RSS | Register/註冊 | Log in/登入
Site search:
Home>FFTC Document Database>Extension Bulletins>Sustainable Land Management in Tropical Tree-Crop Ecosystems
Sustainable Land Management in Tropical Tree-Crop Ecosystems
Ghulam M. Hashim
Malaysian Agricultural Research and Development Institute
P.O. Box 12301, 50774 Kuala Lumpur
Malaysia, 1996-08-01


Soil erosion is a major process of land degradation. Effective soil erosion management is therefore a vital part of the quest for sustainable agricultural production. In tree-crop ecosystems in the humid tropics, the processes of soil erosion operating under the tree canopy are associated with the large-scale flow of surface water during high-intensity rainstorms. Research carried out in Malaysia has shown that a cocoa-Gliricidia ecosystem which maintains dense surface vegetation, together with circle weeding, satisfied many indicators of sustainability. In comparison with an adjacent clean-weeded area, soil loss associated with the improved practice was extremely low, peak runoff rate was low, runoff was delayed, loss of nutrients was reduced, and plant growth enhanced. In many tropical tree-crop ecosystems, leaf litterfall returns nutrients to the soil. However, under slopeland conditions during wet spells, there was rapid mobilization of leaf litter. This is an important process of nutrient displacement. Living surface vegetation effectively controlled this process.

Abstracts in Other Languages: 中文(1040), 日本語(1071), 한국어(1199)


Concern about the extent and rate of land degradation has often been expressed. There are many forms of land degradation, ranging from excessive accumulation of chemicals to sedimentation in lowlands and in marine environments. Several forms of land degradation have been listed by E1-Swaify (1994). Of these, soil erosion by water accounted for 55% of total degraded land (Oldeman 1994). Ayoub (1994) found that in Africa at present, erosion is the main form of degradation, having increased 20 times in the last 30 years. Soil losses of 200 to 300 mt/ha/year are common. Similarly, Sehgal and Abrol (1994) reported erosion by water to be the major form of land degradation in India.

The cost of soil erosion is very high. The financial loss arising from the shortfall of achievable tea yields due to soil erosion in Sri Lanka is in the range of US$60 - 80 million per annum. The estimated cost of replacing nutrients removed by soil erosion is more than US$20 million per annum (Somasiri 1994).

In view of the probable doubling of the world's population in the next century, steps to arrest land degradation are needed. Efforts to achieve sustainable production should be encouraged. Population pressure and conflicting demands on land exclude the possibility of relying on any increase in cultivated area to meet future food needs. Fortunately, mankind's experience in the recent past has shown that various science-based inputs can help to increase yields per unit area of cultivated land (Lal 1994). Some of these innovations are improved cultivars, better nutrient recycling mechanisms, judicious use of fertilizers, and effective soil conservation measures. Surface cover management practices that use natural vegetation to prevent large-scale erosion and yet do not limit plant growth hold the promise of contributing to sustainability. This paper discusses cover management practices in tree-crop cultivation in a tropical slopeland area. The various parameters used to assess sustainability associated with an improved cover management practice are discussed. It is hoped that an understanding of the factors influencing soil movement in relation to different forms of soil cover will help towards the adoption of sustainable practices.

Sustainable Agriculture

Sustainable agriculture is a complex concept, although many attempts have been made to define it. Some of these definitions vary widely (see e.g. Wallace 1994). However, there are a number of common elements, including

  • The efficient use of inputs;
  • The enhancement of environmental quality;
  • The maintenance of the natural resource base;
  • An adequate supply of human food and fibre needs;
  • Enhancement of the quality of life;
  • The assurance of profitability.

The present major concern with respect to sustainability is whether the world can continue to produce food and fiber for its growing population, while preserving the quality of its natural resources. To achieve these objectives, agriculture must emphasize improving plant growth and performance, and maintaining low levels of soil erosion. Wallace and Wallace (1994) stated that "agricultural sustainability comes down mostly to soil and water conservation".

Soil Erosion Control in Tree-Crop Ecosystems

In many parts of the humid tropics, vast areas of slopelands are planted in tree crops such as rubber ( Hevea brasiliensis), oil palm ( Elieases guiniensis), cocoa ( Theobroma cocoa), tea ( Camellia sinensis) and coffee ( Coffea sp.). Important features of these species are their perennial nature, which excludes the need for frequent cultivation, the presence of a canopy of leaves, and their production of leaf litter. Leaf litter accumulates as a result of periodic litterfall, providing a form of surface cover over the soil. However, during large storm events the flow of water over the surface soil means that soil and water conservation measures are necessary. For example, in rubber and oil palm holdings, leguminous cover crops and terracing help to reduce soil erosion while the trees are still immature (Soong et al. 1980). In Sri Lanka, tea is planted in the "high country", where steep slopes are common. To achieve sustainability, a number of biological, chemical and physical measures are practiced simultaneously. This integrated approach ensures that soil fertility and structure are restored and soil loss reduced (Sivapalan 1994). Under mature oil palms in Malaysia, soil loss is minimized by the practice of placing pruned fronds in the inter-row areas (Maene et al. 1980).

A Case Study from Malaysia

A slopeland area in Kemaman, Trengganu, in eastern Peninsular Malaysia was planted in cocoa. Since cocoa is a tree that requires shade (Wessel 1985), Gliricidia maculata had been established a year earlier to provide shade. Detailed measurements of soil erosion, hydrology and plant growth with respect to two contrasting cover management practices were taken. The experiment was carried out in an area with an average slope of 17.6%. The soils, identified as Plinthic Orthoxic Tropudult, are derived from iron-poor shales. They are acidic and infertile, containing about 20% clay.

Four plots, each 0.1 ha in area, were laid out. Each plot extended to the top of a ridge, and had at its lower end a 50 m long concrete channel, which emptied into a 2 m x 2 m x 0.5 m stilling basin. The height of water flow through a Parshall flume during each runoff event was continuously recorded by a mechanical height recorder. The sediments which were deposited in the stilling basin were weighed, and corrected for moisture content. Proportional samples of the suspended load were obtained via a slotted copper tube fixed vertically to the wall of the stilling basin. The extent of surface contact cover, (i.e. vegetation and other material on the ground which is close enough to the soil surface to interfere with overland water flow) was estimated periodically. The girth of about one-third of all cocoa trees, a randomly chosen sample, were measured monthly at 7.6 cm above the base ( Table 1(1317)). Other details of methodology are given in Hashim et al. (1995).

Although two basic cover management practices, i.e. clean weeding and the maintenance of live surface vegetation, were studied, each of the four plots had a different type of ground cover ( Table 1(1317)). As the aim of this discussion is to compare only two ways of managing surface cover, the results from plots T1 and T2 alone (intercropping with banana), are presented. Overall results which include plots T3 and T4 (monocropping of cocoa) have been discussed in Hashim (1995) and Hashim et al. (1995).

Reduced Soil Loss As a Result of Improved Cover Management

Annual soil loss and runoff data are shown in Table 2(1232). It is apparent that throughout the period under review, there are three distinct phases of soil loss management. In water year 1989 - 90, the difference in soil loss between T1 and T2 (Phase 1) was small. Phase 2 is the period from 1990 - 91 to 1992 - 93, when soil loss differences increased markedly. In Phase 3, 1993 - 94, the difference in the rate of soil loss between different treatments decreased again.

During Phase 1, the surface contact cover of grass regrowth in T2 was similar to that in T1. The difference in soil loss and runoff reflects the difference in surface contours; T1 had a higher density of preferred flow pathways. From 1990 - 91 onwards (Phases 2 and 3) there were distinct differences in ground cover, as T1 was subjected to clean weeding with chemicals, and T2 became covered with dense surface vegetation, (although in T2, circles 0.5 m in radius around each tree were kept free of weeds). The dense ground cover helped to keep soil loss in T2 relatively low during Phase 2. During the same period, the ground cover in T1 gradually increased due to the accumulation of leaf litter. From mid-1993 onwards, the percent contact cover of T1 was in the range of 70 to 80%. However, during wet periods when large storms occurred, the overland flow mobilized the leaf litter, and reduced contact cover in the major preferred flow pathways to <5%. Thus, soil loss was extremely high even though the contact ground cover in T1 was almost as high as in T2. This shows that to achieve low levels of soil loss under these conditions, contact cover should be anchored, and should be well distributed over major pathways where erosion processes are particularly active.

During Phase 3, the rate of soil loss in T2 was uncharacteristically large. This was mainly because of careless management in the latter part of 1993, so that the weeded circles became too large and merged, forming "weeded rows". This had the effect of concentrating overland flow, resulting in very high soil losses ( Table 3(1182)). The importance of an even distribution of contact cover must be emphasized.

A Reduction in the Mobility of Leaf Litter

In tropical tree-crop ecocystems, such as rubber plantations or a combination of cocoa and Gliricidia, leaf litter plays an important role in nutrient recycling. Ling (1984) reported that under a mature cocoa- Gliricidia stand, the leaf litter from the cocoa trees returned 75 - 94 kg nitrogen/ha/year, 4 - 5 kg phosphorous/ha/year, 84 - 100 kg potassium/ha/year, 28 - 34 kg magnesium/ha/year and 58 - 78 kg calcium/ha/year. Under mature rubber trees, the amount of litterfall is in the range 4620 to 5320 kg/ha/year (Moris 1993).

On slopelands, leaf litter is mobilized by surface water flow during large storm events (Ghulam M. Hashim, unpublished data). Living surface vegetation slows down the surface flow and induces deposition, helping to retain leaf litter on the soil surface. The decomposition of this litter provides organic matter and nutrients to the soil.

Reduced Runoff Associated with Improved Cover Management

Runoff in T2 was low relative to that in T1 ( Table 2(1232)). Its cover of living vegetation acted as a barrier to surface water flow, and facilitated infiltration. In addition, it also delayed the onset of runoff. This phenomenon is illustrated by pairs of hydrographs (time vs. flow height graphs) for four events in 1992 ( Fig. 1(1132)). In A, B and C ( Fig. 1(1132)), the size of T2 hydrographs were much smaller than those of T1. However, in relatively large rainstorms such as that of November 17, 1992 ( Fig. 1(1132) D), the differences between T1 and T2 were not as marked, showing that, under such conditions, the effectiveness of live vegetation in controlling runoff was reduced (Hashim 1995).

Runoff rate has an important effect on soil erosion. It determines the speed at which detached soil particles are transported away from the site of detachment. An increase in runoff rate increases "streampower" (Rose 1984). When streampower exceeds a certain threshold value, sediment is removed from the soil surface, increasing the sediment concentration of surface water flow (Rose 1984). High runoff rates are an indication that processes of soil removal and deposition are active.

Peak runoff rates associated with the hydrographs shown in Fig. 1(1132) are given in Table 4(1063). Those in T2 are much lower than those in T1, except for the rainstorm on November 17. Corresponding soil losses show a similar trend: differences in soil loss between T1 and T2 were marked in the first three examples, although relatively small on November 17.

Improvement in Plant Growth Associated with Improved Cover Management

Monitoring of plant growth and performance involved monthly measurements of tree girth, while the number of cocoa trees bearing fruit was also recorded ( Fig. 2(1334) Table 5(1093)). Later, the number of fruit was counted.

Plant growth was initially faster in T1 than in the other plots, although gradual increases in average tree girth were observed in both plots. This was probably due to better initial soil conditions. The advantage continued over time, and it was not until after the 17th month that more pronounced increases in average girth in T2 were observed ( Table 6(1159)). The trend in fruit production, as indicated by the number of plants bearing fruit ( Table 5(1093)), also suggested that there was a gradual improvement in plant performance in T2.

The maintenance of living surface vegetation in T2 over a considerable period of time minimized soil loss and ensured the retention of a large part of the topsoil. In addition, more leaf litter, an important source of nutrients (Ling 1986), was retained, as living vegetation reduced its mobility. This is reflected in the nutrient content of the surface soil. Table 7(986) shows that the average nitrogen and potassium contents in the period July 1992 to November 1993 were higher in T2 than in T1.

Thus, although during the early part of the experiment, the growth rate of trees in T2 was lower than in T1 ( Table 6(1159)), improved soil conditions resulted in an improvement in the growth rate of the trees in T2 so that it approached that of T1 after the 17th month.


Identifying and understanding the erosion processes operating on the soil surface are a prerequistite for designing measures for sustainable slopeland agriculture. Measuring parameters such as the amount and rate of runoff, and the rate of soil loss and plant growth, help in identifying and understanding these processes. To achieve a high level of sustainability, measures should be designed to reduce the velocity of surface water flow, and to provide contact cover that is both anchored to the ground and well-distributed over the ground surface.

The maintenance of a dense surface cover of living vegetation, with circle weeding around individual cocoa trees, was associated with low and delayed runoff, and low soil loss. This helped to retain topsoil and leaf litter, which in turn contributed to an improvement in the nutrient content of the soil and its organic matter status. Plant growth gradually improved, indicating the attainment of a relatively high level of sustainability.


  • Ayoub, A.T. 1994. Human-induced soil degradation in Africa. Paper presented at the 8th International Soil Conservation Conference. December 4-8, New Delhi, India. (Unpublished mimeo.).
  • Ciesiolka, C. A., K.J. Coughlan, C.W. Rose, M.C. Escalante, G.M. Hashim, E.P. Paningbatan, Jr. and S. Sombatpanit. 1995. Methodology for a multi-country study of soil erosion management. Soil Technology. 8,3: 179-192.
  • E1-Swaify, S.A. 1994. Problems of land degradation in humid and sub-humid regions. Paper presented at the 8th International Soil Conservation Conference, December 4-8, New Delhi, India. (Unpublished mimeo.).
  • Hashim, Ghulam M. 1995. Important considerations in the management of soil erosion. The Planter (Kuala Lumpur) 71, 826: 5-17.
  • Hashim, Ghulam M., Cyril Ciesiolka, Wan Abudullah Yusoff, A. Wahab Nafis and M. Zain Mustapha. 1990. Methodology and instrumentation for the measurements of rates of runoff and erosion. In: Developments in Soil Research in Malaysia, (M.Y. Khanif, S.R. Syed Omar and J. Shamshuddin (eds.). Malaysian Society of Soil Science, pp. 173-182.
  • Hashim, Ghulam M., Cyril A.A. Ciesiolka, W. Abdullah Yusoff, A. Wahab Nafis, M. Radzali Mispan, Keppel J. Coughlan and Calvin W. Rose. 1995. Soil erosion processes on sloping land in the east coast of Peninsular Malaysia. Soil Technology. 8,3: 215-233.
  • Lal, R. 1994. Soil degradation and agri-cultural sustainability. Paper presented at the 8th International Soil Conservation Conference. December 4-8, 1994, New Delhi, India. (Unpublished mimeo).
  • Ling, A.H. 1986. Litter production and nutrient cycling in a mature cocoa plantation on inland soils of Peninsular Malaysia. In: Cocoa and Coconuts: Progress and Outlook, E. Pushparajah and P.S. Chew, (eds.). Incorporated Society of Planters, Kuala Lumpur, Malaysia, pp. 541-466.
  • Maene, L.M., K.C. Thong, T.S. Ong and A.M. Mokhtaruddin. 1979. Surface wash under mature oil palms. In: Proceedings of the Symposium on Water in Agriculture. Malaysian Society of Soil Science, pp. 203-216.
  • Moris, N. 1993. Nutrient returns via Hevea leaf litterfall. In: Soil Science Conference of Malaysia 1993. Penang, Malaysia, April 19-21, 1993, Malaysian Society of Soil Science.
  • Oldeman, L.R. 1994. An international methodology for a world soils and terrain database and a global assessment of soil degradation. In: The Collection and Analysis of Land Degradation Data. Regional Office for Asia and the Pacific, FAO Bangkok, pp. 35-60.
  • Rose, C.W. 1984. Progress in research in soil erosion processes and a basis for soil conservation practices. In: Soil Erosion Management, (E.T. Craswell, J.V. Remenyi and L.G. Nallana (eds.). ACIAR Proceedings, Series No. 6, pp. 32-41.
  • Sehgal, J. and I. P. Abrol. 1994. Soil degradation in India: Assessment and impact. Paper presented at the 8th International Soil Conservation Conference, New Delhi, India December 4-8, 1994. (Unpublished mimeo).
  • Sivapalan, P. 1994. Strategies for erosion control in tea plantations. Paper presented at the 8th International Soil Conservation Conference, New Delhi, India December 4-8, 1994. (Unpublished mimeo).
  • Somasiri, S. 1994. Land degradation: Causes and impacts. Paper presented at the 8th International Soil Conservation Conference, New Delhi, December 4-8, 1994. (Unpublished mimeo).
  • Soong, N.K., G. Haridas, C.S. Yeoh, P.H. Tan. 1980. Soil Erosion and Conservation in Peninsular Malaysia. Rubber Research Institute of Malaysia, Kuala Lumpur. 69 pp.
  • Wallace, Arthur. 1994. Sense with sustainable agriculture. Communications in Soil Science and Plant Analysis 25, 1&2: 5-13.
  • Wallace, Authur and Garn A. Wallace. 1994. Can we save this fragile earth? Introduction: Environmental challenges related to soil. Communications in Soil Science and Plant Analysis 25, 1 & 2: 1-4.
  • Wessel, M. 1985. Shade and nutrition. In Cocoa, G.A.R. Wood and R.A. Lass (eds.). 4th Ed. Tropical Agriculture Series, Longman, London, U.K., pp. 166-194.


One participant commented that the annual rainfall of Malaysia is about 3500 mm, and asked about the level of rainfall intensity during the rainy season, and in particular, how much rain falls in one hour.

Dr. Hashim replied that Malaysia's rainfall is not evenly distributed: 40% of a typical year's rain falls in November and December, and during these two months, rainstorms can be very heavy. Rainfall of up to 150mm in hour has been experienced. Much of the rain falls in "wet spells", i.e. periods of highly intensive rain occurring on two or three consecutive days. In any one year, there are likely to be five or six wet spells, and it is these which contribute most to erosion. Dr. Hashim estimated that about 70% of soil loss took place at such times.

It was noted that although circle weeding gave some good results, such as a marked improvement in the level of exchangeable P in the soil, in other cases the difference was not significant, as in the case of nitrogen and potassium levels. Dr. Hashim agreed that differences in some cases were not very marked, but felt that the improved practice of circle weeding does mean that soil has a higher nutrient content. The important point is that the practice gives some improvement in return for only a minimal investment of time and energy.

Index of Images

Table 1 Physical and Biological Features of the Plots

Table 1 Physical and Biological Features of the Plots

Figure 1 T1 and T2 Hydrographs Showing Runoff Differences during Four Events

Figure 1 T1 and T2 Hydrographs Showing Runoff Differences during FourEventsFigure 2 Growth Rate of Cocoa Plants

Figure 2 Growth Rate of Cocoa Plants

Table 2 Summary of Soil Loss and Runoff

Table 2 Summary of Soil Loss and Runoff

Table 3 Monthly Soil Loss in the Latter Half of 1993 Showing Uncharacteristic Increases in Soil Loss in T2 from October 1994

Table 3 Monthly Soil Loss in the Latter Half of 1993 Showing Uncharacteristic Increases in Soil Loss in T2 from October 1994Table 4 Peak Runoff Rates and Soil Loss Associated with Hydrographs Shown in Figure 1

Table 4 Peak Runoff Rates and Soil Loss Associated with Hydrographs Shown in Figure 1

Table 5 Cocoa Performance According to Surface Cover Treatment

Table 5 Cocoa Performance According to Surface Cover TreatmentTable 6 Average Rate of Increase in Girth of Cocoa Trees

Table 6 Average Rate of Increase in Girth of Cocoa Trees

Table 7 Nitrogen and Exchangeable Potassium Contents of the Topsoil (Top 2 CM) in Pathways (P) and Interpathway Areas (I). (Means of 9 Sampling Periods between July 1992 and November 1993)

Table 7 Nitrogen and Exchangeable Potassium Contents of the Topsoil (Top 2 CM) in Pathways (P) and Interpathway Areas (I). (Means of 9 Sampling Periods between July 1992 and November 1993)

Download the PDF. of this document(942), 100,307 bytes (98.0 KB).