Changes and Disturbance in Tropical Rainforest in South-East Asia

The following sections are included:

• Preface

• Contents

Investigations of land management impacts on hydrology are well developed in South-East Asia, having been greatly extended by national organizations in the last two decades. Regional collaborative efforts, such as the ASEAN–US watershed programme, have helped develop skills and long-running monitoring programmes. Work in different countries is significant for particular aspects: the powerful effects of both cyclones and landsliding in Taiwan, the significance of lahars in Java, of small-scale agriculture in Thailand and plantation establishment in Malaysia. Different aid programmes have contributed specialist knowledge such as British work on reservoir sedimentation, Dutch, Swedish and British work on softwood plantations and US work in hill-tribe agriculture. Much has been achieved through individual university research projects, including PhD and MSc theses. The net result is that for most countries there is now good information on changes in the rainfall–run-off relationship due to forest disturbance or conversion, some information on the impacts on sediment delivery and erosion of hillslopes, but relatively little about the dynamics and magnitude of nutrient losses. Improvements have been made in the ability to model the consequences of forest conversion and of selective logging and exciting prospects exist for the development of better predictions of transfer of water from the hillslopes to the stream channels using techniques such as multilevel modelling. Understanding of the processes involved has advanced through the detailed monitoring made possible at permanent field stations such as that at Danum Valley, Sabah.

Dynamics of the Pasoh forest in Peninsular Malaysia were assessed by drawing a comparison with a forest in Panama, Central America, whose dynamics have been thoroughly described. Census plots of 50 ha were established at both sites using standard methods. Tree mortality at Pasoh over an eight-year interval was 1.46% yr⁻¹ for all steins ≥10 mm diameter at breast height (dbh), and 1.48% yr⁻¹ for stems ≥ 100 mm dbh. Comparable figures at the Barro Colorado Island site in Panama (BCI) were 2.55% and 2.03%. Growth and recruitment rates were likewise considerably higher at BCI than at Pasoh. For example, in all trees 500–700 mm in dbh, mean BCI growth over the period 1985–1995 was 6 mm yr⁻¹, whereas mean Pasoh growth was about 3.5 mm yr⁻¹. Examining growth and mortality rates for individual species showed that the difference between the forests can be attributed to a few light-demanding pioneer species at BCI, which have very high growth and mortality; Pasoh is essentially lacking this guild. The bulk of the species in the two forests are shade-tolerant and have very similar mortality, growth and recruitment. The Pasoh forest is more stable than BCI's in another way as well: few of its tree populations changed much over the eight-year census interval. In contrast, at BCI, over 10% of the species had populations increasing or decreasing at a rate of >0.05yr⁻¹ (compared to just 2% of the species at Pasoh). The faster species turnover at BCI can probably be attributed to severe droughts that have plagued the forest periodically over the past 30 years; Pasoh has not suffered such extreme events recently. The dearth of pioneer species at Pasoh is associated with low-nutrient soil and slow litter breakdown, but the exact mechanisms behind this association remain poorly understood.

Ten years' hydrological investigations at Danum have provided strong evidence of the effects of extremes of drought, as in the April 1992 El Niño southern oscillation event, and flood, as in January 1996. The 1.5 km² undisturbed forest control catchment experienced a complete drying out of the stream for the whole 1.5 km of defined channel above the gauging station in 1992, but concentrated surface flow along every declivity from within a few metres of the catchment divide after the exceptional rains of 19 January 1996. Under these natural conditions, erosion is episodic. Sediment is discharged in pulses caused by storm events, collapse of debris dams and occasional landslips. Disturbance by logging accentuates this irregular regime. In the first few months following disturbance, a wave of sediment is moved by each storm, but over subsequent years, rare events scour sediment from bare areas, gullies and channel deposits. The spatial distribution of sediment sources changes with time after logging, as bare areas on slopes are revegetated and small gullies are filled with debris. Extreme storm events, as in January 1996, cause logging roads to collapse, with landslides leading to surges of sediment into channels, reactivating the pulsed sediment delivery by every storm that happened immediately after logging. These effects are not dampened out with increasing catchment scale. Even the 721 km² Sungai Segama has a sediment yield regime dominated by extreme events, the sediment yield in that single day on 19 January 1996 exceeding the annual sediment load in several previous years. In a large disturbed catchment, such road failures and logging-activity-induced mass movements increase the mud and silt in floodwaters affecting settlements downstream. Management systems require long-term sediment reduction strategies. This implies careful road design and good water movement regulation and erosion control throughout the logging process.

Changes in species composition in two 4-ha plots of lowland dipterocarp rainforest at Danum, Sabah, were measured over ten years (1986–1996) for trees ≥10cm girth at breast height (gbh). Each included a lower-slope to ridge gradient. The period lay between two drought events of moderate intensity but the forest showed no large lasting responses, suggesting that its species were well adapted to this regime. Mortality and recruitment rates were not unusual in global or regional comparisons. The forest continued to aggrade from its relatively (for Sabah) low basal area in 1986 and, together with the very open upper canopy structure and an abundance of lianas, this suggests a forest in a late stage of recovery from a major disturbance, yet one continually affected by smaller recent setbacks.

Mortality and recruitment rates were not related to population size in 1986, but across subplots recruitment was positively correlated with the density and basal area of small trees (10−< 50 cm gbh) forming the dense understorey. Neither rate was related to topography. While species with larger mean gbh had greater relative growth rates (rgr) than smaller ones, subplot mean recruitment rates were correlated with rgr among small trees. Separating understorey species (typically the Euphorbiaccae) from the overstorey (Dipterocarpaceae) showed marked differences in change in mortality with increasing gbh: in the former it increased, in the latter it decreased. Forest processes are centred on this understorey quasi-stratum.

The two replicate plots showed a high correspondence in the mortality, recruitment, population changes and growth rates of small trees for the 49 most abundant species in common to both. Overstorey species had higher rgrs than understorey ones, but both showed considerable ranges in mortality and recruitment rates. The supposed trade-off in traits, viz slower rgr, shade tolerance and lower population turnover in the understorey group versus faster potential growth rate, high light responsiveness and high turnover in the overstorey group, was only partly met, as some understorey species were also very dynamic.

The forest at Danum, under such a disturbance-recovery regime, can be viewed as having a dynamic equilibrium in functional and structural terms. A second trade-off in shade-tolerance versus drought-tolerance is suggested for among the understorey species. A two-storey (or vertical component) model is proposed where the understorey-overstorey species' ratio of small stems (currently 2:1) is maintained by a major feedback process. The understorey appears to be an important part of this forest, giving resilience against drought and protecting the overstorey saplings in the long term. This view could be valuable for understanding forest responses to climate change where drought frequency in Borneo is predicted to intensify in the coming decades.

The effects of selective logging on the diversity and species composition of moths were investigated by sampling from multiple sites in primary forest, both understorey and canopy, and logged forest at Danum Valley, Sabah, Malaysia. The diversity of individual sites was similar, although rarefied species richness of logged forest was 17% lower than for primary forest (understorey and canopy combined). There was significant heterogeneity in faunal composition and measures of similarity (NESS index) among primary forest understorey sites which may be as great as those between primary understorey and logged forest. The lowest similarity values were between primary forest understorey and canopy, indicating a distinct canopy fauna. A number of species encountered in the logged forest were confined to, or more abundant in, the canopy of primary forest. Approximately 10% of species were confined to primary forest across a range of species' abundances, suggesting this is a minimum estimate for the number of species lost following logging. The importance of accounting for heterogeneity within primary forest and sampling in the canopy when measuring the effects of disturbance on tropical forest communities are emphasized.

A synthesis is presented of sampling work conducted under a UK government-funded Darwin Initiative grant undertaken predominantly within the Danum Valley Conservation Area (DVCA), Sabah, East Malaysia. The project concerned the assemblage structure, gas physiology and landscape gas fluxes of termites in pristine and two ages of secondary, dipterocarp forest. The DVCA termite fauna is typical of the Sunda region, dominated by Termes-group soil-feeders and Nasutitermitinae. Selective logging appears to have relatively little effect on termite assemblages, although soil-feeding termites may be moderately affected by this level of disturbance. Species composition changes, but to a small extent when considered against the background level of compositional differences within the Sunda region. Physiologically the assemblage is very like others that have been studied, although there are some species that do not fit on the expected body size–metabolic rate curve. As elsewhere, soil-feeders and soil–wood interface-feeders tend to produce more methane. As with the termite assemblage characteristics, gross gas and energy fluxes do not differ significantly between logged and unlogged sites. Although gross methane fluxes are high, all the soils at DVCA were methane sinks, suggesting that methane oxidation by methanotrophic bacteria was a more important process than methane production by gut archaea. This implies that methane production by termites in South-East Asia is not contributing significantly to the observed increase in levels of methane production worldwide. Biomass density, species richness, clade complement and energy flow were much lower at DVCA than at a directly comparable site in southern Cameroon. This is probably due to the different biogeographical histories of the areas.

Community resistance to, and resilience from, perturbation will determine the trajectory of recovery from disturbance. Although selective timber extraction is considered a severe disturbance, fish communities from headwater streams around Danum Valley Field Centre, Sabah, Malaysia, showed few long-term changes in species composition or abundance. However, some species showed short-term (< 18 months) absence or decrease in abundance. These observations suggested that both resistance and resilience were important in maintaining long-term fish community structure. Resistance to perturbation was tested by monitoring fish communities before and after the creation of log-debris dams, while resilience was investigated by following the time-course of recolonization following complete removal of all fish. High community resistance was generally shown although the response was site-specific, dependent on the composition of the starting community, the size of the stream and physical habitat changes. High resilience was demonstrated in all recolonization experiments with strong correlations between pre- and post-defaunation communities, although there was a significant difference between pool and riffle habitats in the time-course of recovery. These differences can be explained by the movement characteristics of the species found in the different habitats. Resilience appeared to be a more predictable characteristic of the community than resistance and the implications of this for ensuring the long-term persistence of fish in the area are discussed.

Understanding the extent and causes of insect diversity in the humid tropics is one of the major challenges in modern ecology. We review some of the current approaches to this problem, and discuss how future progress may be made. Recent calculations that there may be more than 30 million species of insect on earth have focused attention on the magnitude of this problem and stimulated several new lines of research (although the true figure is now widely thought to be between five and ten million species). We discuss work based on insecticidal fogging surveys; studies of herbivore and parasitoid specificity; macroecological approaches; and the construction of food webs. It is argued that progress in estimating insect diversity and in understanding insect community dynamics will be enhanced by building local inventories of species diversity, and in descriptive and experimental studies of the trophic structure of communities. As an illustration of work aimed at the last goal, we discuss the construction and analysis of quantitative host-parasitoid food webs, drawing on our work on leaf miner communities in Central America.

A nutrient addition experiment was set up in August 1993 in a species-rich primary lowland dipterocarp forest in Barito Ulu, Central Kalimantan, Indonesia. The following treatments were applied: control, +N, +P and +NP. There were five blocks of four 50m × 50m plots with a separate treatment for each plot. Fine litterfall was measured on all the plots from 1 May 1994 for 12 months. Litterfall mass and phosphorus concentrations were significantly higher in all the fertilizer treatments compared with the controls. All trees (≥ 10 cm dbh) were measured in August 1993 and in August 1998, and there was no significant girth increment response to fertilization in dipterocarps or non-dipterocarps. Dipterocarps of the red meranti group showed a doubling of girth increment in the +NP treatment, however, the difference from the control fell short of significance.

The ability to model the suspended sediment flux (SSflux) and associated water flow from terrain affected by selective logging is important to the establishment of credible measures to improve the ecological sustainability of forestry practices. Recent appreciation of the impact of parameter uncertainty on the statistical credibility of complex models with little internal state validation supports the use of more parsimonious approaches such as data-based mechanistic (DBM) modelling. The DBM approach combines physically based understanding with model structure identification based on transfer functions and objective statistical inference. Within this study, these approaches have been newly applied to rainfall–SSflux response. The dynamics of the sediment system, together with the rainfall–river flow system, were monitored at five nested contributory areas within a 44 ha headwater region in Malaysian Borneo. The data series analysed covered a whole year at a 5 min resolution, and were collected during a period some five to six years after selective timber harvesting had ceased. Physically based and statistical interpretation of these data was possible given the wealth of contemporary and past hydrogeomorphic data collected within the same region.

The results indicated that parsimonious, three-parameter models of rainfall–river flow and rainfall-SSflux for the whole catchment describe 80 and 90% of the variance, respectively, and that parameter changes between scales could be explained in physically meaningful terms. Indeed, the modelling indicated some new conceptual descriptions of the river flow and sediment-generation systems. An extreme rainstorm having a 10–20 year return period was present within the data series and was shown to generate new mass movements along the forestry roads that had a differential impact on the monitored contributory areas. Critically, this spatially discrete behaviour was captured by the modelling and may indicate the potential use of DBM approaches for (i) predicting the differential effect of alternative forestry practices, (ii) estimating uncertainty in the behaviour of ungauged areas and (iii) forecasting river flow and SSflux in terrain with temporal changes in rainfall regime and forestry impacts.

A series of growth experiments and observations on natural populations have been carried out on dipterocarp species of contrasting ecology growing in artificial gaps and the forest understorey. These studies have demonstrated that although differences exist between species in photosynthetic and growth responses to the high-light environment, competition for light in canopy gaps is highly asymmetrical and tends to reinforce any pre-existing dominance hierarchy. We propose that differences in seedling persistence in forest canopy shade are highly influenced by species-specific biotic and abiotic interactions. Our experiments suggest that as seedlings, dipterocarp species trade off traits which enhance persistence and growth in shade against those that enhance their ability to exploit gaps. Less competitive species survive for progressively longer periods of time after a gregarious fruiting event. This leads to significant shifts with time in the number of species present in the seedling bank and hence in the importance of interspecific competition in determining which species dominates regrowth in gaps. We propose that this special case of dispersal limitation is more likely to account for coexistence of dipterocarp species than differences in growth responses to gaps of different size, with stochastic and environmental variables interacting to determine species distribution and abundance.

The Krakatau Islands, Indonesia, have provided an opportunity for ecologists to track primary succession from the “clean slate” of 1883, through forest closure in the 1920s, to the contemporary period, in which successional changes take the form of alterations in composition and stature of forest stands rather than gross changes in ecosystem type. This paper reports on permanent forest plots established on the islands in 1989, and fully surveyed again in both 1992 and 1997. Since 1989, the plots have been subject to natural disturbance phenomena in the form of varying combinations of, for example, deposition of volcanic ejecta, landslides, lightning strikes, storm damage and drought. These effects have been concentrated between 1992 and 1997, during which the volcano Anak Krakatau has deposited ash on the islands of Sertung and Panjang, but not on Rakata. Data on stand responses are presented for growth rates (dbh (diameter at breast height, 1.3 m) increment), stem recruitment and mortality, biomass changes (partitioned into mortality, ingrowth and growth of established trees), and compositional shifts. The discussion focuses on evaluation of questions and successional models framed earlier in the programme. One general finding is that the stand dominants as of 1989 have tended to decline in number within the plots, generally through low levels of recruitment failing to balance rates of mortality. The effects of disturbance to the plots appear to be evident in terms of mortality and recruitment, dbh increment, and changes in biomass. The patterns of change in the eight plots are quite varied, such that relatively few generalizations are possible. The difficulties of establishing meaningful baseline rates for tree growth and stand biomass are discussed.

Climatic records for Danum for 1985–1998, elsewhere in Sabah since 1879, and long monthly rainfall series from other rainforest locations are used to place the climate, and particularly the dry period climatology, of Danum into a world rainforest context. The magnitude frequency and seasonality of dry periods are shown to vary greatly within the world's rainforest zone. The climate of Danum, which is aseasonal but subject, as in 1997–1998, to occasional drought, is intermediate between less drought-prone north-western Borneo and the more drought-prone east coast. Changes through time in drought magnitude frequency in Sabah and rainforest locations elsewhere in South-East Asia and in the Neotropics are compared. The 1997–1998 ENSO-related drought event in Sabah is placed into a historical context. The effects of drought on tree growth and mortality in the tropics are assessed and a model relating intensity and frequency of drought disturbance to forest structure and composition is discussed.

The visual uniformity of tropical peat swamp forest masks the considerable variation in forest structure that has evolved in response to differences and changes in peat characteristics over many millennia. Details are presented of forest structure and tree composition of the principal peat swamp forest types in the upper catchment of Sungai Sebangau, Central Kalimantan, Indonesia, in relation to thickness and hydrology of the peat. Consideration is given to data on peat geochemistry and age of peat that provide evidence of the ombrotrophic nature of this vast peatland and its mode of formation. The future sustainability of this ecosystem is predicted from information available on climate change and human impact in this region.

The following sections are included:

• Index