Early successional process of a fallow forest in East Kalimantan, Indonesia

Transcription

1 TROPICS Vol. () Issued ay, Early successional process of a fallow forest in East Kalimantan, Indonesia Toru ASIOTO 1,, Takeshi TANGE 2, asaya ASUORI 2, isayosi YAGI 2 and Katsumi KOIA 3 Nutrient Dynamics Laboratory, orestry and orest Products Research Institute. atsunosato, Tsukuba, Ibaraki, apan Graduate School of Agricultural and Life Sciences, The University of Tokyo, Tokyo, apan Asian Natural Environmental Science Center, The University of Tokyo, Tokyo, apan Corresponding author: Toru ASIOTO. Tel: +, ulin@ffpri.affrc.go.jp ASTRACTWe examined the early successional process in a fallow forest following slash-andburn agriculture in East Kalimantan, Indonesia, to determine the ecological response of tropical secondary forest to human disturbance. Community dynamics and growth of the vegetation were studied during the initial 2 months of succession following the abandonment of slash-and-burn agriculture. The vegetation grew rapidly, and 2 months after the abandonment, plants with diameters 1 cm at breast height had a community density (number of stems) of 4.1 m 2 and a basal area of 1.2 cm 2 m 2. Populations of some dominant species began to decrease, and replacement of dominant species occurred within 2 months after the abandonment. Weeding of seedlings and sprouts during cultivation affected the composition and growth of the fallow vegetation. The results showed that fallow forests in the tropics have high recuperative potential. Key words: orneo, Southeast Asia, pioneer tree, secondar y succession, shifting cultivation, slashand-burn agriculture INTRODUCTION Tropical secondary forests have been increasing as a result of the expansion of human activities. rown & Lugo () reported that the area covered by secondary forests exceeds million ha in the tropics and accounts for about % of the total area of tropical closed forest. Tropical secondar y forests have many ecological characteristics that make them valuable and useful to people (rown & Lugo, ). They often produce timber as well as firewood and charcoal materials (inegan, ), and they stabilize the environment through soil conser vation and the sequestration of atmospheric carbon (earnside & Guimarães, ; ashimoto et al. ). In addition, secondary forests play important roles to conserve biodiversity as refuges or green corridors that link isolated virgin forests for use by wildlife in agricultural landscapes (DeWalt et al. ). If we can accelerate the process of secondary succession and restore degraded secondary forests, we will be able to increase these values for secondary forests and utilize them more effectively; hopefully, this will then reduce human pressures on the remaining virgin forests. Although the study of secondary forests in the Neotropics has progressed (Guariguata & Ostertag, ; Kammesheidt, ; yster, ), there have thus far been few studies of these forests in Southeast Asia. Various forms of human disturbance create secondary forests, and slash-and-burn agriculture is one of the main ones, especially in Southeast Asia (AO, ). Slash-and-burn agriculture used. ha of land worldwideabout % of the world, s exploitable soilsand produced food for more than people (auck, ). It is important to accumulate fundamental knowledge of the ecological processes in the fallow forests that regenerate after farmers move on to new sites to elucidate their mechanisms and to manage fallow forests. In the present paper, we describe the early successional process that occurs in fallow forests after the abandonment of slash-and-burn agriculture in East Kalimantan, Indonesia. STUDY AREA The study was carried out at Sebulu in East Kalimantan, in the Indonesian part of orneo. The study site ( S, E, about m above sea level) is located on a low undulating plateau and includes some small rivers and shallow basins. This topography is common in the lowlands of East Kalimantan. The soil around the study area is a red acid soil that is common in the Southeast Asian tropics.

2 Toru ASIOTO, Takeshi TANGE, asaya ASUORI, isayosi YAGI and Katsumi KOIA Daily mean temperature changes little throughout the year. The yearly mean temperature at alikpapan ( S, E), which is located about km southwest of the study site, is. C (National Astronomical Observatory of apan, ), and that at Sepinggan ( S, E), which is close to alikpapan, is. C (Ohta et al. ). The yearly average precipitation at Sebulu from to was mm (Soda, ). In, the monthly precipitation from ay to October was < mm, which is lower than precipitation in the previous years, resulting in a total annual precipitation of only mm. This period of relative drought was related to a pronounced El Niño Southern Oscillation (ENSO) event in (Toma et al. ). The natural forest in this area was dominated by species in the family Dipterocarpaceae (Sukardjo et al. ). ecause extensive commercial logging operations have been encouraged since in Indonesia (Kartawinata et al. ), the area of natural forest has gradually been decreasing in this region. In addition, a large-scale wildfire occurred in the study area during (alingreau et al. ) and most of the natural forest was lost to the fire. After the fire, illegal logging and shifting cultivation spread widely throughout the burnt area. The deforested lands then changed into either secondary forests of pioneer species or into alangalang (Imperata cylindrica (L.) P. eauv.) grasslands (ashimoto et al. ). ETODS Slash-and-burn agriculture In this area, slash-and-burn agriculture was generally conducted as follows: in September or thereabouts, farmers slash all trees over an area of one to several hectares, burn the slashed trees to enrich the soil with ash (after allowing the wood to dr y for about to weeks), and then gather the remaining chips of unburned wood and burn them again. Where upland rice cultivation is practiced, they dig holes in the ground and sow the rice seeds. armers weed once or twice in the months or so before the rice harvest. inally, they harvest the rice ears. Some farmers plant vegetables or fruit trees after the rice harvest, but we rarely heard about or saw repeated cultivation of rice at the same site during our field study. We asked some farmers to carry out rice cultivation using methods similar to that described above. To examine the results of this cultural practice, we established a m plot in September. We set up a -m-wide buffer zone along the inner edges of the plot and divided the remaining core zone of m into a rice cultivation area (RCA) of m and a noncultivation area (NCA) of m. In September, we felled all vegetation in the plot to a height of about cm from the ground and then dried and burned it. The chips of unburned wood remaining after the first burn were then burned. Rice seeds were sown in the RCA in October. All seedlings and sprouts of competing vegetation in the RCA were weeded completely once, in December. The rice was harvested in April. After the rice harvest, no further treatments were applied. Data collection About months after the harvest (une ) we set up quadrats, each m (Nos. ), in the RCA and two quadrats of the same size (A and ) in the NCA. In each quadrat, we attached numbered tapes to every plant whose stem had a diameter at breast height (D) cm. We recorded the species name, D, and the length of each stem on uly and November, and on anuary, une, and December. RESULTS AND DISCUSSION Vegetation development Changes in the mean lengths of the longest stems in each quadrat are shown in igure. ean stem length ranged from about. to. m by months after rice harvesting in the RCA, versus. and. m in the NCA. The rate of stem elongation in each quadrat decreased after une. The mean lengths of the longest stems in the two NCA quadrats were higher than those in the RCA quadrats months after harvest, but this difference had disappeared by months after harvest. The total community density (number of stems m ) in the RCA quadrats increased rapidly after the abandonment of farming, and by years after weeding (December ) it had reached an average of. m (ig. ). Community density increased, but the rate of increase slowed in the last months of the study period. Drought conditions from ay to December may have caused this reduction. The changes in community density in the NCA differed from those in the RCA: the density increased more rapidly in the NCA than in the RCA for year after burning, but thereafter the difference between the two areas decreased (ig. ). Self-thinning was thought to have been the cause of this decrease in the NCA from year after burning. We considered that the faster increase in community density in the NCA than in the RCA was

3 Early successional process of a fallow forest in East Kalimantan, Indonesia Stem length (m) Slash-and-burn Weeding arvesting Ç É Ç É Ç É Ç É Quadrat No. Rice cultivation area É 6 Ç Non-cultivation area A onth ig. 1. Changes in the mean lengths of the 1 longest stems in each quadrat. Community density (No. of stems m -2 ) Slash-and-burn Weeding É Ç arvesting É Ç É É ÉÇ É Ç Ç Ç onth Quadrat No. Rice cultivation area É 6 Ç Non-cultivation area A ig. 2. Changes in the community densities in each quadrat.

4 Toru ASIOTO, Takeshi TANGE, asaya ASUORI, isayosi YAGI and Katsumi KOIA caused by differences in the time when recovery began and in the frequency of disturbance. The vegetation in the NCA was only subjected to the October slash-andburn, whereas that in the RCA underwent not only slashand-burn but also weeding (in December ). The community density in quadrat A was the highest among the quadrats in anuary (ig. ), and the dominant species there were icus sp. and Piper aduncum L. (data not shown). These species can sprout from cut stumps. On average, there were fewer sprouting stumps of icus sp. and P. aduncum in the RCA than in the NCA (Table ). These results suggest that the difference in the rate of population increase between the two areas was caused by weeding, which removed sprouts and seedlings from the RCA. Comparison of the population densities of woody plants cm tall months after slash-andburn only (Uhl et al. ) and after slash-and-burn plus cropping (Uhl et al. ) revealed that the latter was lower. Uhl et al. () pointed out that repeated weeding had apparently exhausted the sprouting reser ves of many of the cut forest trees. Sprouts play important roles in forest restoration during the initial successional stages (Kammesheidt, ). Our results also suggest that weeding, as an anthropogenic modification, affected the successional process, because it appeared to lead to a decrease in the number of sprouting stumps and an inhibition of population increases in the fallow forest. In quadrat, Trema tomentosa (Roxb.) ara dominated. This difference in the dominant species between quadrats A and was caused by stochastic factors which are pronounced in small quadrats, and also by environmental factors that were related to the sampling position in the NCA. Whether or not sprouting species dominated a plot accounted for the difference in community density between quadrats A and. Changes in species composition The changes in the proportions of population density and basal area accounted for by the dominant species in the RCA (all quadrats combined) are shown in igure. All species with D cm were woody plants, with the exception of lumea balsamifera (L.) DC., a perennial herb. The population densities of icus sp., Geunsia pentandra (Roxb.) err., and P. aduncum continued to increase during the study period, but the rate of increase in the densities of icus sp. fell after une (ig. A). The population densities of omalanthus populneus (Geisel.) Pax and T. tomentosa began to decrease during the study period after an initial increase, and the proportion of the population accounted for by. populneus decreased conspicuously. The proportion of the population accounted for by. balsamifera also began to decrease after an initial increase. The proportion of elastoma malabathricum L. increased rapidly after une, and this may have resulted from slower initial growth or subsequent recruitment of seedlings of this species. The basal area (cm m ) of the RCA was. (cm m ) years after weeding (ig. ). The proportions of total basal area accounted for by. populneus, T. tomentosa, and. balsamifera all decreased during the study period after initial increases similar to those in the proportions of total population density. These results suggest that the initially dominant species were replaced Table 1. ean numbers per m 2 of sprouting stumps whose branch 1 cm D and stems 1 cm D grown from seeds. Area Sp. (m ) ul. Nov. an. un. Rice cultivation area icus sp. Sprouting stumps (with weeding) Stems grown from seeds Piper aduncum Sprouting stumps Stems grown from seeds Quadrat A A A A Non-cultivation area icus sp. Sprouting stumps (without weeding) Stems grown from seeds Piper aduncum Sprouting stumps The numbers in Rice cultivation area indicate mean SD of quadrats. The numbers in Non-cultivation area indicate data of each quadrat A and. Stems grown from seeds

5 Early successional process of a fallow forest in East Kalimantan, Indonesia Proportion of population (% ) A allow forest Others lumea balsamifera elastoma malabathricum Proportion of basal area (%) onth allow forest Piper aduncum Geunsia pentandra icus sp. Trema tomentosa omalanthus populneus ig. 3. Changes in the proportions of population density (A) and basal area () accounted for by the dominant species in the rice cultivation area (all 1 quadrats combined). Right bar on the each panel shows the proportions of population and basal area accounted for by dominant species in a fallow forest about years old at the same study site from ashimoto et al. (), respectively. Numbers on the bar indicate community densities (number of stems m ; A) and basal areas (cm m ; ) at each measurement. over the short period of the study. In, ashimoto et al. () recorded the floristic compositions of plots in this area, including the same site used in the present study, and showed that. populneus, T. tomentosa, and. balsamifera dominated only in secondary forests to years old. At the time of that study, the sites were covered by a - to -year-old secondary forest that had previously been disturbed (ig. ; fallow forest). omalanthus populneus, T. tomentosa, and. balsamifera were scarce in the fallow forest, whereas P. aduncum, G. pentandra, and icus sp. were dominant. Populations of the former three species may therefore become established during the initial stages of secondary forest succession and then disappear. omalanthus populneus,. malabathricum, Trema spp., acaranga spp., and allotus spp. seem to be common pioneer species in East Kalimantan, as they appeared not only at our present study sites but also in

6 Toru ASIOTO, Takeshi TANGE, asaya ASUORI, isayosi YAGI and Katsumi KOIA other secondary forests. Kiyono () reported that small trees such as allotus paniculatus (Lam.) üll. Arg., acaranga tanarius (L.) üll. Arg.,. populneus, and P. aduncum entered the succession and formed rather pure colonies of each species after the first slashand-burn cropping around ukit Soehar to in East Kalimantan. Tagawa et al. () listed Anthocephalus chinensis (Lam.) A. Rich. ex Walp., Glochidion capitatum.. Sm., acaranga trichocarpa (Rchb. f. & Zoll.) üll. Arg., acaranga gigantea (Rchb. f. & Zoll.) üll. Arg., Croton argyratus lume, Tristania whitiana Grif f., allotus spp., and. populneus as woody pioneer species that form monodominant stands along the roadsides around Kutai National Park in East Kalimantan. Watanabe et al. () similarly listed Vitex pubescens Vahl, Trema cannabina Lour.,. trichocarpa,. gigantea, A. chinensis,. macrophyllus (the genus name was not provided by these authors), Cratoxylum sp.,. populneus,. malabathricum, and. balsamifera. These similarities of species composition imply that these species are very common in secondary forests in the Kalimantan area. In conclusion, we found that succession of woody dominant species occurred during the first years after abandonment of slash-and-burn agriculture. This fast successional advance appears to be characteristic of this area. ACKNOWLEDGENTWe thank the Sumitomo orestry Co., Ltd. for their support and their maintenance of the research site. REERENCES rown, S. & Lugo, A.E.. Tropical secondary forests. ournal of Tropical Ecology, 6:. DeWalt, S.., aliakal, S.K. & Denslow,.S.. Changes in vegetation structure and composition along a tropical forest chronosequence: implications for wildlife. orest Ecology and anagement, 182:. AO. Global issues. In: Global orest Resources Assessment (orestry Paper )(ed. ood and Agriculture Organization), pp., AO, Rome. earnside, P.. & Guimarães, W... Carbon uptake by secondary forests in razilian Amazonia. orest Ecology and anagement, 8:. inegan,.. The management potential of neotropical secondary lowland rain forest. orest Ecology and anagement, 4:. Guariguata,.R. & Oster tag, R.. Neotropical secondary forest succession: changes in structural and functional characteristics. orest Ecology and anagement, 148:. ashimoto, T., Kojima, K., Tange, T. & Sasaki, S.. Changes in carbon storage in fallow forests in the tropical lowlands of orneo. orest Ecology and anagement, 126:. auck,.w.. Introduction. In: Shifting Cultivation and Soil Conservation in Africa: Papers Presented at the AO/SIDA/ARCN Regional Seminar eld at Ibadan, Nigeria, uly, (ed. ood and Agriculture Organization), pp., AO, Rome. Kammesheidt, L.. The role of tree sprouts in the restoration of stand structure and species diversity in tropical moist forest after slash-and-bur n agriculture in eastern Paraguay. Plant Ecology, 139:. Kammesheidt, L.. Perspectives on secondary forest management in tropical humid lowland America. Ambio, 31:. Kartawinata, K., Adisoemarto, S., Riswan, S. & Vayda, A.P.. The impact of man on a tropical forest in Indonesia. Ambio, 1:. Kiyono, Y.. Degeneration of tropical rain forests of the East Kalimantan by man, s impactforests in areas of swidden agriculture. The Tropical orestry, 29:. (in apanese) National Astronomical Obser vator y of apan.. Chronological Scientific Tables. pp. aruzen, Tokyo. (in apanese) alingreau,.p., Stephens, G. & ellows, L.. Remote sensing of forest fires: Kalimantan and Nor th orneo in. Ambio, 14:. yster, R.W.. Post-agricultural invasion, establishment, and growth of neotropical trees. The otanical Review, :. Ohta, S., Ef fendi, S., Tanaka, N. & iura, S.. Characteristics of major soils under lowland dipterocarp forest in East Kalimantan, Indonesia, Tropical Rain orest Research Project TA-(a)-. pp. PUSREUT Special Publication No., ulawarman University, Samarinda, Indonesia. Soda, R.. Rainfall pattern and forest fires. In: Research Report on the Sebulu Experimental orest (eds. orestry Research and Development Agency, inistry of orestry, Republic of Indonesia; PT Kutai Timber Indonesia; Sumitomo orestry Co., Ltd. & Laboratory of Silviculture, The University of Tokyo), pp.. Sumitomo orestry Co., Ltd., Tokyo.

7 Early successional process of a fallow forest in East Kalimantan, Indonesia Sukardjo, S., agihara, A., Yamakura, T. & Ogawa,.. loristic composition of a tropical rain forest in Indonesian orneo. ulletin of Nagoya University orest, 1:. Tagawa,., Suzuki, E., Wirawan, N., iyagi, Y. & Oka, N.P.. Change of vegetation in Kutai national park, East Kalimantan. In: A Research on the Process of Earlier Recovery of Tropical Rain orest after a Large Scale ire in Kalimantan Timur, Indonesia (Occasional Paper No. ) (eds. Tagawa,. & Wirawan, N.), pp. -. Kagoshima University Research Center for the South Pacific, Kagoshima. Toma, T., arjenah & astaniah.. Climate in ukit Soeharto, East Kalimantan. In: Rainforest Ecosystems of East Kalimantan (eds. Guhardja, E., atawi,., Sutisna,., ori, T. & Ohta, S.), pp.. Springer-Verlag, Tokyo. Uhl, C., Clark,., Clark, K. & aquirino, P.. Successional patterns associated with slash-andburn agriculture in the upper Río Negro region of the Amazon basin. iotropica, 14:. Uhl, C., Clark, K., Clark,. & urphy, P.. Early plant succession after cutting and burning in the upper Río Negro region of the Amazon basin. ournal of Ecology, 69:. Watanabe, R., irabuki, Y. & Oka, N.P.. Note on the road-side shrub vegetation as the early stages of the tropical rain forest succession. In: Early Recovery Process of Kutai EcosystemA Preliminary Report (Kyoto University Overseas Research Report of Studies on Asian Non-uman Primates: Special Number) (eds. Azuma, S. & Wirawan, N.), pp.. Kyoto University Primate Research Institute, Kyoto. Received th uly Accepted th Nov.