RATTAN IN EAST KALIMANTAN, INDONESIA: SPECIES COMPOSITION, ABUNDANCE, DISTRIBUTION AND GROWTH IN SOME SELECTED SITES

  Johan L.C.H. van Valkenburg

1. Introduction

Apart from timber, rattan is the most important forest-derived commodity in East Kalimantan. The multitude of species, that all have their specific use in both the traditional way of life and commercial trade, makes rattan stand out among non-timber forest products.

Despite the longstanding exploitation of rattan and early records of rattan gardens in East Kalimantan (Endert, 1927; Van Tuil, 1929), the resource remains poorly known. Whereas rattan floras for Sabah and Sarawak exist (Dransfield, 1984, 1992a) none is available for East Kalimantan. This is a serious handicap since rattan species tend to display a rather high degree of endemism, e.g. of the 107(--109) species and varieties reported for Sarawak, 63 species and 4 varieties are endemic in Borneo (Dransfield, 1992a, 1992b), and 23 species and 1 variety are endemic in Sarawak (Pearce, 1989; Dransfield, 1992a). For the island of Borneo as a whole, so far 146 rattan species have been recorded (Dransfield, 1992c). For Sabah, 82 species and varieties are reported, with 8 endemic species and 2 endemic varieties (Dransfield & Johnson, 1989). Species richness in West Kalimantan is assumed to be similar to that of Sarawak, and in East Kalimantan similar to that in Sabah (Dransfield, 1992c).

If rattan is to be used as a renewable resource, knowledge of the growth of rattan plants is essential to devise sustainable exploitation practices. The growth of non-exploited rattan in forest of various successional stages (see 3.2) is compared with the growth of rattan plants subjected to harvesting (see 3.3). Knowledge with respect to impact of harvesting on the rattan resource is essential for sustainable exploitation.

The important role of light for establishment and growth of rattan plants was observed in planting trials of four species (Calamus caesius, C. manan, C. scipionum and C. trachycoleus) in Malaysia and Indonesia (Wan Rhazali Wan Mohd. et al., 1992; Wong & Manokaran, 1985). The effect of logging resulting in an increase of light, however, appeared to have negative effects on the rattan resource (Abdillah Roslan & Phillips, 1989; Kiew & Hood, 1991), whereas harvesting intensity affects the vitality of plants (Nandika, 1938; Kiew & Hood, unpublished)

In this paper attempts are made to answer the following questions:

· What is the species composition and abundance of rattan in the various sites?

· Does logging affect the species composition and abundance of rattan?

· Does logging have an effect on the subsequent increment of rattan both at the population and individual plant level?

· What is the effect of harvesting on the performance of a rattan clump?

Figure 1. Map of the study area

2. Methods

2.1 Permanent plots

Permanent plots were set up in 1992 to assess species composition, abundance and growth of rattan. Plots were located in the Wanariset area, the ITCI concession, and the Apo Kayan (Figure 1; for details see Van Bremen et al., 1990; Van Valkenburg, 1997). Within each site the plots were situated in a topo-sequence from ridge top to valley floor. In the ITCI concession both primary and logged-over plots were studied.

In the Wanariset forest part of plot Matthijs (5100 m²) was inventoried. In the ITCI concession permanent plots established by the concessionaire were used, and either the entire plot or part of the plot was inventoried as follows: 72-1 (5000 m²), 72-2 (5000 m²), 76-3a (5000 m²), 76-3b (7500 m²), 76-4 (3200 m²) and 77-2 (1600 m²). In the Apo Kayan four plots of 1600 m² were established.

At all sites rattan plants in the plots were labelled with numbered aluminium tags using either nylon fish-line or iron-wire. Within a population, only plants with a minimum cane length of 50 cm were included. The growth stage of all shoots was recorded.

The abundance of rattan in the permanent plots was studied from three different aspects:

· number of plants/hectare

· number of plants with mature canes/hectare

· total number of mature canes/hectare

Although from an ecological point of view the total number of plants is most important, from the economic point of view the number of mature canes is more relevant to this study.

2.2 Additional plots

As the total area of the permanent plots in the Apo Kayan was considered not sufficient to obtain a good impression of rattan abundance over larger areas (i.e. the Apo Kayan in general), a line plot of 4x1200 m was set up in October 1992. A greater variation in topography and growth stages of the forest was included and effects of the often very local occurrence of rattan plants were compensated. Rattan plants with mature canes of all species and the total number of mature canes were recorded. Plants belonging to (potentially) commercial species were labelled and the number of shoots belonging to the different growth stages counted.

Also in the ITCI concession some additional inventory work was carried out to compensate for effects of the often local occurrence of rattan plants and species. In 1994 two plots of 1600 m² each were inventoried inside the permanent plot 72-8 situated in primary forest. All rattan plants with a minimum cane length of 50 cm were labelled and growth stage of all shoots was recorded.

2.3 General collecting

In addition to recording rattan species in the permanent plots, supplementary collections and observations were made in the three main research sites, Wanariset forest, ITCI concession and Apo Kayan in the vicinity of Long Sungai Barang. Other qualitative inventories of varying intensity were also conducted in the villages of Dilang Puti (June/July 1993) and Eheng (May 1994) upstream from the lakes on the Mahakam river and in the Bahau region. In the Bahau region, the inventory was carried out during a visit at the basecamp of World Wildlife Fund Indonesia, in the Kayan Mentarang Nature Reserve near Long Alango (November/December 1991). The listing of rattan species in the Meratus Mountains is based on herbarium specimens available at the Wanariset Herbarium.

Solving taxonomical problems was not the aim of this research. However, three new species were described (Van Valkenburg, 1995). Plants belonging to the species complex of Calamus pogonocanthus, C. erioacanthus, and C. semoi have for convenience been divided into three artificial taxa and are referred to as Calamus pogonocanthus 1, 2 and 3. The use of all three taxa is the same: namely as strips for good quality binding

2.4 Growth measurements

As a rattan seedling grows, there is a gradual increase in stem diameter, before the stem starts significantly to grow upwards. This process can take many years depending on the species and the light condition in which it grows. This initial phase is known as establishment growth, and a vigorous plant in this initial phase is referred to as rosette stage/plant.

Since rattan stems often become entangled in the canopy, the length of the canes could not be accurately recorded without pulling them down. This was not done because it would result in an altered light environment for the crown-leaves and would thereby influence subsequent growth of the plant. An alternative method for recording growth over a 2-year period at time intervals of 12 months was therefore developed. All shoots were classified as belonging to a specific growth stage as defined in Table 1. Growth was defined as the passage from one stage to the next.

At the beginning of the study only plants with a minimum cane length of 50 cm were recorded. If in the following years a plant no longer had a shoot with a minimum cane length of 50 cm, but suckers were present, they were still monitored. Not all mature canes are harvestable, since a minimum length of 3 meters of sufficiently mature cane is required. The distinction into harvestable and non-harvestable canes was only made at the end of the study, when canes were harvested in the Apo Kayan.

However, shoots of some species could not be classified in this way; e.g., Calamus conirostris and Daemonorops hystrix often flowered when cane length was still less than 200 cm and the lower leaf sheaths were still green. For these plants the flowering shoots were classified as mature. The lower leaf sheaths of Calamus laevigatus are normally still green when the cane has reached the canopy and is more than 30 m long. Since the cane can be harvested in this state, it was classified as mature.

The mature plants present in the Apo Kayan line plot were remeasured after 17 months.

Table 1. Growth stages of rattan

Stage	Length of cane	Appearance of lower leaf-sheath
Sucker	0 - 49 cm	green
Juvenile	50 - 199 cm	green
Immature	> 200 cm	green
Mature	> 200 cm	dead or bare cane exposed

2.5 Effect of disturbance on recruitment

In order to judge the significance of disturbance on the recruitment of rattan, the twelve plots were compared using a procedure fit-curve (Genstat 5, 1990). Each plot was classified as either disturbed or undisturbed depending on the presence of degrading canopy trees, a chablis, or serious effects of drought. Furthermore plot 72-2 was studied in detail, whereby all 50 subplots were tallied according to the degree of disturbance (three classes distinguished: 0, 1 and 2, with disturbance increasing from 0 to 2).

2.6 Harvesting experiment

In a non-exploited state a rattan clump might invest in elongation of already mature canes that have reached the canopy. Or in response to changes in the environment, an acceleration of the growth of dominant shoots might prevail. If the mature canes are harvested, growth of the remaining shoots may be released. Acceleration of the growth stage of the minor/young shoots might be induced, or new suckers could be formed.

In addition to the harvesting of canes in the Apo Kayan permanent plots and line plot at the end of the study in 1994, rattan plants outside these plots were subjected to maximum harvesting. The experiment was not only conducted to obtain a better estimate of harvestable length per plant but in particular to supply information on the effect of harvesting on the growth of remaining shoots.

In the Apo Kayan rattan plants of three locally important and (potentially) commercial species: Calamus javensis (n=31), Calamus ornatus (n=33), and Daemonorops sabut (n=33) were harvested in May 1993. Each plant was labelled with an aluminium tag. The total number of shoots belonging to the different growth stages was recorded and habitat of the plant was classified in terms of silvigenetic stage (Oldeman, 1980, 1990) of the surrounding forest, geomorphology/topography (valley floor/middle slope/upper slope/ ridge crest) and steepness of slope. Harvestable mature canes, with a minimum length of 3 m, were cut and their total length measured. The remaining shoots were tagged according to their growth stage. After 12 months the growth response of the plants and the individual shoots was recorded, and canes that had attained harvestable size were cut.

3. Results

3.1 Species composition, abundance and distribution

In general neither species composition nor abundance of rattan are evenly distributed.

Although in the Middle Mahakam region distribution and abundance of rattan species is strongly influenced by man, in other areas of the province it is presumed to be the result of natural phenomena. Influence of man on abundance and distribution is increasing due to the extensive logging activities and accompanying agricultural development.

Species can be geographically and/or ecologically (e.g. Korthalsia flagellaris confined to peat swamp forest) limited in their distribution. Variation in climate, natural boundaries or recent speciation may be reasons for their geographical limitation. As yet no information on distribution of rattan in East Kalimantan is available, because the province has been poorly collected. Of the 59 species I collected, only 21 were already represented by herbarium specimens from East Kalimantan in the collection at Leiden or Wanariset Herbarium. The information on species distribution presented in Table 2 is, therefore, based on the surveys in a limited number of sites in East Kalimantan.

Personal observations indicate that the following species are of common occurrence in East Kalimantan: Calamus javensis, C. ornatus, C. pilosellus, C. pogonocanthus 3, Ceratolobus concolor, Daemonorops didymophylla, D. fissa, D. hystrix, D. korthalsii, D. sabut, Korthalsia echinometra, K. ferox and Plectocomiopsis geminiflora. However, there is a difference in preferences for moisture regime and degree of disturbance.

Some species are restricted to one of the areas where they were locally common whereas other species were simply very rare. Daemonorops atra, D. crinita, D. pumila, Calamus nigricans and Ceratolobus subangulatus were common species where they occurred. While Calamus gonospermus, C. hispidulus, C. paspalanthus, C. sarawakensis, Korthalsia hispida and Plectocomiopsis mira occurred at very low densities.

Calamus mattanensis and C. tomentosus appear to be confined to the western part of East Kalimantan whereas C. blumei, C. pandanosmus, C. rhytidomus, C. fimbriatus and Korthalsia furtadoana were only encountered in the eastern part.

Some species, e.g. Calamus laevigatus, and C. ornatus, are more or less evenly distributed. While others are always encountered in groups e.g. Calamus convallium, C. pandanosmus, C. caesius and Korthalsia cheb. On dry ridges Calamus conirostris can become the dominant species; while in the Wanariset area C. marginatus takes this dominant role.

Table 2. List of rattan species recorded in different areas of East

Kalimantan in the period 1991-1994 (for details see text)

– : Only present outside permanent plots

Wanariset: Plants present in the Wanariset research forest (incl. plot Matthijs) and vicinity

including Sungai Wain area and Inhutani I concession

ITCI : Plants present in the permanent research plots and vicinity of these plots

Apo Kayan: Plants present in the Apo Kayan primarily vicinity of Long Sungai Barang,

and the primary forest eastward of the Kayan River

Meratus : Some sporadic collections of the Meratus mountains in the PT ITCI concession area

Bahau : Plants collected in the Kayan Mentarang Nature reserve, during a visit at the WWF

basecamp near Long Alango

Mahakam: Plants collected in the vicinity of Dilang Puti and Eheng

Wanariset forest

In plot Matthijs 20 rattan species were found on 5100 m². Of these 18 species had mature canes (Table 3). Three species were not encountered in the Apo Kayan or ITCI permanent plots: Calamus nigricans, C. pogonocanthus 2, and Plectocomiopsis mira.

Calamus marginatus, usually a single-stemmed species of dry upper and middle slopes, was the dominant species accounting for 36 % of all plants. More than half of these plants had mature canes representing 25 % of total mature canes present in the plot (Table 3). Korthalsia rigida is second in frequency (15 % of all plants) and almost half of its plants had mature canes, representing 15 % of total mature canes.

Five species each represented 5-10 % of all plants: Daemonorops sabut, Korthalsia furtadoana, Calamus flabellatus, C. nigricans, and C. pogonocanthus 2. Four clustering species each represented 5-10 % of the mature canes in the plot: C. javensis, C. flabellatus, K. furtadoana and D. sabut.

ITCI concession area

A total of 24 rattan species (see Table 1) were encountered in the ITCI permanent plots (30,500 m²). Six of these species (Calamus caesius, C. pandanosmus, C. pogonocanthus 3, C. rhytidomus, C. scipionum and Daemonorops cristata) were present neither in the Apo Kayan or Wanariset permanent plots. The primary and logged-over plots had 14 species in common.

Table 3. Presence of rattan species in primary forest, Wanariset, plot Matthijs (in 1994)

Number of plants ha^-1 (a), plants with mature canes ha^-1 (b) and total number of mature canes ha^-1(c)

Primary plots

A total of 18 species were encountered on 18,900 m² (Table 4). Daemonorops cristata was only found in the ITCI primary plots. Two other species were restricted to the ITCI permanent plots: Calamus pandanosmus and C. rhytidomus.

On ridge crest and upper slope, the abundance of rattan plants appears to be lower than on middle slopes. The total number of plants in the plots was low, and differences between plots in similar habitats were considerable.

In plot 76-3b a dense under-storey of Borassodendron palms was present in 21 % of the 100 m² subplots. Eighty-seven percent of the subplots with Borassodendron was void of rattan as compared with 47 % of the other subplots. The dense leaf litter that prevents seedling establishment and competition for light are plausible explanations for the absence of rattan.

Table 4. Presence of rattan species in permanent plots in primary forest in the ITCI concession area (in 1994)

Number of plants ha^-1 (a), number of plants with mature canes ha^-1 (b) and total number of mature canes ha^-1(c)

Plots arranged in a topo-sequence from ridge crest to valley floor

Logged-over plots

A total of 20 species were encountered on 11,600 m² (Table 5). Two species, Calamus caesius, C. scipionum were found only in these plots, although they are reported to be common in Southeast Asia/Borneo (Dransfield, 1992b). The low number of species in plot 77-2 as compared with the other plots can be ascribed to the smaller size (1600 m² versus 5000 m²).

The abundance of rattan plants in the logged-over plots was highest in the subplots with a canopy of pioneer trees or in the transitional zone with primary forest. On the other hand, some subplots with a canopy of pioneer trees were void of rattan (in plot 77-2).

The logged-over and primary plots at ITCI have many species in common. Only four species present in the primary plots were not recorded in the logged-over plots, two of which were only seldom encountered (Calamus fimbriatus, Daemonorops cristata); one species was present in all of the primary plots and is typical of dry upper and middle slopes (Calamus conirostris). The fourth species (Calamus ornatus) was very rare and only represented by a single plant.

The similarity between logged-over and primary plots is caused by/has to be ascribed to the fact that a logged-over forest consists of patches of logged and primary forest. The balance between logged and primary patches depends on logging intensity and previous distribution of timber trees. In the logged-over plots, rattan species associated with primary forest can still survive. The difference is in an invasion of rattan species preferring more open growth conditions.

Table 5. Presence of rattan species in permanent plots in logged-over forest in the ITCI concession area (in 1994). Number of plants ha^-1 (a), number of plants with mature canes ha^-1 (b) and total number of mature canes ha^-1(c)

The abundance of plants in primary forest is on average lower than in logged-over forest (102 versus 230 plants ha^-1, see Table 7). The difference is higher when plants with mature canes are compared (29 versus 109 plants ha^-1), and highest when the total number of mature canes is compared (34 versus 262 canes ha^-1). The high contrast in mature cane numbers is the result of the presence of profusely clustering species (Calamus pandanosmus, Korthalsia echinometra and K. furtadoana) in the logged-over plots.

Apo Kayan

In total 26 rattan species were found in the Apo Kayan area. Local variation in abundance and species composition of rattan in apparently similar habitats was encountered in primary as well as secondary forest. Areas with reported fertile soils in the vicinity of the village of Long Sungai Barang are all used in the agricultural cycle. Primary forest is therefore no longer found on fertile land. Young secondary vegetation is poor both in species composition and abundance of rattan, as a result of burning that follows clearing of the land for a swidden. No seed-bank is present as rattan seeds quickly loose their viability (Yap, 1992) and the weeding during the rice cycle prevents regeneration by means of resprouting or establishment of new plants (pers. observ.).

Typical species of older re-growth are Plectocomia mulleri, Daemonorops fissa, Ceratolobus concolor and Calamus pogonocanthus 3. On riverbanks bordering the Kayan river Daemonorops hystrix was found, but the species was absent in the primary forest plots situated on upper slope and ridges. In swampy sites locally dense stands of Korthalsia cheb are found. This species is sought for its durable cane that is particularly suited as binding material for fish-traps, and the shoot is edible.

The nearest stands of the highly prized Calamus caesius are two days' (by boat/on foot) Southeast of the village, along the Boh river. In Long Sungai Barang, Daemonorops sabut is used as a substitute for C. caesius in the weaving of backpacks and rice-mats. Rattan is neither planted nor protected by the people of Long Sungai Barang.

The permanent plots

The four permanent plots harbour 18 species on 6400 m². Of these 16 species have mature canes (Table 6). Eight species present were not encountered in the ITCI or Wanariset plots: Calamus hispidulus, C. mattanensis, C. muricatus, C. tomentosus, C. pogonocanthus 1, Daemonorops atra, D. fissa, and D. pumila.

Species composition and abundance of individual species gradually change going down from ridge crest to valley floor. The abundance of rattan plants is lowest at the valley floor even when plants of Calamus conirostris (the dominant species of upper and middle slope) are excluded. The abundance of plants with mature canes at the valley floor is only 16 % or 36% (if C. conirostris is excluded) of the average abundance on upper and middle slope. The difference is smaller when the total number of mature canes is compared, 23 % and 47 % respectively.

When all permanent plots are combined in four groups (Apo Kayan, ITCI primary, ITCI logged-over and Wanariset) and compared, only three species are seen to occur in all four groups: Ceratolobus concolor, Daemonorops didymophylla and D. sabut. Both C. concolor and D. sabut respond to disturbance by more vigorous growth. If the two ITCI groups are combined, the number of species that all sites have in common is six with the addition of Calamus javensis, C. laevigatus and C. ornatus (Table 2). Both Calamus laevigatus and C. ornatus are species that are often encountered in primary forest as robust plants in the rosette stage. Calamus javensis is actually a complex polymorphic species that occurs from almost swampy valley floors to moist depressions on slopes.

The fact that 8 out of 18 Apo Kayan, and 6 out of 24 ITCI species were not found in the other research sites further emphasises the great regional variation in species composition.

Table 6. Presence of rattan species in four permanent plots in primary forest in the Apo Kayan

(in 1994)

Number of plants ha^-1 (a), number of plants with mature canes ha^-1 (b) and total number of mature canes ha^-1(c)

Plots arranged in a topo-sequence from ridge crest to valley floor

Comparison of species composition and abundance

Species composition

The Calamus pogonocanthus complex represented by three taxa also illustrates regional variation. C. pogonocanthus 1 is restricted to primary forest in the Apo Kayan. C. pogonocanthus 2 is restricted to plot Matthijs and other parts of the Wanariset research forest. C. pogonocanthus 3 (the taxon that most resembles C. pogonocanthus Becc. ex H. Winkl.) occurs in primary as well as disturbed forest in all sites (at Apo Kayan and Wanariset outside permanent plots), with a higher abundance in disturbed forest.

As regards the (locally) more common species, some general comments on ecology can be given. The species that are almost exclusively encountered in logged-over/old secondary forest are Calamus caesius, C. pandanosmus, C. scipionum and Daemonorops fissa. A number of other species which are also present in primary forest show a clear positive response (in numbers of mature canes) to disturbance namely Calamus pogonocanthus 3, Ceratolobus concolor, Daemonorops sabut, Korthalsia echinometra, K. furtadoana, K. rigida and Plectocomiopsis geminiflora. Certain species are restricted to primary forest, some preferring dry sites like Calamus conirostris, C. marginatus, C. muricatus, Daemonorops atra, D. hystrix and D. pumila, while others, like Calamus javensis, C. ornatus and C. tomentosus, prefer moist conditions.

For species encountered on only a few occasions, it was difficult to define their preferences i.e.: Calamus hispidulus, C. mattanensis, Daemonorops korthalsii, Daemonorops cristata and Plectocomiopsis mira.

Calamus pilosellus although only found in two of the Apo Kayan plots and in plot Matthijs, locally formed dense entanglements in old canopy gaps on well-drained slopes in the Apo Kayan. Mature plants of Calamus blumei, C. laevigatus, C. tomentosus and Korthalsia ferox always occurred at very low densities.

The species richness in primary forest in the Apo Kayan plots and plot Matthijs is similar. Species richness, of a primary forest plot of comparable size, in the ITCI concession is lower. However, logging appears to have resulted in an increase in species richness in the ITCI plots (18 species on 18,900 m² in primary forest versus 20 species on 11,600 m² in logged-over forest) due to an influx of species adapted to disturbed sites.

Abundance

The abundance of rattan differs considerably between plots of the same site as well as between sites. This variation between plots may be attributed to the limited size of the plots. The size of the plots was not sufficient to comprise the variation in site conditions and successional stages, present in the area. Differences between research sites however point to a common trend.

The Apo Kayan plots have the highest abundance of rattan. This applies to the total number of plants, the total number of plants with mature canes, and the total number of mature canes (Table 7). Plot Matthijs is a good second with roughly half the number of plants and half the number of mature canes. The number of plants in the ITCI primary plots is only a tenth of the Apo Kayan value, and the total number of mature canes is a mere 4 % of the Apo Kayan value.

Calamus conirostris dominates the 'dry' Apo Kayan plots, where it accounts for over 50 % of all plants. Even when C. conirostris is excluded, the Apo Kayan plots still have the highest abundance of plants with mature canes and total number of mature canes.

Although the abundance in primary forest of the Apo Kayan and plot Matthijs is higher than the ITCI plots, the logging has (probably) resulted in an increase in abundance. However, it still needs to be emphasised that local differences in abundance are substantial, even in similar forest types.

Table 7. Rattan abundance in East Kalimantan (in numbers ha^-1)