Engineering properties of Leucaena leucocephala for prevention of slope failure

e c o l o g i c a l e n g i n e e r i n g 3 2 ( 2 0 0 8 ) 215–221 available at www.sciencedirect.com journal homepage: www.elsevier.com/locate/ecoleng Engineering properties of Leucaena leucocephala for prevention of slope failure O. Normaniza ∗ , H.A. Faisal, S.S. Barakbah Institute of Biological Sciences, Faculty of Science, University of Malaya, 50603 Kuala Lumpur, Malaysia a r t i c l e i n f o a b s t r a c t Article history: The potential of Leucaena leucocephala as an erosion control plant was investigated in terms Received 4 June 2007 of its capacity of root reinforcement, root profile and root shear strength. The species stud- Received in revised form ied exhibited extensive, dense rooting and depth of penetration. These aspects, to some 30 October 2007 extent, could provide surficial as well as deep-seated erosion control. A high water absorp- Accepted 5 November 2007 tion capacity would increase suction, thus potentially extracting water at the greater depth. The soil–root matrix significantly affects cohesion factor but not the angle of friction. The effect varies with increasing depth and age of plant depending on the root length density. Keywords: After 6 months of growth, the cohesion factor had tremendously increased by two- to five- Soil erosion fold (0.1–0.5 m soil depth). This range almost reached the values of those in the 12-month Leucaena leucocephala treatment, indicating a high contribution of the root system to soil–root reinforcement. © 2007 Elsevier B.V. All rights reserved. Root length density Soil–root matrix Root reinforcement 1. Introduction In the past few decades, topics on ecological restoration have been much discussed due to the increase in awareness of environmental conservation issues. Various terminologies and techniques have been established, integrated and implemented in order to obtain a structurally sound, ecologically sustainable and socio-economic method for restoring the ecosystem. One of the techniques of interest is slope bioengineering, the use of living plant materials to perform some engineering and ecological functions for slope stabilization and site restoration of riverbanks, which is widely investigated and discussed (e.g. Haigh and Gentcheva-Kostadinova, 2002; Li and Eddleman, 2002; Li et al., 2006). The mechanism used includes the enhancement of soil shear strength using vegetation–soil systems and limiting soil particle move- ments on slope via utilizing the effects of root systems on soil structure. It is hence of great interest to explore further the influence of the engineering properties of the root system on slope stability and shallow landsliding. Much has been investigated and written about root growth, phenology and function but very little attention has been given to aspects of roots concerning stabilization of slopes. This is mainly due to destructive harvesting and the difficulties in collecting representative samples and separating the roots from soil in the field or on the slope. However, the progress made in the past few years on the contribution of the root system in reinforcing mass-stability of the slopes is an eye-opener. Various studies have documented the important effects of root density in preventing landslides (Gray, 1995; Abe, 1997). The mechanical or reinforcing effect of plant roots on the stability of slopes is mainly attributed to, inter alia, the Corresponding author. Tel.: +60 3 79674185; fax: +60 3 79674178. E-mail address: normaniza@um.edu.my (O. Normaniza). Abbreviations: RLD, root length density; WAC, water absorption capacity. 0925-8574/$ – see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.ecoleng.2007.11.004 ∗ 216 e c o l o g i c a l e n g i n e e r i n g 3 2 ( 2 0 0 8 ) 215–221 increase of shear strength of the soil (Wu, 1995). Roots increase soil shear strength by anchoring a soil layer and by forming a binding network within the layer (Ziemer, 1981). Hence, for the purpose of slope vegetation screenings, a root profile study is indeed essential. Amongst the versatile leguminous trees, Leucaena leucocephala has been determined as a potential slope plant (Parera, 1982). It is widespread throughout the tropics and is abundant in villages in the northern part in Malaysia. It is a multipurpose tree which profusely produces propagules (beans) and has been used as an erosion control plant. It was also recently reported that L. leucocephala showed high tolerance and survival and is a potential plant in the revegetation of lagoon ash in China (Cheung et al., 2000). There is, however, limited documentation on the contribution of this species in terms of slope stability enhancement in Malaysia as well as in other countries. Hence, L. leucocephala was chosen in this project based on these reports, coupled with the fact that there is a woeful lack of documentation on its contribution to slope stabilization. In view of this, the objective of this paper is to quantify the root effect to slope stability in terms of root profile and root reinforcement capacity of the species studied. The shear box test was carried out to investigate the strengthening effects of plants on soil shear strength in terms of shear strength properties and to compare L. leucocephala with two other potential slope species, namely, Bauhinia purpurea and Bixa orellana. The latter two species have shown promising physiological characteristics as slope plants in a previous preliminary experiment (Normaniza, 1998). 2. Materials and methods 2.1. Root profile experiment 2.1.1. Plant materials L. leucocephala seedlings were grown in PVC pipes (each 200 cm in length; 30 cm in diameter) filled with slope soil. In a PVC, three seeds were inoculated and germinated and later thinned to one seedling when a height of 15 cm was reached. The experiments were conducted at 6 and 12 months at three replications per treatment. The pipes were placed in a completely randomized design (CRD) under prevailing atmospheric conditions (relative humidity 60–81% RH, maximum photosynthetically active radiation (PAR) 2000 ␮E m−2 s−1 and temperature 21–34 ◦ C). The plants were watered twice a day to maintain turgidity. Root length densities as well as water absorption capacity (WAC) were assessed. 2.1.2. Table 1a – Physical properties of the soil Atterberg limits Liqiud limit Plastic limit Plasticity index 26.9% 14.59% 21.31% Linear shrinkage Specific gravity 3.23 2.61 Compaction Optimum moisture content Maximum dry density 13.5% 1.8515 mg/m3 Table 1b – Grain size distribution Type Size distribution (%) Gravel (2–60 mm) Sand (0.06–2 mm) Silt (0.002–0.06 mm) Clay (<0.002 mm) 10.0 79.5 7.5 3.0 mented into 40 cm of soil depth. This gave rise to a soil volume of 0.0283 m3 . 2.1.4. Water absorption capacity (WAC) The WAC was formulated based on Baker’s theory (Baker, 1994). According to this theory, 98% of the water absorbed by roots is transpired to the atmosphere. This statement leads to the formula of calculating WAC = [(transpiration rate × 100)/98]/total root length. 2.2. Soil–root matrix experiment 2.2.1. Plant materials The species used in this trial were L. leucocephala, B. purpurea and B. orellana. Three seeds of each plant species were germinated and later thinned to one seedling when a height of 10 cm was reached. The plant density is one plant per 300 mm × 300 mm, planted in the center of the shear box (Fig. 1). All the species studied were grown for a duration of 6 and 12 months under prevailing conditions (RH 70–90%, temperature 32–38 ◦ C, and maximum PAR 2025 ␮E m−2 s−1 ). Soil The soil was collected from a slope and subjected to a number of standard tests to determine the basic physical properties. Based on the grain–size distribution curve, the soil is described as silty sand and its physical properties are shown in Tables 1a and 1b. 2.1.3. Root length density (RLD) The RLD was calculated as total root length/soil volume. Root length was measured using a leaf area image analyser (Image Analyser, Delta-T devices, UK). The 200 cm PVC pipe was seg- Fig. 1 – Soil sample was prepared in a stacked perspex box. 217 e c o l o g i c a l e n g i n e e r i n g 3 2 ( 2 0 0 8 ) 215–221 Fig. 3 – Root length density (RLD) of Leucaena leucocephala at 6 and 12 months of growth. Fig. 2 – A schematic diagram of the automatic modified direct shear box machine. The shear boxes were placed in a CRD and the plants were watered every morning to avoid stress conditions. Each of the plant samples as well as the controls was replicated three times. Each plant sample was cut off near the lowest part of the stem prior to measurement. Plant height and root circumference were also measured. The shoot was partitioned into stem and leaf and both were oven-dried at 80 ◦ C (120 h) to obtain dry weight partitioning and total shoot biomass. 2.2.2. Shear box preparation Three perspex boxes, each 300 mm × 300 mm × 200 mm, were stacked to form a box of 60 cm in height (Fig. 1). During compaction the perspex box was placed inside a specially designed steel box to avoid deformation. Perspex was chosen in order to observe the development of the root system. The soil used in this trial was similar to that of the previous experiments (slope soil). The soils at a moisture content of 15% were compacted and the compaction effort was 29.43 J kg−1 . Ten kilograms of soil were compacted using a 2.5 kg rammer falling freely from a height of 300 mm for 40 blows. This process was repeated until the soil filled into the box of 60 cm in height. The mean dry density of the soil samples was 1800 kg m−3 . Prior to testing, the soil samples were divided into three parts to determine shear strengths at depths of 0.1, 0.3 and 0.5 m. 2.2.3. 3. Results and discussion 3.1. Root profile of L. leucocephala 3.1.1. Root length density (RLD) Root length density (RLD) decreased with increasing soil depth (Fig. 3). The results indicate that a favourable condition for root growth varies at different depths of soil. After 6 months of treatment, all plants showed the highest percentage of RLD at the first 80 cm of soil depth, which contributed to 53% of the total RLD. After 12 months, however, the highest was observed at 40 cm of soil depth, contributing 74.0% of the total RLD. A similar root distribution, higher density at a soil depth of 40–80 cm, was also observed in soybean (DelCastillo et al., 1989) and cotton (Reddy, 1997). The total RLD at 12 months was observed to be twice that of the plants at 6 months. This considerable increase implies that the contribution of root density to slope stabilization increases with time. 3.1.2. Water absorption capacity (WAC) Water absorption capacity (WAC) of L. leucocephala after 12 months of treatment was about fivefold that of 6-month plants (Table 2). The result may be due to the increment of its RLD, which was observed to be doubled in the 12 months (Fig. 3). In comparison with the other potential slope plants (Normaniza, 1998), WAC of L. leucocephala displays the highest value (Fig. 4). In fact, its value is almost fourfold and eightfold that of vetiver and Justicia betonica, respectively. With a high capacity to absorb more water, 0.18 L H2 O/m root/day, soil water becomes less saturated, and in turn slope soil becomes drier, thus reducing the potential of slope failure. A model established Laboratory testing The shear strength of each soil sample was determined using a modified 300 mm × 300 mm × 200 mm direct shear box machine (Fig. 2). Each test was carried out according to BS 1377: Part 7: 1990: 5 (British Standard Methods, 1990). The normal stresses used were 10, 20 and 30 kPa. An automatic data logging system was used to record all required data, which were then transferred to a personal computer using the software PC208W. Since the dimension of each sample box is in the same size as the shear box, each sample was placed in the shear box just by carefully and slowly pushing the sample downward. Table 2 – Water absorption capacity and related parameters Treatment (months) 6 12 E WAR WAC 12.96 ± 0.86 135.0 ± 41.3 13.36 ± 1.1 137.78 ± 42.16 0.037 ± 0.09 0.18 ± 0.04 E = transpiration rate (L H2 O/plant/day); WAR = (transpiration rate × 100)/98 (L H2 O/plant/day); WAC = WAR/total root length (L H2 O/m root/day). 218 e c o l o g i c a l e n g i n e e r i n g 3 2 ( 2 0 0 8 ) 215–221 Fig. 4 – Water absorption capacity (WAC) of L. leucocephala compared to other potential slope plants. All plant species except L. leucocephala had been previously examined using the same experimental protocol and growing duration (Normaniza, 1998). Fig. 6 – Angle of friction (◦ ) after 6 months of growth. by Pollen (2007) showed that root reinforcement was at its lowest when soil moisture was high and soil shear strength was low. Therefore, the root reinforcement is likely to be at its minimum value when the soil is saturated, and vice versa. However, further investigation is essential, as the relationship between soil moisture and shear strength is dependent on spatial and temporal variability in several factors, including soil matric suction and geotechnical properties (Pollen, 2007). 3.2. Soil–root matrix L. leucocephala 3.2.1. Shear strength, cohesion and friction factors Six months after planting, the reinforcement of the roots of all the species studied had improved the shear strength of the soil. This result implies that the root systems of the species studied contribute to the enhancement of soil shear strength. Roots of L. leucocephala increased the cohesion factor of shear strength by twofold (0.1 m), threefold (0.3 m) and fivefold (0.5 m) that of the control soil (Fig. 5). The results imply that the root system of the species studied enforces high inter- Fig. 5 – Values of cohesion factor (kPa) after 6 months of growth. Fig. 7 – Values of coheson factor (kPa) after 12 months of growth. nal forces that hold soil particles together in a solid mass. The cohesion factor of L. leucocephala was enhanced by six- and sevenfold (at 0.1 m soil depth) that of B. purpurea and B. orellana, respectively. B. purpurea exhibited higher cohesion factor than B. orellana by 143.7% at a soil depth of 0.5 m. At a soil depth of 0.3 m, B. orellana showed higher cohesion factor by 33.5% than that in the B. purpurea. The results also indicate that no root system of any of the species studied had much effect on the angle of friction (Fig. 6). It implies that the friction between the soil particles for the control as well as the different species studied is more or less consistent. The reinforcement of the roots of all species studied had improved the shear strength of the soil after 12 months of observation. The cohesion factor of shear strength increased by eightfold (0.1 m), fourfold (0.3 m) and sixfold (0.5 m) compared to those in the control (Fig. 7). Amongst the species studied, L. leucocephala had the highest cohesion at both the soil depths of 0.1 and 0.5 m. B. purpurea showed the lowest value of cohesion except at a depth of 0.5 m. The results also reveal a significantly higher angle of friction in both L. leucocephala and B. purpurea at 0.3 m soil depth 219 e c o l o g i c a l e n g i n e e r i n g 3 2 ( 2 0 0 8 ) 215–221 after 12 months of observation (Fig. 8). However, a consistent value was observed at the other soil depths. The results showed that the cohesion factor of all the species studied increased with time (Fig. 9). In 12 months, L. leucocephala was observed to increase its cohesion factor by more than threefold of those in the 6-month trial at a depth of 0.1 m. However, there was no distinct effect at other depths (0.3 and 0.5 m). The considerable increase of root reinforcement at 0.1 m implies that L. leucocephala is a good species for providing surficial erosion control. In contrast, B. orellana and B. purpurea showed markedly increased value in cohesion between the two experimental periods at all soil depths. The values of cohesion in both species are in fact lower than their controls during 6 months of observation. The loosening effect may be greater than the strengthening effect whereby the roots crack the soil at initial growth. Furthermore, the root had not been established in the early growth period as compared to the plants grown for 12 months. These results also indicate that it takes 12 months for both species to reach a higher level of reinforcement effect. The root systems of all species studied exhibit an inconsistent trend and do not seem to improve the angle of friction significantly with respect to time (Fig. 10). 3.2.2. Relationship between shear strength and root profiles There is a positive correlation between shear strength (at all normal pressure) and RLD (Fig. 11), implying that a dense root density at a soil depth of the first 60 cm for all the species studied helps to reinforce the soil by increasing its shear strength. This helps in soil reinforcement at the surface level, thus reducing surface erosion. This increase in soil strength by an increase in cohesion is particularly brought about by the binding action in the fine roots or soil composite and adhesion of the soil particles to the roots (Styczen and Morgan, 1995). Studies by O’Loughlin (1984) have also indicated that shear strength provided by fine roots, 1–20 mm in diameter, contributes most to soil reinforcement. The root system penetrates the soil mass, reinforces it, bringing about an increase in cohesion and, hence, in soil shear strength. Apart from that, the increase in cohesion, is probably partly due to an increase amount of fiber (fine) roots of the species studied. In addition, a fine root mat close to the soil surface may act like a low-growing vegetation cover and protect the soil from erosion. Rickson and Morgan (1988) also reported that an increase in the percentage area occupied by fine roots produces an exponential decay in the soil loss ratio. Thus, the combined effect of primary and fine root systems resulted in the enhancement of soil–root matrix shear strength. Table 3 – Root reinforcement of Leucaena leucocephala and its increment (ıs) throughout the experimental period Soil depth (m) 6 months (kPa) 12 months (kPa) ıs (%) 9.90 14.85 23.46 34.0 16.34 26.06 243.4 9.91 9.98 0.1 0.3 0.5 4. Discussions Both the root profile and the reinforcement capacities of L. leucocephala have been discussed thoroughly. The shear strength of the species studied increases with increasing normal stress in both trials (6 and 12 months). This is in line with law of soil mechanics that soil shear strength increases with the increase pressure on shear surfaces (Wu, 1995). The root reinforcements of the species studied contribute significantly to soil shear strength. The results also indicate that the shear strength rapidly increases, in line with the cohesion factor, with increasing root length density. It may be due to the increase interaction between the root and soil particles as reported earlier (Tobias, 1995; Wulfsohn et al., 1996). In addition, vegetation has its greatest effect close to the soil surface where the root length density is generally the highest. Amongst the species, L. leucocephala reveals the highest root reinforcement by a significant increase in the cohesion factor with respect to time. The cohesion factor of L. leucocephala after 6 months of planting increased by almost doublefold (0.1 m), triplefold (0.3 m) and fivefold (0.5 m) that of the control by almost eightfold (0.1 m), fourfold (0.3 m) and sixfold (0.5 m) that of the control after 12 months of growth. These observations brought to the increase (ıs ) between the two trials are about 243% (0.1 m) and 10% at both the soil depths of 0.3 and 0.5 m (Table 3). The extensive growth and development of taproots in L. leucocephala has probably caused the tremendous enhancement of shear strength at a soil depth of 0.1 m. Amongst the species studied, L. leucocephala has the highest ıs (%) of root reinforcement at all soil depths, except at 0.3 m, after 12 months of growth (Table 4). B. orellana seems to provide a promising root reinforcement capacity, however, only to the short distance of soil depth. Apart from the mechanical attribution of the species, its hydrological role is also outstanding. It can partly help in strengthening the slope via its capacity to absorb water. This is shown by its prominent WAC compared to other potential slope plants. It can be envisaged that with higher RLD, the soil water content of the slope becomes lower as a consequence of the higher WAC, thus, in turn, lowering the risk of landslide and erosion (Normaniza and Barakbah, 2006). Table 4 – Root reinforcement (kPa) of all species studied (as compared to control) after 12 months of growth Soil depth (m) 0.1 0.3 0.5 Control (kPa) 4.39 4.39 4.39 L leucocephala (LL) (kPa) 34.0 16.34 26.06 Bauhinia purpurea (BP) (kPa) 22.10 7.27 18.67 Bixa orellana (BO) (kPa) 26.76 19.93 9.60 (LL) ıs (%) (BP) (BO) 674.5 272.2 493.6 403.4 65.6 325.3 509.6 354.0 118.7 220 e c o l o g i c a l e n g i n e e r i n g 3 2 ( 2 0 0 8 ) 215–221 Fig. 11 – Correlations between shear strength (at all normal pressure) and root length density (RLD). Fig. 8 – Angle of friction (◦ ) after 12 months of growth. Fig. 9 – The increment of cohesion factor of all the species studied throughout the observation (() 6 months; () 12 months). Fig. 10 – The increment of angle of friction of all species studied throughout the observation (() 6 months; () 12 months). e c o l o g i c a l e n g i n e e r i n g 3 2 ( 2 0 0 8 ) 215–221 5. Conclusion This project was the first project investigating a leguminous tree as a slope plant in terms of its engineering capacities. This initial attempt shows that the species studied has the potential to play a major mechanical role as well as hydrological role in stabilizing slopes and protecting against soil erosion. It is suggested that the high capacity of the root reinforcement and high water absorption capacity of L. leucocephala rank it as an outstanding future slope remedy to prevent slope failure. Acknowledgements The author would like to thank The Ministry of Science, Technology and Innovation, Malaysia, and Universiti Malaya for a fellowship and vote-F research grant, respectively, throughout this project. The author is also grateful to the North-South Expressway Project Berhad (PLUS Bhd) for the use of their equipment. references Abe, K., 1997. A method of evaluating the effect of trees roots on preventing shallow-seated landslides. Bull. Forest. For. Prod. Res. Inst. 1 (373), 1105–1181. Baker, D.A., 1994. Water relation. In: Malcolm, B.W. (Ed.), Advance Plant Physiology. English Language Book Society, Longman, pp. 297–312. British Standard Methods of Tests for Soils for Civil Engineering Purposes, 1990. Part I: General Requirements and Sample Preparation. B.S. Institution, UK. Cheung, K.C., Wong, J.P.K., Zhang, Z.A., Wong, J.W.C., Wong, M.H., 2000. Revegetation of lagoon ash using the legume species Acacia auriculiformis and L. leucocephala. Environ. Pollut. 109, 75–82. DelCastillo, D., Acock, B., Reddy, V.R., Acock, M.C., 1989. Elongation and branching of roots on soybean plants in a carbon dioxide-enriched aerial environment. Agron. J. 81, 692. Gray, D.H., 1995. Influence of vegetation on the stability of slopes. In: Barker, D.H. (Ed.), Vegetation and Slopes Stabilisation, Protection and Ecology. Thomas Telford House, London, pp. 2–23. Haigh, M.J., Gentcheva-Kostadinova, S., 2002. Ecological erosion control on coal-spoil banks: an evaluation. Ecol. Eng. 18, 371–377. 221 Li, M.H., Eddleman, K.E., 2002. Biotechnical engineering as an alternative to traditional engineering methods—a biotechnical streambank stabilization design approach. Landscape Urban Plan. 60, 225–242. Li, X.P., Zhang, L.Q., Zhang, Z., 2006. Soil bioengineering and the ecological restoration of riverbanks at the Airport Town, Shanghai, China. Ecol. Eng. 26, 304–314. Normaniza, O., 1998. The influence of plant water relations and vegetation cover on slope stability. M.Sc. Thesis. Department of Botany, University Malaya. Normaniza, O., Barakbah, S.S., 2006. Parameters to predict slope stability—soil water and root profiles. Ecol. Eng. 28, 90–95. O’Loughlin, C.L., 1984. Effectiveness of introduced forest vegetation for protecting against landslides and erosion in New Zealand’s steeplands. In: Paper Presented for a Symposium on Effects of Forest Land Use on Erosion and Slope Stability, Honolulu, Hawaii. Parera, V., 1982. Leucaena for erosion control and green manure in Sikka. Leucaena research in the Asian-Pacific region. In: Proceedings of a Workshop held in Singapore, November 23–26, 1982, pp. 169–173. Pollen, N., 2007. Temporal and spatial variability in root reinforcement of streambanks: accounting for soil shear strength and moisture. Catena 69, 197–205. Reddy, V.R., 1997. Root growth of cotton as influenced by CO2 and temperature. In: Altman, A., Waisel, Y. (Eds.), Biology of Root Formation and Development. Plenum Press, New York and London, pp. 237–241. Rickson, R.J., Morgan, R.P.C., 1988. Approaches to modelling the effects of vegetation on soil erosion by water. In: Morgan, R.P.C., Rickson, R.J. (Eds.), Commission of the European Communities, pp. 237–254 (DG. VI EUR. 10860 EN). Styczen, M.E., Morgan, R.P.C., 1995. Engineering properties of vegetation. In: Morgan, R.P.C., Rickson, R.J.E. (Eds.), Slope Stabilisation and Erosion Control: A Bioengineering Approach. E and FN Spon, London, pp. 5–58. Tobias, S., 1995. Shear strength of the soil root bond system. In: Barker, D.H. (Ed.), Vegetation Slopes Stabilisation Protection Ecology. Thomas Telford House, London, pp. 280–285. Wu, T.H., 1995. Slope stabilization. In: Morgan, R.P.C., Rickson, R.J. (Eds.), Slope Stabilisation and Erosion Control: A Bioengineering Approach. E & FN Spon, London, pp. 221–264. Wulfsohn, D., Adams, B.A., Fredlund, D.G., 1996. Application of unsaturated soil mechanics for agricultural conditions. Can. Agric. Eng. 38, 173–181. Ziemer, R.R., 1981. Roots and the stability of forested slope. In: Proceedings of the International Symposium On Erosion and sediment Transport in Pacific Rim Steeplands, International Association of Hydrological Sciences, Christchurch, New Zealand. Publication vol. 132, pp. 343–361.