Part B
Sumatra, area 476,000 km2, probably has over 10,000 species of higher (seed) plants, most of which are found in lowland forest. For comparison, the British Isles is 65% of the area but has only about 13% of the number of species. The number of tree species per unit area in the lowland forest (p. 189) of Malesia and particularly Western Malesia (which includes Sumatra) is probably greater than in similar forest in west Africa or South America. Indeed, these Sunda Region forests are probably the most species-rich plant communities in the world (Whitmore 1975).
Part of the reason for the very large number of plant species is that in the near-perfect growing conditions of warmth and adequate moisture, the tall trees provide a large framework for a wide and environmentally diverse range of structural niches. These are filled by smaller trees, shrubs, herbs, scramblers, climbers, epiphytes, parasites, etc. The underlying causes of high plant diversity are discussed on pages 189-192.
The vast number of plant species present in Sumatra and elsewhere in the Sunda Region represent a magnificently rich source of natural resources, many of which have proven or potential economic value (Williams 1975). About one-third of the 7,500 species of plants known from Peninsular Malaysia are listed by Burkill (1966) as having some economic value, and a similar proportion would be expected from Sumatra's plants. The dipterocarp timber trees are the most obvious example and these constitute about one-quarter of all the hardwood timber on international markets. Unfortunately, this reflects the productivity of timber companies rather than the productivity of the trees themselves, and accessible supplies of dipterocarp timber will be exhausted by the end of this century or before (Ashton 1980). In addition to the timber species already exploited, there are many tree species whose potential for timber, fibre or cellulose remains unexploited. In addition to these are hundreds of species in the forests whose economic potential for food or chemicals remains unassessed (Whitmore 1980). People of rural areas generally have a deeper understanding of the value of wild or semi-domesticated forest plants but, as Sastrapradja and Kartawinata (1975) have pointed out, knowledge of these can be lost within the timespan of a generation. Useful summaries of species with economic potential in Southeast Asian forest and of the need for genetic resource conservation are given by Sastrapradja et al.(1980), Whitmore (1980) and Jacobs (1980).
The economic value of forest vertebrates is less well studied but the vital role that macaque species played in the early stages of the mass production of polio vaccine should not be forgotten (Medway 1980). That role was largely unanticipated and it would be a brave person who would claim now that some other wild vertebrate could not be of similar value to human welfare. In addition, cave bats are vital to the production of durians (p. 329). The potential uses and values of invertebrates - which are more numerous, more diverse, less well known - are clearly even greater. The economic value of the goods and services provided by coastal swamps and natural ecosystems, particularly lowland forest, is described elsewhere (Burbridge 1982; Farnworth et al. 1981; Krutilla and Fisher 1975).
Considering the aesthetic value, economic importance, and diversity of Sumatra's forest and other natural ecosystems, it is surprising that studies of their components and the way these components function and interact has not been particularly intense or widespread. From discussions with staff at the University of North Sumatra there seem to be three major reasons for this situation - tigers, snakes and leeches - all of which have reputations unrelated to the facts.
a) Tigers. The tigers reported in the newspapers as killing stock animals or even humans are rogue individuals and are not typical of the species. Adventurous young or ailing old animals may take to seeking food around villages, at least in part because their forest habitat is being reduced in area. Field workers who have spent over two years conducting research in the forest of the Mount Leuser National Park, where tigers are regarded as common, have counted themselves 'lucky' to have had a fleeting glimpse of a fleeing tiger. It is worth reporting that during the CRES ascent of Mt. Kemiri, one or two tigers visited the first camp site (at 1,500 m) on two consecutive nights - passing within 30 cm and 50 cm of the sleeping party - and the only loss was a pair of sandals, later found thoroughly chewed at the side of the path.
b) Snakes. It surprises many people to learn that less than 18% of the world's 2,300-2,700 snake species are poisonous, and only a fraction of these actually endanger human life (Duellman 1979; C. McCarthy, pers. comm.). Sumatra has about 150 snake species (de Hass 1950). Of this total 6% are poisonous sea snakes, 67% are harmless land snakes, and 14% are poisonous land snakes. Of this last group, however, most are too small or weak to present any hazard to humans. However, the reticulated python, Python reticulatus, which constricts its prey, must be regarded as dangerous to man. In fact, the percentage and density of snakes that pose any sort of threat to people is probably considerably higher in towns and villages than in forest. The information above should not lead to carelessness with snakes in forests; however, they are only rarely seen.
c) Leeches. Most leeches, whose closest relatives are earthworms, are found in fresh water but some species have adapted to life on land or in the sea. Some of the larger freshwater (swamp) species are known to transmit protozoan blood parasites to their hosts (usually mussels, fish and frogs), or cause serious blood loss to cattle or buffalo which drink from infested water. The leeches seen in lowland forest are small (2-4 cm) and are not known to transmit any disease to humans or other animals (Sharma and Fernando 1961). Of the two commonest species, the dull-coloured Haemadipsa zeylandica gives a painless bite, but the brighter-coloured H. picta has a fiery bite. Bleeding from a land leech bite looks spectacular but the amount of blood lost is very small. Proper dress and perhaps a repellent (based on dimethyphthalate) provides protection, but if a leech manages to bite, a dab of antiseptic to prevent secondary infection is all that is required.
The concern for and interest in natural ecosystems expressed by some people (often foreigners) baffles others. "Why bother?" they might ask. Van Steenis (1971) has replied to such a rhetorical question by answering, "Come with an observant eye and the question will answer itself." In the next nine chapters it is hoped that the reader will be able to 'observe' a little from his chair and then be sufficiently enthused to venture out and find out more himself about the forest and natural ecosystems of Sumatra.
Chapter Two
INTRODUCTION
Mangrove forests1 are sometimes regarded as a very distinct ecosystem virtually requiring a separate discipline of science. In fact, they are just one of the many types of forest in Sumatra having a closed, even canopy made up of tree species which are predominantly evergreen. The environmental conditions in which they grow are extreme, however, because they are subject to soil-water salinity and waterlogging. Mangrove forests have important environmental roles concerned with land, wildlife and fisheries management (table 2.1) and have been exploited by man for numerous natural products (table 2.2). This exploitation by man has had important effects on the ecosystem.
On Sumatra, mangrove forests have been studied more than almost any other natural ecosystem. Every Sumatra province has mangrove fringing some of its shores, if only in a few sheltered bays or river entrances or around offshore islands, but of the 1,470,000 ha of mangrove forests in 1982, over 60% were in Riau and South Sumatra (p. 20) and this is where the studies have concentrated. Major studies have been conducted by Sabar et al. (1979), Soeriaatmadja (1979), Sukardjo (1979), and Sukardjo and Kartawinata (1979).
Mangrove forests form a protective and productive margin to much of Sumatra's coastline. It is important, from many points of view, that the ecology of mangrove areas should be understood as fully as possible and as soon as possible, but particularly because so many development projects are in the coastal region and create serious conflicts of land use.
An excellent review of mangrove ecosystems in Indonesia is given by Kartawinata et al. (1979). They conclude:
Almost all species of plant which make up Indonesian mangrove swamps are now known. However, many fields such as variations in species composition, forest structure, seed dispersal, phenology, biology, flowering and fruiting, composition of the fauna, mineral cycling, productivity and ecosystem dynamics are barely understood. Because of this, studies of the above topics need to be encouraged. Exploration, inventory and mapping must be given priority too.
As with other chapters, results of studies from Peninsular Malaysia and elsewhere are used below in order to increase the depth of the information.
Although there is no generally accepted definition of a mangrove ecosystem, Saenger et al. (1981) proposed that it should consist of the following:
1) one or more of the exclusive mangrove tree species (see p. 79 and table 2.4);
2) any non-exclusive plant species growing with (1);
3) the associated biota - the terrestrial and marine animals, the lichens, fungi, algae, bacteria, etc., whether temporary, permanent, casual, incidental or exclusive in the area of (1);
4) the processes essential for its maintenance whether or not within the area of (1).
To this should be added:
5) and the foreshore below the margin of vegetation where colonization by pioneer mangrove trees will occur and where the fauna has a terrestrial component.
An extremely comprehensive bibliography of references to mangroves totalling 5,608 entries has been published by UNESCO (Rollet 1981).
From Saenger et al. 1981
From Saenger et al. 1981
IMPORTANT PHYSICAL FEATURES
Tides
The most important feature of the physical environment for the mangrove biota is the tide. Tides usually occur twice each day and are caused by the gravitational pull of the moon and to a lesser extent the sun. In the open ocean the amplitude (height) of the tides is not more than about 0.5 m but in shallow seas around Sumatra it is commonly up to 3 m. The tides of greatest amplitude, the spring tides, occur when the earth, sun and moon are in a straight line (i.e., about the times of the new moon and the full moon). Neap tides, those of smallest amplitude, occur when the sun is opposed (is at right angles) to the gravitational pull of the moon (i.e., about the times of the first and last quarters of the moon).
As the moon orbits the earth in the same direction as the earth's rotation, a period of rather more than one day (in fact, 24 hours 50 minutes) elapses between successive occasions when the moon is vertically above the same meridian. Thus, the interval between successive high or low tides is generally half this period (i.e., about 12 ½ hours). This general pattern is modified when the tidal waters move through complex archipelagos or irregularities in the coastline, and the fauna and flora of different locations will respond to these different regimes of inundation.
The tidal regimes for the waters around Sumatra are shown in figure 2.1, and it can be seen that Sumatra experiences four types of tide which are described below and illustrated in figure 2.2.
• Semi-diurnal tide - two high waters and two low waters daily with similar amplitudes;
• Mixed tide, mainly semi-diurnal - two high waters and two low waters daily with different amplitudes;
• Mixed tide, mainly diurnal — sometimes only one high water and one low water daily, but sometimes also two high waters which differ greatly in height;
• Diurnal tide - only one high water and one low water daily.
Some of the mangrove fauna do not avoid the alternate wetting and exposure caused by tides. Others climb into the trees during high tides and then descend to feed during low tides. One such animal is the snail Cerithidea decollata whose pattern of activity follows the tidal rhythm and is not associated with light or dark (Cockcroft and Forbes 1981). The snails descend during low tide periods and ascend before the next high tide. Note that the rising tide does not force them up, but that the tidal rhythm is part of their own internal rhythm.
Figure 2.1. Distribution of tidal tvpes around Sumatra. Leaend as in figure 2.2.
After Wyrtki 1961
Figure 2.2. Fluctuations in tide level through a cycle of 16 days for four different tidal regimes. A - semi-diurnal tide; B - mixed tide, mainly semi-diurnal; C - mixed tide, mainly diurnal; D - diurnal tide. Refer to text for further explanation.
After Wyrtki 1961
Salinity
As might be expected, the salinity of seawater around the coast decreases during the rainy season because of the increased volumes of fresh water flowing out of the rivers, and is greatest during the dry seasons. The average annual variation in surface salinity (in parts per thousand or % NaCl) for the waters around Sumatra is shown in figure 2.3. Patterns of water movement in estuaries are extremely complex. Where fresh water meets seawater, mixing does not automatically occur but rather the heavy seawater sinks below the fresh water to form a 'saltwater wedge'. This configuration varies with tides and also with river flow.
For most of the mangrove forest trees and burrowing fauna, the salinity of tidal water is probably less important than the salinity of the soil water. The salinity of this interflow (p. 133) is generally less than the water above it because of dilution by fresh rain water seeping through the soil. For tree roots and burrowing fauna the crucial factor is not so much the concentration of NaCl alone but the osmotic potential. This depends partly on soil type, being greater in clayey soils than in sandy soils. If it is not feasible to measure osmotic potential with the equipment available, then salinity and conductivity measurements represent a good second best.
Temperature
In tropical coastal waters the surface temperature is generally between 27° and 29°C. In shallow water, however, temperatures can reach at least 30°C, and the mud flats may become much hotter than this (Stebbins and Kalk 1961). Inside the actual mangrove forest the temperature is lower (dela Cruz and Banaag 1967) and the variation is more or less the same as for other shaded coastal areas. The forest areas therefore represent a rather more equitable environment for the mangrove biota than the mud flats.
Surface Currents
In general terms, seawater currents change with the rainy seasons but figure 2.4 shows that the Straits of Malacca experience a southeast current throughout the year and the west coast of Sumatra has a north/northwest current. However, the coast of southeast Sumatra has a north/northwest current in February and a south/southeast current in August. The western coast of Aceh and some of North Sumatra experience the opposite change in current. It can be seen that most of the water from rivers on the east coast will flow to the northwest, and water from rivers on the west coast to the southeast.
Figure 2.3. Average annual variation of surface salinity (%o)
After Wyrtki 1961
Figure 2.4. Surface currents around Sumatra in February and August.
After Wyrtki 1961
Nutrients and Dissolved Oxygen
The greatest concentration of dissolved oxygen in the coastal environment is at the open water edge where wave action repeatedly agitates the water. Oxygen concentration also fluctuates with tides, the highest values being found at high tide. The abundance of life in mangrove forests and the constantly replenished supply of nutrients, leads to a very high biological demand for oxygen, and this tends to lower the levels of available oxygen. Nevertheless, neither oxygen concentrations in free water nor nutrient levels impose any limitations on the productivity of the mangrove biota.
Summary of Water Quality in a Mangrove Forest
Variations in water quality at different locations from open sea to land through mangrove forest are shown in table 2.3. In general there is a gradient of decreasing nutrient concentrations moving from the mangrove area to the open sea. This is caused by dilution in the greater volume of water and increased incorporation of the nutrients into the sediments.
After Gomez 1980
ENERGY FLOW AND THE IMPORTANCE OF MANGROVE VEGETATION TO FISHERIES
Mangrove forests are highly productive ecosystems but only about 7% of their leaves are eaten by herbivores (Johnstone 1981). Most of the mangrove forest production enters the energy system as detritus or dead organic matter (fig. 2.5). This detritus plays an extremely important role in the productivity of the mangrove ecosystems as a whole (Lugo and Snedaker 1974; Ong et al. 1980a,b). Its importance to offshore ecosystems is not clear (Nixon et al. 1980). Leaves and other small litter (twigs, flowers, and fruit) fall throughout the year and are broken down by macroorganisms (mainly crabs) and microorganisms (mainly fungi) into smaller particles that form the detritus. The detritus becomes rich in nitrogen and phosphorus because of the fungi, bacteria and algae growing on and within it and is therefore an important food source for many 'detri-tivore' animals such as zooplankton, other small invertebrates, prawns, crabs, and fish. These detritivores are eaten in turn by carnivores which are dependent to varying degrees on these organisms. Those carnivores not dependent on detritivores would be directly or indirectly dependent on planktonic benthic algae (p. 134). It is known that in the turbid coastal waters characteristic of at least the northern half of the Malacca Straits, the productivity of phytoplankton is low. It is probable, therefore, that most of the micro- and macrofauna in the mangroves and surrounding coastal areas are dependent on the productivity of litter from mangrove forests (Ong et al. 1980a,b).
The high productivity of mangroves and the physical structure and shading they provide form a valuable habitat for many organisms, some of which are of commercial importance. The most valuable mangrove-related species in Indonesia are the penaeid prawns, which support an export market worth over $150 million annually (Anon. 1979a). The juvenile stages of several of these prawn species live in mangrove and adjacent vegetation, while the adults breed offshore (Soegiarto and Polunin 1980).
The influence of mangroves extends far beyond the prawn fisheries themselves. In Sumatra, for example, the profits from the export of prawns subsidises the sale of fish which are caught in trawlers' nets (Turner 1977). The average coastal fishpond (tambak) produces 287 kg fish/ha/yr, which is more than the offshore shrimp yield, but the loss of one hectare of mangrove to tambak actually leads to an approximate net loss of 480 kg offshore fish and shrimp per year. Thus, widespread loss of mangrove to tambak is likely to cause the loss of jobs (Turner 1977). An assessment of the impact of a mangrove reclamation project on the south Java coast conservatively estimated that the development would lead eventually to loss of employment for 2,400 fishermen in the area and the loss of $5.6 million in annual income (Turner 1975).
Figure 2.5. Major pathways of energy flow in a mangrove-fringed estuary.
After Saenger et al. 1981
Apart from the prawns, many other species of economic importance are associated with mangroves. These include the edible crab Scylla serrata, the shrimp Acetes, from which 'belacan' paste is made, and fishes such as Chanos, Mugil and Lates (MacNae 1968; Soegiarto and Polunin 1980).
THE VEGETATION
Tree Species
Mangrove forests are found in almost all tropical areas and it is likely that the centre of evolution was the Indo-Malesian region. This region (India, Burma, Thailand, Malaysia, Indonesia and northern Australia) has the most species-rich mangrove forests (Chapman 1977a,b) but even so contains many fewer species than other natural forest types. There are only 17 tree species (belonging to seven genera and only four families) which would be generally encountered in Sumatran mangroves (table 2.4), but over 20 different plant communities, many dominated by a single tree species, have been identified in Southeast Asian mangroves (Chapman 1977b). Details of the species found at different locations along the east coast of Sumatra and west coast of Peninsular Malaysia can be found elsewhere (van Bodegom 1929; Bunning 1944; DKPD I Riau 1979; Endert 1920; Jonker 1933; Lugo and Snedaker 1974; Luythes 1923; Ong et al. 1980a; Samingan 1980; Soeriaatmadja 1979; Steup 1941; Sukardjo 1979; Sukardjo and Kartawinata 1979; Tee 1982a; Versteegh 1951). Further Dutch references can be found in Kartawinata et al. (1979). A practical key to mangrove and estuarine trees is available (Wyatt-Smith 1979).
Mangrove trees are 'halophytes'. That is, they are able to withstand the saline soil in which they are rooted, and repeated inundation by seawater. This does not imply that they can grow only under such conditions; three species thrive next to a freshwater pond at the Botanic Gardens in Bogor (Ding Hou 1958). The occurrence of mangrove forest only along coasts indicates that these species exist there because of a lack of successful competition from other plant species which have not adapted to the salty environment.
Clearly the mangrove species are in turn unable to compete successfully against freshwater vegetation in other locations. This illustrates well the difference between fundamental niche and realised niche which is discussed on page 159. Some mangrove trees can be found growing on riverbanks over 100 km from the sea. The water they grow in may appear to be fresh, but a wedge of salt water (heavier than fresh water) may extend a great distance inland. Thus the roots of a plant may be in saline water although it grows through fresh water. A review of the response of mangrove plants to salt water is given by Walsh (1974).
A number of mangrove forest trees have peculiar rooting systems (fig. 2.6). Rhizophora spp. have stilt roots which help to support the trees and possibly prevent other seedlings from becoming established too close. The spike pneumatophores (roots that stick out of the ground) of Sonneratia and Avicennia, and the bent pneumatophores of Brugaiera, allow oxygen to enter the root system. The mangrove soil is generally waterlogged and unless there is extensive burrowing by crabs and other animals, roots a few centimetres below the soil surface are in essentially anoxic (no oxygen) conditions.
The roots of Sonneratia and Avicennia are composed of four distinct parts. The cable root runs beneath the soil surface and is held in place by anchor roots which grow downwards. The pneumatophores grow up from the cable root and small nutritive roots grow horizontally from these (fig. 2.6). In addition to the function of pneumatophores in gas exchange, it is thought that they also allow nutritive roots to grow quickly into new sediments should the soil surface suddenly rise, causing the nutritive roots to be buried within the anoxic layer (MacNae 1968; Troll and Dragendorf 1931).
The pollination ecology of mangrove trees is extremely diverse with wind, bats, birds, butterflies, moths and other insects all being pollinators. The pollination ecology of Rhizophoraceae (Rhizophora, Bruguiera, etc.) is described by Tomlinson et al. (1979).
A few mangrove tree species have evolved an unusual, though not unique, form of reproduction. Generally speaking, a fruit develops on a plant and when it is ripe or fully developed, the fruit or the seed(s) inside it is then dispersed and the seed germinates when or if it comes to rest in suitable conditions. In most of the Rhizophoraceae such as Rhizophora and Bruguiera, however, the fruits ripen and then, before leaving the parent tree, the seeds germinate inside the fruit, possibly absorbing food from the tree. The embryonic root or hypocotyl of the seedling pierces the wall of the fruit and then grows downwards. The cotyledons (first leaves) remain inside the fruit. Eventually, in Rhizophora mucronata, for example, the root may reach a length of 45 cm. When the seedling is fully grown it drops off by separating itself from the cotyledon tube, the scar of which forms a ring around the top of the fallen seedling, and the small leaf-bud can be seen above this scar (fig. 2.7). Bruguiera behaves similarly, but the break occurs at the stalk of the fruit. These types of fruit are described in more detail by MacNae (1968).
Figure 2.6. Different types of roots in mangrove trees: a- knee roots as found in Bruguiera spp.; b- spike roots as found in Sonneratia spp., Avicennia spp. (and sometimes Xylocarpus moluccensis); and c- still roots as found in Rhizophora spp.
Zonation
Watson recognised five major divisions of mangrove forest in Peninsular Malaysia based on tidal regimes, such that the first division (nearest the sea) was inundated by all tides, and the fifth division was inundated only by abnormal tides (Watson 1928). These tidal zones supported different types of vegetation, the species composition of which varied with the position relative to rivers and the sea (fig. 2.8). The major divisions were:
Figure 2.7. The propagule of Rhizophora mucronata showing the root (often mistaken for part of the fruit) and the top of the part that detaches itself from the parent plant.
1. Forest nearest the sea dominated by Avicennia and Sonneratia, the latter growing on deep mud rich in organic matter. A. marina grows on comparatively firm clayey substrate which is easy to walk on, whereas A. alba grows on softer mud (Ding Hou 1958; Sukardjo and Kartawinata 1979; Troll and Dragendorf 1931).
2. Forest on slightly higher ground is often dominated by Bruguiera cylindrica and can form virtually pure stands behind Avicennia forest. It grows on firm stiff clays out of the reach of most tides.
3. Forest further inland dominated by Rhizophora mucronata and R. apiculata, the former preferring slightiy wetter conditions and deeper mud. These trees can be 35-40 m tall (Ding Hou 1958). Also present are Bruguiera parviflora and Xylocarpus granatum. Mounds of mud built up by the mud lobster are often colonised by the large fern Acrostichum aureum.
4. Forest dominated by Bruguiera parviflora can occur in pure stands although it often invades Rhizophora forest after it has been clear-felled.
5. The final mangrove forest is that dominated by Bruguiera gymnorrhiza. The seedlings and saplings of this tree are tolerant of shade in conditions where Rhizophora cannot perpetuate itself. Like the she-oak Casuarina (p. 117), however, seedlings of B. gymnorrhiza will not develop under the canopy of their parents. The transition between this type of forest and inland forest is marked by the occurrence of Lumnitzera racemosa, Xylocarpus moluccensis, Intsia bijuga, Ficus retusa, rattans, pandans and, at the inland margin, the tall palm Oncosperma tigillaria. This succession is not always visible, particularly where man has disturbed the forest; in such conditions the fern Acrostichum aureum is very common (p. 347).
Figure 2.8. Typical but not invariable distribution of mangrove tree species near the mouth of a large river. Aa- Avicennia alba, Am-A. marina, Be- Bruguiera cylindrica, Bg- B. gymnorrhiza, Bp- 6. parviflora, Bs- B. sexangula, Ct-Ceriops tagal, Fr- Ficus retusa, Ir-Intsia bijuga, Ot- Oncosperma tigillaria, Ra- Rhizophora apiculata, Rm- R. mucronata, Sa- Sonneratia alba, Sg- S. griffithii, Xg-Xylocarpus granatum, Xm- X. moluccensis. Note that some of the names used by Watson have been modernized for this diagram. See table 2.4.
After Watson 1928
The above descriptions refer mainly to gently sloping, accreting shores, but where there are creeks, bays or lagoons, Rhizophora is usually the pioneer tree. This is common in areas near coral reefs (p. 124).
The succession generally followed by mangrove vegetation under different conditions is shown in figure 2.9.
The tendency of the mangrove forest to occur in distinct floral zones has been interpreted by various authors in many different ways. These have been summarised by Snedaker in the following four categories: plant succession, geomorphology, physiological ecology, and population dynamics (Snedaker 1982b).
Figure 2.9. Succession in mangrove forests.
After Chapman 1970 in Walsh 1974
Plant Succession
Plant succession is a classic ecological concept and is defined as being the progressive replacement of one plant community with another (e.g., bare ground-alang-alang-scrub-secondary forest-primary forest [see p. 340]). Much of the early work on mangrove forests focused on its supposed land-building role. It seemed clear from this that one species colonised an exposed bank of mud, and as conditions changed (such as the increase in the organic debris of the mud) so other species took over. Thus at any one spot, a succession of species or communities of species would be observed over a period of time. It is clear however that in most places, including Sumatra, the stages of succession are not always consistent and different local environmental conditions and man's impact on those have an influence.
Budowski (1963) compiled a list of 20 characteristics of secondary succession in tropical forests which distinguish between early and late stages. If this is compared with the stages of succession for mangroves, only seven of the characters apply, nine do not, and four are inconclusive. This suggests that attributing the apparent zonation to succession is not the whole story (Snedaker 1982b).
Geomorphological Change
Early workers on mangrove felt that it was mangroves that 'built land'. However, it is clear from observations that in some areas such as the huge deltas of the Ganges, Indus and Irrawaddy, it is the process of sediment decomposition that builds land. It appears to most workers now that mangroves do not have any influence on the initial development of the land forms. Mangroves may accelerate land extension but they do not seem to cause it (p. 88) Van Steenis tells of engineers working at Belawan (the port north of Medan) who planted mangrove seedlings on a new port extension to stabilise it. The attempt failed, thereby demonstrating that it is not possible to force accumulation of silt (Ding Hou 1958).
From a geomorphological perspective, it is the shape, topography and history of the coastal zone that determine the types and distributions of mangrove trees in the resulting habitats. The position of species relative to tidal levels (and thus soil type) is obviously important, and Watson (1928) considered the pattern of tidal inundation and drainage to be the major factor in mangrove zonation. This idea was developed by Lugo and Snedaker (1974) to include variation in the salinity of the tidal water and the direction of its flow into and out of the forest. Although good correlations exist between salinity and zonation, they are not proofs of direct cause and effect.
Physiological Response to Soil-Water Salinity
If salinity is an actual cause of zonation in mangrove forests, it needs to be shown that the plants actually respond through their physiology to salinity gradients and not to factors such as oxygen levels, which under certain conditions will fall with decreasing salinity (Ding Hou 1958). It has already been mentioned that mangroves are able to live in fresh water (i.e., they are facultative not obligate halophytes) but each species probably has a definable optimum range of salinity for its growth. Indeed, Lugo et al. (1975) found that within each zone the characteristic species had apparently maximised its physiological efficiency and therefore had a higher metabolic rate than any invading species. These invading species would be at a competitive disadvantage due to their lower metabolic efficiency in that habitat. Similarly, it has been found that each species of the mangrove forest grows best under slightly different conditions, such as the amount of water in the mud, the salinity, and the ability of the plant to tolerate shade (Lind and Morrison 1974). This means that the various species are not mingled together in a haphazard way but occur in a fairly distinct zonation.
As was shown on pages 74 and 76, salinity can vary considerably between high and low tides and between seasons, and thereby presents a confusing picture to a scientist conducting a short-term study. Thus, to identify the salinity levels to which the different species are optimally adapted requires long-term and detailed measurements to determine long-term averages and ranges of salinity.
Differential Dispersal of Propagules
The suggestion that differential dispersal of propagules (fruit, etc.) influences zonation of mangroves rests on the idea that the principal propagule characteristics (e.g., size, weight, shape, buoyancy, viability, numbers, and location of source areas) result in differential tidal sorting and therefore deposition. There are as yet few data to support this hypothesis (Rabi-nowitz 1978) and interested readers should refer to Snedaker (1982b) for a thorough discussion.
Geomorphology and environmental physiology thus appear to be the most relevant of the topics discussed above to further the understanding of zonation and plant succession in the mangrove forest. However the impact of human activities, so ubiquitous in coastal regions, plays a major role in modifying species composition, and physical conditions, and so should also be considered in any study of zonation.
The term 'zone' itself needs to be clarified. Bunt and Williams (1981) suggest that recognisable zones may arise through at least two causes: situations where neighbouring vegetation associations have little or no floristic affinity although growing in the same environmental conditions (continuous variation); and situations where environmental gradients exist at a scale to permit sudden changes between related vegetation associations (discontinuous variation). Thus vegetational changes can be continuous, discontinuous or a combination of both. This, they argue, is why it is crucial to consider scale before attempting an analysis of mangrove forest and why, in the absence of such consideration, data from isolated transects, even from the same area, are so hard to interpret. Comparisons of transect data from different sites are useful in compiling inventories and noting similarities but are not the basis for a discussion of zonation.
Biomass and Productivity
Biomass is a term for the weight of living material, usually expressed as a dry weight, in all or part of an organism, population or community (Ricklefs 1979). It is commonly expressed as the biomass density or biomass per unit area. Plant biomass2 is the total dry weight of all living plant parts and for convenience is sometimes divided into above-ground plant biomass (leaves, branches, boughs, trunk) and below-ground plant biomass (roots). It appears that no study of mangrove biomass has been conducted in Sumatra, but several studies have been conducted on the other side of the Straits of Malacca in Peninsular Malaysia. Thus the biomass of trees in the Sungai Merbok Forest Reserve was between 122 and 245 t/ha (Ong et al. 1980a). In the Matang Forest Reserve, however, which has been exploited for timber on a sustained basis for 80 years, and which receives silvicultural management, the biomass of trees was 300 t/ha (Ong et al. 1980b). As explained below, the higher biomass in managed forest is not unexpected.
Biomass is a useful measurement but it gives no indication of the dynamics of an ecosystem. Ecologists are interested in productivity because if the dry weight of a community can be determined at a given moment and the rate change in dry weight can be measured, these data can be converted into the rate energy flow through an ecosystem. Using this information, different ecosystems can be compared, and their relative efficiencies of converting solar radiation into organic matter can be calculated (Brock 1981).
Plant biomass increases because plants secure carbon dioxide from the atmosphere and convert this into organic matter through the process of photosynthesis. Thus, unlike animals, plants make their own food. The rate at which biomass increases is the 'gross primary productivity'. This depends on the leaf area exposed, amount of solar radiation, temperature, and upon the characteristics of individual plant species (Whitmore 1984). During the day and during the night, plants, like all other living organisms, respire and use up a proportion of the production formed through photosynthesis. What remains is called 'net primary productivity' and over a period of time this is termed 'net primary production'. Net primary productivity is obviously greatest in a young forest which is growing, and it should be remembered that a dense, tall forest with a high biomass does not necessarily have a high net primary productivity. Large trees may have virtually stopped growing. Indeed in an old 'overmature' forest, the death of parts of the trees and attacks by animals and fungi may even act to reduce the plant biomass while net primary productivity remains more or less constant. The major aim of silvicultural management in forests or timber plantations is to maximise productivity, and the trees are usually harvested while they are still growing fast and before the net primary productivity begins to decrease.
One means of assessing net primary production is to measure the rate at which litter is produced. The production of litter at Merbok Forest Reserve was found to be 8.1 t/ha/yr for leaf litter and 12.0 t/ha/yr for total small litter (mainly leaves but with twigs, flowers and fruit) (Ong et al. 1980a). At Matang, Ong et al. obtained figures for production of 8.2 and 10.4 t/ha/yr (Ong et al. 1980b). These figures are similar to those obtained in lowland forest (p. 205).
The approximate annual accumulation of litter on the mangrove forest floor at Merbok was calculated to be 0.33 t/ha for leaf litter and 1.13 t/ha for total litter. An experiment at the same site revealed that 40%-90% of fallen leaves were lost after 20 days on the forest floor. The major agents in the disappearance were probably crabs which either bury them (eventually to form peat) or eat them, later to be excreted as detritus. At Matang, the time required for total leaf decomposition in areas with virtually no crabs (i.e., with only microorganisms involved in decomposing) was 4-6 months (Ong et al. 1980a).
In calculations of total productivity the contribution made by soil algae, algae growing on roots, and phytoplankton should not be overlooked, as their combined contribution to the total can be quite significant (Walsh 1974).
Tannins
It used to be thought that the leaves of mangrove trees were exceptionally rich in tannins, a view originating perhaps from the commercial use of mangrove bark for producing tannins for the leather industry (Walsh 1974, 1977). Recent analyses, however, suggest that tannin content of mangrove tree leaves exhibits a wide spectrum of concentrations, with Bruguiera having the highest levels. It may well be that mangroves have a higher-than-average concentration of tannins, but not significantly more than in other forests growing under somewhat adverse conditions (P.G. Waterman, pers. comm.) (pp. 175 and 256). Bark of trunks usually contains more tannins than bark of twigs, which has in turn more tannins than the leaves. Mangrove barks often contain 20%-30% dry weight of tannins (Walsh 1977), and although high, this it is not outside the range found in lowland forest barks (Whitten, unpubl.).
Tannins may play an extremely important role in the interactions between herbivores and leaves (pp. 110, 177, 230 and 258).
THE COASTLINE OF EASTERN SUMATRA AND THE ROLE OF MANGROVES IN LAND EXTENSION
It used to be thought that in historical times the southern Sumatran coast comprised a series of large bays. Obedeyn (1941, 1942) studied old maps dating from 1030 to 1600 A.D. and proposed that both Jambi and Palembang had been ports at the end of promentories. Thus the dashed line in figure 2.10 would correspond to the coast in about 1000 A.D., and the dot-dash line to the coast in about 1600 A.D.. The supposed infilling of the bays by silt since those times required the shoreline to have advanced at a rate of over 100 m per year, but recent studies in the Musi-Banyuasin area had shown that present rates of accretion were over five times less, at only 20 m per year (Chambers and Sobur 1977). This discrepancy was not easy to explain, but Chambers (1980) suggested debris from volcanoes active during earlier periods could have caused the higher accretion rate.
Figure 2.10. Changes in the coastline of part of eastern Sumatra.
After Obedeyn 1941, 1942
It now seems, however, that the coastlines proposed by Obedyn certainly did once exist, but in geological rather than historical time. Investigations have shown that areas near the coast of Jambi had habitation at times when, according to Obedyn, no land would have existed. Cultural remains found 1.75 m below the soil surface at Muara Kumpeh, for example, have revealed stoneware from the twelfth to fourteenth centuries, and a coin from the seventh or eighth century. Chinese sources mention 'Kompel' or 'Kumpeh' during the sixth century (McKinnon 1982). These and other archaeological finds suggest that early rates of coastal accretion were not the same as present rates. Indeed, the current high rates of soil erosion caused by extensive bad land use have doubtless dramatically increased the rates of coastal accretion.
In the southern part of Sumatra, the areas of maximum accretion are not, strangely, directly related to the major drainage systems. For example, opposite Bangka Island lies one of the broadest stretches of 'tidal swamp' in Sumatra, extending over 100 km inland. This appears to be because of the longshore drift carrying sediment eastwards from the Musi/Banyuasin drainage system, and such eastward currents occur in the region for nine months of the year (Wyrtki 1961). Chambers and Sobur (1977) suggest that over half of the deposition of sediment from Musi/Banyuasin may occur outside the river delta itself, because the relatively exposed delta experiences considerable erosion. Thus, no conventional delta can be formed (Chambers 1980).
As described on page 85, mangrove forests do not 'build' land but they do have a vital role in colonising emerging mud banks, consolidating sediments and thereby accelerating the process of land extension (Bird 1982; Bird and Barson 1977; Walsh 1974). The pioneer colonisation by species of Sonneratia produces a network of pneumatophores which have three indirect functions:
• they help to protect the young trees from wave damage,
• they entangle floating vegetation, and
• they provide a habitat for burrowing crabs which help to aerate the soil (Chambers 1980).
If the mud banks were not colonised, it is likely that shifting currents, seasonally increased flows, or strong wave action could wash them away. Thus, at the mouth of rivers, deposition which typifies coastal extension is in the form of gradually enlarging offshore islands which ultimately coalesce (although occasionally separated from the mainland by small tidal channels [Chambers 1980]). This phenomenon can be seen near the mouths of the rivers Besitang, Bile/Barumun, Rokan, Siak, Kampar, and Inderagiri, as well as in the Musi/Banyuasin area.
Along the shore line between the major rivers, the sediment deposition is more or less parallel to the established shoreline, particularly in sheltered bays where wave action is minimal. Species of Sonneratia in these areas are again important as pioneer colonisers (Chambers 1980).
FAUNA
The fauna of mangrove forests can be divided conveniently into two groups: the essentially aquatic component of crabs, snails, worms, bivalves and others which depend directly upon the sea in various ways; and the terrestrial, often 'visiting', component including insects, spiders, snakes, lizards, rats, monkeys and birds (see pp. 109-111) which do not depend directly upon the sea. The majority of what follows, based on Berry (1972), deals with the aquatic component.
The animals mentioned below represent only a fraction of the total mangrove fauna, but the important points to be understood are that there are many different habitats within the mangrove forest, each of which supports a distinct community of animals, and that a proportion of the fauna in mangrove forests is more or less restricted to mangroves. Knowledge of the coastal fauna of the Sunda Region is not yet sufficiently advanced to decide whether any of the mangrove fauna is absolutely restricted to mangroves (Chapman 1976, 1977a; Hutchings and Reicher 1982; MacNae 1968).
The Challenges of Living in Mangroves
In comparison with other shore fauna, the mangrove forest fauna is heavily dominated by many species of crabs and snails. There are few types of worms, and very few bivalve molluscs, coelenterates (sea anenomes), echinoderms (starfish, sea urchins), etc. (fig. 2.11). The dominance of crabs and snails may be largely ascribed to the peculiar conditions found in mangrove forests to which these two groups of animals have evolved successful adaptations.
Animals living on the soil of the intertidal area are subjected to long periods during which they are not covered by the sea. At the mean high water level of spring tides (MHWS), for example, the soil is left exposed for about 270 days per year, and for up to 25 consecutive days; at the lower terrestrial margin, soil is left exposed for about 320 days per year, and for up to 30 consecutive days. Most marine animals cannot stand this because:
• they quickly dry out,
• most cannot respire in air,
• many can only feed on waterborne food, and
• many must release spermatozoa, eggs and larvae into seawater.
Crabs and snails, however, have:
• impervious exoskeletons/shells to restrict water loss,
• many can breathe air,
• many feed on microorganisms or organic materials,
• many climb into the trees to find these foods, and
• they have internal fertilisation and protect their eggs and early developing young in capsules or brood pouches.
Figure 2.11. Percentages of total aquatic animal taxa recorded at three types of shore (omitting microscopic forms). Note that crustaceans and gastropod molluscs account for 74% of the total on the mangrove shore. Note also that these data are not based on complete lists or on equally exhaustive surveys, but they serve to illustrate the major differences.
Data from Berry 1972
The soil in mangrove forests is subjected to salinities ranging from nearly zero to 50%o (Sasekumar 1974). During spring tides, salinities approximate those of tidal seawater, that is, about 27%o-32%o. During neap tides, however, when the landward mangrove soil is not covered by the sea for days at a time, rainfall may reduce salinity to less than 15%o. Conversely, evaporation without rainfall can increase salinity to more than 32%o.
Most marine animals are only able to withstand very minor variations in salinity such as are experienced in the open sea (see again fig. 2.3) because they cannot regulate the salt/water balance of their body fluids except within quite narrow limits. Many crabs are able to cope with salinity variation by precise osmoregulation, whereas many snails can allow the salt concentration of their body tissues to vary without ill effect. These adaptations thus allow crabs and snails to predominate in the mangrove fauna.
Zonation and Characteristics of the Aquatic Fauna
Faunal zonations within mangrove forest are effectively three-dimensional. The vertical component is shown, for example, by the worms and crabs bur-rowing into the mud, some snails living on the mud surface, and other snails and barnacles living in the trees (fig. 2.12). One horizontal component is shown by the different species observed on the foreshore, the seaward mangrove forest and at the limit of tidal influence (fig. 2.13). The other horizontal component is represented by the different features such as small streams, eroding banks, and mud mounds that occur irregularly through the mangrove forest. These forms of zonation set mangrove forest apart from other shore line ecosystems such as sandy shores or rocky cliffs, which lack the large-scale dimension of distance from the sea which can be over 50 km for mangroves on the east coast of Sumatra. The most complete studies of zonation of mangrove fauna in or near the Straits of Malacca are those of Berry (1972), Sasekumar (1974) and Frith et al. (1976).
Figure 2.12. Relative density of several species of tree-dwelling fauna distributed vertically on trees in an Avicennia forest, 10 m from the sea. a-snail Thais tissoti; b-barnacle Balanus amphitrite; c-bivalve mollusc Crassostrea cucullatai d-barnacle Chthamalus withersii; e-mussel Brachyodontes variabilis.
After Tee 1982a
The distribution of the macrofauna (animals visible to the naked eye) is influenced by the nature of the soil (particle size, consolidation, and content of organic matter and moisture), the vegetation and tidal factors. The true mangrove forest (Zone 4 below) has the highest animal abundance and species diversity because of its shade, complexity and abundance of organic detritus (Frith et al. 1976).
Ecologists have found that dividing ecosystems into recognisable subdivisions or zones is fraught with difficulties because vegetation types often intergrade with each other. Nevertheless, as long as the subdivisions are not regarded as rigid, and it is clear on what criteria they are based, such an approach provides a convenient means of describing the different areas (fig. 2.14). The mangrove zones described by Berry (1972) are as follows:
Figure 2.13. Relative abundance of several species of tree-dwelling fauna along a transect through mangrove forest. a-barnacle Balanus amphitrite; b-snail Nerita birmanica; c-snail Littorina scabra; d-barnacle Chthamalus withersii; e-polychaete worm Nereis spp.
After Tee 1982a
Zone 1: Seaward foreshore. As those field scientists who have tried to work in mangroves know to their discomfort, the seashore below the mangrove tree edge is usually extremely soft mud. This soil commonly comprises about 75% very fine sand (particles of 20-210p), most of the remainder being even finer silt (Sasekumar 1974). It extends down from the mean low water level of neap tides (MLWN) and so is covered by every tide of the year and never left exposed for many hours. The fauna is essentially marine with certain crab species, a few snails such as T. telescopium (fig. 2.15) and two or three species of polychaete worms (fig. 2.16) predominating. A variety of mudskippers - an unusual group of fish capable of living out of water for short periods - occur commonly along the water's edge and in burrows in the mud. An excellent review of general mudskipper biology can be found in MacNae (1968).
Underwater, mudskippers breathe just like other fishes but in air they obtain oxygen by holding water and air in their gill chambers. This is supplemented by gas exchange through their skin and fins (Stebbins and Kalk 1961). On land mudskippers move in a variety of ways: they lever themselves forward on their pectoral fins, which move in synchrony with each other (i.e., not true walking), leaving characteristic tracks in the mud, they skip over the mud by flicking their tails, and they climb on vegetation using their pectoral and pelvic fins.
Figure 2.14. Major zones in a mangrove forest. Horizontal distances extremely reduced for clarity. Streams (Zone 5 in the text) are omitted. Figures represent tide coverings per year and days per year with tidal wetting. An eroded bank (Zone 3 in the text) is shown superimposed in dashed lines. Common tree species are indicated above. EHW = extreme high water, MHWS = mean high water of spring tides, MHWN = mean high water of neap tides, MSL = mean sea level, MLWN = mean low water of neap tides.
After Berry 1972
Figure 2.15. Telescopiumtelescopium.
Figure 2.16. A polychaete worm common in mangrove soil -Leiochrides australia; this species commonly reaches 14 cm long.
Although they look very similar, the different species of mudskipper have very different diets; some take mud into their mouths, retain algal material, and blow out the rest; some are omnivorous, eating small crustaceans as well as some plant material; and some are voracious carnivores, feeding on crabs, insects, snails and even other mudskippers (Burhanuddin and Martosewojo 1979; Macintosh 1979).
It can be safely assumed that the modern mudskippers evolved from the same single ancestral mudskipper. It is not possible to determine what its feeding habits were, but it is likely to have been a generalised detritus-feeder. Genetic variations in individuals of this mudskipper made it possible for them to specialise in exploiting one particular food resource or group of food resources with more efficiency and thus greater competitive advantage. Although a taxonomic/ecological study of Sumatran mudskippers has not been undertaken, a study in Queensland has shown that at least five species of predatory mudskippers can live in the same area by successfully partitioning their habitat and its resources between them (Nursall 1981) (see also pp. 155, 240 and 326).
Figure 2.17. Fiddler crab Uca dussmieri.
After Berry 1972
If an observer stands still on the foreshore, animals which have been scared into their burrows by his approach will start to reappear. Hundreds of fiddler crabs Uca emerge, in densities as high as 50 per m2. The larger mudskippers are their main predators (others include snakes, spiders, birds and monkeys) (Macintosh 1979) and it would seem reasonable to ask why the mudskippers 'allowed' such high populations of potential food to exist.
Rates of prey exploitation depend upon the adaptations of both predator and prey. If predators evolved prudence so that they deliberately avoided overexploiting their prey, it would both oppose the pre-eminent goal of evolution which is to maximise the fitness or advantage of the individual, and also require that predator populations were regulated altruistically to maintain an optimum density. Natural selection should favour maximum exploitation of prey by each individual mudskipper at the expense of optimum prey management.
The level of exploitation of a prey population is determined by the ability of the predator to capture prey, balanced by the ability of the prey to avoid being captured. In this case, the speed of attack by mudskippers is balanced by the awareness and speed of retreat into burrows by the crabs. Both of these skill are evolved and tend to result in a more or less stable equilibrium between the numbers of predators and prey.
In an area of mangrove foreshore 10 m wide and 20 m from the sea edge to the first mangrove seedlings, it would not be unusual to find an average of about 10 Uca dussumieri fiddler crabs (fig. 2.17) per square metre - therefore 2,000 over the entire area - and a total of four large mudskippers Periophthalmodon schlosseri. This gives a predator-prey ratio of 1:500. Ricklefs (1979) lists various predator-prey ratios taken from the literature, and their range is 1:30-1:1,263.
Most mudskippers occupy deep burrows, the tops of which are turreted in some species. Their inter- and intraspecific relationships are complex but burrows are usually evenly spaced on the mud, thereby reducing the likelihood of conflicts (Stebbins and Kalk 1961). Males seem to dig the burrows and make displays to attract passing females by jumping, and by erecting their dorsal fins. After pairing, eggs are laid on the sides of the burrow which is then defended by the male. Such territorial defence is rare or absent outside the breeding season (Nursall 1981).
Zone 2: Marine pioneer zone. Actively accreting pioneer zones rise gently through the region of neap low tide and mid-tide levels as bare mud gives way to seedlings and saplings, often of Avicennia and Sonneratia. The soil differs little from that of Zone 1. Most of this zone except the highest parts is covered by all the high tides of the year. This zone includes fewer common marine animals and begins the more typical mangrove fauna. Snails such as Syncera, Salinator, and Fairbankia may occur on the wet soil (fig. 2.18). Mudskippers are common, and the large fiddler crab Uca dussumieri and the smaller U. mani can be very numerous.
Horseshoe crabs (fig. 2.19) are periodically seen in this zone. They are distantly related to spiders but are the sole representatives of a group of animals which dominated the sea during the Palaeozoic period over 250 million years ago. Fossil horseshoe crabs from the Silurian period 400 million years ago are virtually indistinguishable from the five species surviving today. The reason for the lack of change in these animals is hard to explain but it may be that they simply have not experienced effective competition or predation from other species to select for refinements or other adaptations.
Horseshoe crabs live in shallow water along sandy or muddy shores. They can swim through the water by flapping appendages on their abdomens. They feed on soft-shelled molluscs and worms, digging for them using shovelling movements of their shell. When the moon is full and the tide is high, horseshoe crabs come up to the beach to breed. The females, which are larger than the males (van der Meer Mohr 1941b), dig a shallow nest for their thousand or so eggs. The male clings to the back of the female's shell, ready to release his sperm when the eggs are laid. One month later, again at high tide, the eggs hatch and the small larvae join the other marine zooplankton. Sumatran horseshoe crabs are of two species: one with a tail triangular in cross-section, Tachypleus gigas, and one with a tail circular in cross-section, Carcinoscorpius rotundicauda.
An extract of horseshoe crab tissue assists medical science in two unlikely ways. It is used in an extremely sensitive test for checking whether vaccines and intravenous fluids are contaminated by bacterial toxins and in a test for gonorrhoea.
Figure 2.18. Some snails from mangrove soil. a-Fairbankia sp.; b-Syncera brevicula-, c-Cerithidea cingulata; d-Terebralia sulcata-, e-Salinator burmana (a lung-bearing pulmonate snail).
From Berry 1872
Figure 2.19. Horseshoe crab.
Figure 2.20. Peanut worm Phascolosoma.
After Berry 1972
Various species of polychaete worms, occasional bivalve molluscs and the peculiar sipunculid peanut-worm (fig. 2.20) live permanently in the soil of this zone. Peanut worms rest vertically in the soil and sweep their long proboscis over the soil surface, picking up particles of organic matter.
The abundance and type of fauna living on the vegetation depends largely on the age of the tree - older ones have denser populations of more species. Littorina snails (fig. 2.21) occur on almost all the vegetation (Lim 1963), sometimes up to 2 m above the soil surface. Populations of sedentary animals encrust the lower stems of trees as they grow. These animals typically include: barnacles, with the larger ones Balanus amphitrite below and the smaller Chthamalus withersii extending higher; oysters, commonly Crassostrea cucullata; and the small black mussels Brachyodontes sp. which are attached to the tree by 'byssus' threads. A total of 15,401 animals, 9,199 of which were mussels of the genus Brachyodontes, were found on a single Avicennia tree (Tee 1982b). These attached fauna may be eaten off the lower stems by the carnivorous snails Thais and Murex (fig. 2.22). Barnacles and mussels sometimes suffer 50% mortality in the first 10 m above the soil surface, but the number of dead animals decreases upward, indicating that predation higher up the tree is less (Tee 1982b). Figure 2.23 shows the effect of predation by Thais tissoti on the barnacle Balanus amphritrite.
Figure 2.21. Four species of Littorina commonly found in mangrove forest near the seaward edge; a - L. melanostoma; b - L. scabra; c - L. nudulata; d - L. carinifera.
From Berry 1972
Figure 2.22. Two predatory snails Thais sp. (left) and Murex sp. (right).
Figure 2.23. The numbers of the barnacle Balanus amphritrite found dead and alive at different heights up a mangrove tree stem, with the percentage of dead animals. The shaded portion indicated by the arrow represents those animals probably predated upon by the predatory snail Thais tissoti.
After Tee 1982b
Two or three species of hermit crab are found in this zone. These crabs have lost the hard protective carapace over the rear part of their bodies and depend on finding empty shells for protection. The combination of security and mobility of the adopted shells is clearly advantageous and hundreds of hermit crabs can sometimes be seen on a beach. Suitable shells are, unfortunately for the crabs, a limited resource and this affects growth and reproduction. It has been found that crabs with roomy shells put their energy into growth and do not reproduce, whereas crabs with tight shells, stop growing and put their energy into reproduction (Bertness 1981a). The scarcity of shells leads to active competition between species of hermit crabs and this is described by Bertness (1981b). A review of hermit crab ecology is given by Hazlett (1981).
Figure 2.24. Schematic diagram of a meandering river near the sea to illustrate the position of eroded banks (heavy lines).
Zone 3: Eroded banks. The seaward edge of mangrove forest is often marked by a nearly vertical bank 1-1½ high instead of a sloping pioneer zone. This is caused by current sweeping away the silt and is most obvious on the outer bend of rivers near the sea, where the current flows faster than on the inner bend (fig. 2.24). The bank may be broken in places and mangrove trees at its edge often fall into the sea as the soil is slowly eroded away. The top of this bank is usually at or slightly above the mean high water of neap tides, and may sometimes be left 9-10 days without tidal cover. The soil of an eroded bank resembles that of the mangrove forest floor behind it, with less fine sand (commonly about 65%) than in Zones 1 and 2.
The bank is burrowed into by various crab species (Berry 1963).
Zone 4: True mangrove forest. The mangrove fauna of this zone differs in several respects from the preceding zones. Most of this zone is very flat and the soil surface is exposed to the air for an average of 27 days per month. However, since the trees provide heavy shade, the humidity is very high, so the soil rarely dries out. There is also abundant leaf litter and other organic matter so that detritus-feeders abound. The soil contains even less fine sand and more of the finer silt and clay particles. There is also more organic matter than in soils near the water's edge.
Figure 2.25. Nerita birmanica.
Figure 2.26. Two ellobid snails, a - Ellobium aurisjudae; b - E. aurismidae.
From Berry 1972
About 75% of the fauna of this zone is not found in the other zones (Frith et al. 1976), and is best divided into three groups: tree fauna, soil surface fauna and burrowing fauna.
Tree fauna. Perhaps the most striking change is the rapid decrease of encrusting animals - which rely on frequent inundation - on the lower stems and trunks of the trees. The remaining (mobile) tree fauna is largely composed of snails such as species of Littorina and Nerita birmanica (fig. 2.25), which are also found in Zone 2. Further back from the sea Cerithidea obtusa and Cassidula are found, and even further inland, species of Ellobium occur, (fig. 2.26) (Lim 1963). Most of these feed on algae and move up trees when tides wet the soil, but Littorina very rarely leaves the tree trunks. All these snails are able to breathe efficiently in air and some, the pulmonates such as Ellobium, have lungs.
Soil surface fauna. These animals comprise mostly crabs (see below) and snails, although the medium-sized mudskippers Periophthalmus vulgaris can be common. Among the snails, actual distance from the sea seems to matter less than details of ground conditions. In wetter areas such as where water drains into small gullies, Syncera brevicula is more common than anywhere else in the mangrove forest.
Molluscs at the seaward edge (Zone 1 and 2) of the mangrove forest comprise a mixed sample of gastropods and bivalves. Further back in the mangrove forest, however, carnivores and filter-feeding molluscs disappear. The vast majority of the molluscs in the true mangrove forest feed by grazing on algae or microorganisms on the soil surface. Little is known about molluscs' reproductive behaviour in mangrove forests but most have internal fertilisation of eggs and, unlike many aquatic snails, have eggs that develop directly into small snails rather than waterborne larvae.
Burrowing fauna. Nearly all the mangrove crustaceans and worms make burrows which reach down to the water table. Many different types of tunnels are constructed (fig. 2.27), and the elliptical tunnels of the edible crab may slope down from the bank of a river for as far as 5 m.
In general, these burrows serve as:
• a refuge from predators at the surface,
• a reservoir of water,
• a source of organic food,
• home for pairing and mating, to be defended, and
• a place for brooding eggs and young, although no one species uses a burrow for all these purposes.
Figure 2.27. Different types of animal burrows in mangrove soil. A- crab Scylla serrata (with transverse section of burrow); B- mud lobster Thalassina anomala (not all burrows open below ground as shown here); C- crab Uca spp. (burrows may reach water table nearer the low-tide level); D- crab Sesarma spp.; E- peanut worm Phascolosoma; F- large mudskipper Periophthalmodon schlosseri; G- smaller mudskipper Boleophthalmus boaerti, H- pistol prawns Alphaeus spp.; I- smaller mudskipper Periophthalmus vulgaris.
From Berry 1972
Figure 2.28. Sesarma sp.
As mentioned on page 91, most mangrove crabs are able to tolerate long periods out of water, and their respiratory systems are obviously specialised. Typical crabs inhale water under the sides of the carapace, between the legs and into the gill chambers. Deoxygenated water is expelled on either side of the mouth. Out of water, sesarmid crabs (small crabs with a characteristic square carapace) allow air into the moist gill chambers. Sesarma (fig. 2.28) can even recirculate the water around the carapace where it becomes reoxygenated (Malley 1977).
The most abundant, though largely invisible, source of food on the forest floor is a mixture of organic deposits, much of it originating from decomposed leaves and other vegetable material, mixed with a flora of diatoms, bacteria and other microorganisms growing on the deposits. Many crabs and other mangrove animals feed upon either the organic matter, the microorganisms, or both, picking up this 'mud' with their pincers or chelipeds. Male fiddler crabs have one huge pincer which is useless for feeding and the males are forced to eat half as fast as the females, which have two normal-sized pincers; presumably the males have to eat for twice as long. This seemingly unnecessary encumbrance is used to attract females for mating and for warning away males. It would serve a mutant male fiddler crab little if its larger pincer was reduced in size and could be used for feeding, if the animal could not then attract females for mating.
In the landward areas of true mangrove forest, the first signs are seen of an animal that is itself rarely seen, the mud lobster. This animal builds volcano-like mounds of mud which can reach over a metre high (see again fig. 2.27), and feeds on mud, digesting algae, protozoa and organic particles (Johnson 1961). The burrow below the mound is 1-3 m long, extending down to the water table. The entrance leading to the main burrow is generally plugged with layers of earth. The habit of burrowing deeply in anoxic mud, closing itself off in poorly oxygenated air and water, suggests that it may have evolved means of anaerobic respiration (Malley 1977). Research into this secretive creature would not simply be for academic interest, because its mounds are often regarded as a nuisance. Managing the numbers of distribution of mud lobsters requires a knowledge of their biology and ecology.
The bivalve mollusc Geloina (fig. 2.29) lives buried in mud and can occasionally be found in this zone, but is more common in mangrove forests on islands in or near river deltas. Geloina is remarkable in its ability to feed, respire and breed so far from open water and at levels where it is sometimes not covered by tidal seawater for several weeks at a time.
Zone 5: Rivers, streams and gullies. The banks and beds of water courses in the mangrove forest have a fauna which is generally distinct from that on the general forest floor. For example, many of the forest floor species of polychaete worms and crabs are missing, whereas juvenile fiddler crabs and some snail species are more common. The edible crab with paddle-like rear legs occurs in the larger streams and rivers in the mangrove forest where it is caught in traps.
Figure 2.29. The bivalve mollusc Geloina ceylonica.
From Berry 1972
Zone 6: The terrestrial margin. Far back from the sea, mangrove soil is covered by fewer and fewer tides and suffers longer and longer intervals of exposure. Unlike other types of shore, there is virtually no wave action in mangrove swamps to carry seawater higher than the true tidal level, because the wave energy is absorbed by the abundant vegetation. Thus, animals in this zone live on a salt-impregnated soil but are covered by seawater only at irregular and infrequent intervals. Insects, snakes, lizards and other typically terrestrial animals are much more common here than further seaward. Sesarma and some large crabs occur here and in the Nypa palm swamps behind. Their basic requirement is that their burrows must reach down to the water table. The mud lobster has one of the deepest burrows and is therefore seen some distance from the mangrove forest, for instance in coconut plantations. Where there is moving water some snails, such as Fairbankia and Syncera, and the large mudskipper Periophthalmodon schlosseri may be found.
Biomass of Aquatic Fauna
Only one estimate of biomass of aquatic mangrove fauna seems to have been determined in the Sunda Region. Tee (1982b) calculated the dry-tissue biomass of tree-dwelling aquatic fauna in four mangrove forest zones in Selangor, Peninsular Malaysia (table 2.5). The results show a general reduction of biomass with increasing distance from the sea. It is supposed that the turnover rate (time for one generation to replace another) must be quite fast because most of the tree-dwelling aquatic fauna are in the lower trophic group (i.e., filter-feeders), none of which have long life spans.
Terrestrial Fauna
Insects, birds and mammals live chiefly in the canopy of mangrove forest. Ground-living animals such as wild pig, mouse deer, wild cats, rats and lizards only venture into the landward edge of mangrove for brief periods. The two mangrove mammals seen most often are the silvery leaf monkey Trachypithecus cristata and the long-tailed macaque Macaca fascicularis. The leaf monkeys are largely arboreal and eat leaves, shoots and fruit, whereas the macaques are mainly ground-dwelling and eat crabs, peanut worms and small vertebrates as well as some leaves, shoots and fruit. Faecal analyses have also shown that both species eat the nectar-rich flowers of Sonneratia (Lim and Sasekumar 1979).
Leaves are a poor source of food for most mammals because the cellulose molecules are very long and are therefore difficult to break down without special digestive adaptations. Leaf monkeys have evolved a highly sacculated stomach, like a cow's, in which special bacteria ferment and break down cellulose so that it becomes digestible.
From Tee 1982b
Many tree leaves are defended against being eaten by herbivores by defence compounds, the most common of which are fibre (lignin) and phenols, particularly tannins (p. 88). In a normal acidic digestive tract (the content of a human stomach are generally at pH 3), tannins will bind strongly onto any protein, both food and enzymes, thereby hindering the digestive process.3 In the leaf monkey's complex stomach, however, the pH is probably maintained between 5.0 and 6.7 (Bauchop and Marrucci 1968), and in such conditions not only are the fermentation bacteria able to thrive, but the attachment of tannins onto proteins is weaker. Both silvery leaf monkeys and long-tailed macaques have been observed to eat bark from the twigs of Bruguiera (Lim and Sasekumar 1979), which have considerable quantities of tannins - perhaps 30% dry weight. For the macaques, which have a simple stomach, this may have been eaten as a stomach purgative (p. 232).
Fruit bats, particularly the flying fox Pteropus vampyrus, roost in the mangrove forest canopy and other bats such as the cave fruit bat Eonycteris spelaea and the long-tongued fruit bat Macroglossus sobrinus are very important in the pollination of the flowers of Sonneratia (Start and Marshall 1975) (p. 329).
The mangrove frog Rami cancrivora is exceptional among amphibians in being able to live and breed in weakly saline water. The tadpoles are more resistant to salt than the adults and metamorphosis into adults will only occur after considerable dilution of the salty water (MacNae 1968).
Mangroves are inhabited by a variety of reptiles such as the monitor lizard Varanus salvator, the common skink Mabuya multifasciata, mangrove pit viper Trimeresurus pupureomaculatus and the common catsnake Boiga den-drophila. Most snakes seen in or near mangroves are not in fact sea snakes (Hydrophidae), which live primarily in open water (Voris et al. 1978; Voris and Jayne 1979; Voris and Moffett 1981). Potentially the largest animal of the mangrove swamps is the estuarine crocodile Crocodilus porosus. Persecution for centuries has reduced its numbers to a very low level and it is possible that large specimens (they can exceed 9 m) no longer exist around Sumatra (p. 185).
The most conspicuous of the insects are mosquitoes, but only one of these, Anopheles sundaicus, carries malaria. The larvae of this species can live in water with a salinity of 13%o (Chapman 1977b). Species of Aedes mosquitoes have been seen feeding on mudskippers, but they are also attracted to human skin. Another potential insect hazard is the leaf-weaving ant Oecophylla smaragdina, the workers of which, if their nest is disturbed, can inflict a very painful bite. The making of the nest is extraordinary because the adult ants have no means of producing silk to join the leaves of the nest together. The larvae do have silk glands, however, originally intended so they could weave cocoons to pupate within, but they are not now used for this. Instead, when a leaf is to be added to the nest or a tear repaired, some of the worker ants seize the edges to be joined and hold them in the required position. Other workers enter the nest and collect a larva. The larva is held in the worker's jaws and while it produces the sticky silk, it is moved back and forth between the leaf edges, like a tube of quick-drying glue (Tweedie and Harrison 1970).
One insect, the caterpillar of the moth Olethreutes leveri, has managed to extend its niche and avoid competition by spinning together leaves of Sonneratia alba. It is able to survive periodic immersion in seawater because air is trapped within the web. Experiments showed it could withstand 8 1/2 hours immersion within the web, but died quickly if immersed in seawater without the web (Lever 1955).
BIRDS
Mangrove forests on the west of Peninsular Malaysia are used by at least 121 species of birds which are probably almost identical to those found in Sumatran mangroves. The most important trees for nesting are probably Sonneratia, which are both tall (±30 m) and have occasional holes in their trunks which are used by woodpeckers. The landward side of a mangrove forest generally has more birds of more species than the seaward side, but birds are found throughout.
Nisbet (1968) divided the list of bird species observed in mangrove forests on the west coast of Peninsular Malaysia into 10 groups (table 2.6) according to the manner in which they use the ecosystem. Mangroves are thus important to birds in at least four ways:
• they provide nesting sites for a number of large species of herons and storks (fig. 2.30), and also for at least three birds of prey and two owls. Many of these species nest nowhere else, and presumably would be subjected to much greater predation (certainly by young humans) if they attempted to do so.
• they provide roots for many migrant species (and a few residents) which feed on or over the tidal mud flats. These roosts are particularly important for the migrant waders.
• they are visited seasonally by a number of pigeons and parrots, both from inland and outlying islands. One of these, the pied imperial pigeon Ducula bicolor (fig. 2.31), is thought to be the main source of seeds reaching Jarak Island in the Straits of Malacca. Of the plants growing on the island, 90% have fruit dispersed by birds (Wyatt-Smith 1951).
• they support a varied fauna of resident and migrant land-birds. Eight species (groups A and B in table 2.6) occur in no other habitat and 17 others (group E) occur largely within mangrove forest. This assemblage of birds is distinctive in that it is largely made up of predatory kingfishers, insectivorous woodpeckers and warblers and nectar-drinking sunbirds. Some of the dominant frugivores of adjacent forest (pheasants, hornbills, barbets, leaf-birds, bulbuls and babblers) are more or less absent. Interestingly, most of the species in groups A and B occur outside mangrove forests in other parts of their range.
Figure 2.30. Storks, egrets and herons commonly seen near mangrove forests. a-woolly-necked stork Ciconia episcopus; b-little egret Egretta garzetta; c-Chinese egret Egretta eulophotes; d-plumed egret Egretta intermedia; e-great egret Egretta alba; f-grey heron Ardea cinerea; g-purple heron Ardea purpurea; h-great-billed heron Ardea sumatrana.
Figure 2.31. Pied Imperial Pigeon Ducula bicolor.
A significant feature of the resident land-bird fauna is that most of the common species (in group E and I) are those which, in other parts of Malaysia, and indeed Sumatra, are characteristic of semi-open habitats -villages, towns, gardens, riversides and clearings. This is perhaps most surprising because mangrove forest is dense and much more similar in structure to secondary forest. Since semi-open habitats are largely man-made, it is likely that many of these species were, until recently, confined to a narrow coastal strip of mangroves and scrub and to clearings and river-banks in the forest. It should be said, however, that the species which have benefited most from the clearance of inland forest are not very numerous in mangrove, while most of the dominant mangroves species are still primarily coastal in distribution.
Some of the species which are most specialised for life in mangrove forest are replaced in inland areas by an obvious potential competitor. For example, the ruddy kingfisher Halcyon coromanda breeds in forest in the northern part of its range (India, E. China, Japan, Taiwan) but is confined to mangroves as a breeding species in Peninsular Malaysia and Sumatra where the forest is inhabited by the rufous-collared kingfisher Halcyon concreta. Thus it seems that, as inland forest was opened up by man, the unspecialised coastal species (group I) were able to spread inland, whereas the species which were more adapted to life in mangrove forest remained confined to the mangroves and nearby coastal scrub.
From Nisbet 1968
EFFECTS OF THE FAUNA ON THE VEGETATION
Very little is known about the effects of the activities of animals on the mangrove forest. The holes constructed by crabs and mudskippers may affect rates of accretion or erosion and increase aeration of the soil, the mounds of the mud lobster and the activities of polychaete worms may significantly influence the rate of nutrient/energy flow in the ecosystem, inshore fish that graze on young seedlings may cause slower colonisation, etc., but there have been no studies in the Sunda Region on these or related topics.
Some of the known effects are rather surprising. Root tips of Rhizophora trees which have not yet reached the ground are sometimes attacked by small isopod crustaceans. They burrow into the soft root tip, form a chamber and breed. Such destruction of growing points, as with most plants, results in the formation of multiple side roots. It had been suggested that the activity of these isopods may benefit Rhizophora by increasing the number of roots and therefore the number of channels by which nutrients can reach the tree. A higher production of roots per tree would either result in a faster expansion of adjacent areas of Rhizophora or in a higher number of roots per unit area. In Florida, the number of roots per unit area was indeed higher where these trees were parasitised by isopods than at sites where they were absent (Ribi 1981, 1982). The effect of the increased root density on silt settlement and thence on land formation has not been measured.