt et al., 1997) with abundances of up to 10,000 m^-2, and live biomass densities up to 100 g m^-2 (Eggleton et al., 1996). Across the Isoptera, a wide range of dietary, foraging and nesting habits occur, but many species show a high degree of resource specialization (Wood 1978; Collins 1989; Sleaford et al., 1996). A few species feed on living plant tissue but most are detritivores, feeding on dead plant material along a humification gradient, from dead wood and leaf-litter to humus in the soil (Donovan et al., in press). As the dominant arthropod detritivores, termites are important in decomposition processes (Wood & Sands 1978; Matsumoto & Abe 1979; Collins 1983) and play a central role as mediators of nutrient and carbon fluxes (Jones 1990; Lawton et al., 1996; Bignell et al., 1997; Tayasu et al., 1997; Eggleton et al., 1999). Termite activity, such as mound-building, subterranean tunnelling and soil-feeding, is thought to have a positive effect upon soil structure and quality (Lee & Wood 1971; Lobry de Bruyn & Conacher 1990; Black & Okwakol 1997; Holt & Lepage 2000; Donovan et al., 2001). To date, about 2,650 species of termites have been described (Kambhampati & Eggleton 2000), and less than 3% of these cause significant economic damage to buildings or related human-made structures (Pearce 1997). A similar proportion are serious pests of crops (Wood 1996). The termite fauna of urban environments is usually depauperate and characterised by wood-feeding species, unlike natural habitats that often support great termite diversity. For example, 136 species have been recorded in a single forest site in Cameroon, 73% of which feed on soil (Jones & Eggleton 2000). The impact of termites on ecosystem processes in natural habitats and in agroforestry systems is likely to be governed by the species composition and abundance of the local termite assemblage. Therefore, to quantify the influence of termites, it is necessary to accurately characterize the structure of that assemblage. In response to the need for standardized methods for sampling insects (Sutton & Collins 1991; Stork & Samways 1995) a termite sampling protocol has been developed. The protocol, described and tested by Jones & Eggleton (2000), produces samples that are representative of the taxonomic and functional diversity of the local termite assemblage. The protocol is based on standardised sampling effort, thus ensuring the samples from different sites are directly comparable.

Termite studies in Borneo

Borneo is the world’s third largest island, and is divided politically between Indonesia, Brunei Darussalam, and the Malaysian states of Sabah and Sarawak. Almost everything published on the termites of Borneo, whether ecological or taxonomic, describes work undertaken in the north and north-west of the island. Kalimantan covers 73% of the island but the ecological structure and species composition of its termite fauna has never been studied. Our current understanding of Bornean termite assemblages is based almost entirely on studies conducted at just four sites (Figure 13). Danum Valley, in southeast Sabah, has arguably the most intensively studied termite fauna on Borneo, due to the investigations of Eggleton et al., (1997, 1999) and Homathevi (1999). To date, a total of 93 species have been recorded from the Danum Valley area, the highest termite species richness of any site in Southeast Asia (Jones & Eggleton, 2000). The other significant assemblage-level studies have been conducted at Mulu, northern Sarawak (Collins, 1980, 1983, 1984), Maliau Basin, south-central Sabah (Jones et al., 1998; Jones 2000) and Belalong, Brunei (Jones 1996). The taxonomic literature has a similar geographical bias, with most species descriptions and locality records originating from collections made in Sabah (Thapa, 1981) and Sarawak (e.g. Haviland 1898; Ahmad 1968). The survey reported here will help to address this imbalance.

Bioindicators and the assessment of sustainable forestry

The field of bioindication has expanded rapidly over the past two decades as scientists and environmental managers have sought efficient ways of assessing the impact of anthropogenic disturbance on ecosystems. However, the subject is seen by many to lack scientific rigour, partly because policy initiatives are running far ahead of ecological knowledge. Recent reviews (McGeoch 1998; Caro & O’Doherty 1999; Hilty & Merenlender 2000; Lindenmayer et al., 2000) have highlighted many practical short comings, and emphasised two themes that are central to the successful implementation of bioindication. First, the scientific objectives for which the bioindicator is to be used must be clearly defined, as this will determine the criteria employed when selecting suitable bioindicators. Adopting general objectives such as monitoring “ecosystem health” has provoked a heated debate in the literature (e.g. Callicott & Mumford 1997; Simberloff 1998) because the term can be defined in many ways and may be difficult to quantify and interpret. Second, many studies have failed to test the predictive power of the species used, with the result that the efficacy of these species as bioindicators can be questioned. To be accorded the status of bioindicator, a species must indicate a relationship with another biotic or abiotic variable. This relationship should be statistically strong so as to provide a spatially and temporally robust prediction. Unfortunately, the detailed knowledge of species responses that is required to quantify these relationships is not always available, and extrapolation from limited ata may prove spurious when relationships between and within biotic and abiotic variables are non-linear.