Comparison of soil macro-invertebrate communities in Malaysian oil palm plantations with secondary forest from the viewpoint of litter decomposition

Forest Ecology and Management 381 (2016) 63–73

Contents lists available at ScienceDirect

Forest Ecology and Management journal homepage: www.elsevier.com/locate/foreco

Comparison of soil macro-invertebrate communities in Malaysian oil palm plantations with secondary forest from the viewpoint of litter decomposition Mum-Keng Wong a,b, Jiro Tsukamoto c,⇑, Yusufujiang Yusuyin d, Sota Tanaka c, Kozo Iwasaki c, Ngai-Paing Tan e a

Felda Global Ventures Research and Development Pte. Ltd., Kuala Lumpur, Malaysia University of Malaya, Kuala Lumpur, Malaysia c Faculty of Agriculture, Kochi University, Nankoku 783-8502, Japan d The United Graduate School of Agricultural Sciences, Ehime University, 3-5-7 Tarumi, Matsuyama, 790-8566, Japan e Faculty of Agriculture, Universiti Putra Malaysia, 42400 UPM Serdang, Selangor, Malaysia b

a r t i c l e

i n f o

Article history: Received 26 May 2016 Received in revised form 2 September 2016 Accepted 12 September 2016

Keywords: Deforestation Earthworms Management practice Microhabitats Termites

a b s t r a c t Biodiversity decline in rapidly expanding oil palm plantations is of global concern. Many studies have demonstrated that fauna species diversity is lower in oil palm plantations than forests. However, information about the flow-on effects of these declines in species diversity on ecosystem functioning is scarce for oil palm plantations. Litter decomposition performed by soil organisms is a vital ecosystem function that regulates nutrient cycling and carbon sequestration. Some studies have found a high level of redundancy among litter decomposing species. In order to evaluate the effects of the conversion of forests to oil palm plantations on decomposition, we investigated the abundance and biomass of soil macroinvertebrates at sites in two oil palm plantations and a secondary forest in Malaysia. Biodiversity of soil macro-invertebrates were lower in the oil palm plantations than in the secondary forest. The abundance and biomass of surface–living litter transformers was lower in oil palm plantations than forest, probably due to the isolated piles of frond litter that occur in plantations, instead of the more continuous litter layer observed in forests. However, we found dense populations of wood (litter)–feeding termites in the thick rachises of fronds heaped on the ground surface. A pantropical earthworm species, Pontoscolex corethrurus, which buries the litter through cast deposition, abounded more in the oil palm plantations than in the secondary forest. These characteristics of soil macro-invertebrates have also been reported in other oil palm plantations. Thus, we conclude that the conversion of forests to oil palm plantations may reduce diversity of soil macro-invertebrates, increase the heterogeneity of macroinvertebrates distribution and decrease populations of some functional groups of soil macroinvertebrates. However, overall, forest conversion does not appear to have a negative impact on the decomposition process to a great extent, owing to the colonization of plantation sites by other groups of decomposer animals that are favored by disturbance and/or the great amount of localized input of fresh fronds pruned at the time of fruit harvesting. Ó 2016 Elsevier B.V. All rights reserved.

1. Introduction To date, most of the major oil palm plantations comply with an international certification system that sets voluntary standards for producers and provides assurances to consumers, such as the Roundtable on Sustainable Palm Oil, one of the aims of which is to reduce the impact of deforestation on ecosystem services ⇑ Corresponding author. E-mail address: [email protected] (J. Tsukamoto). http://dx.doi.org/10.1016/j.foreco.2016.09.011 0378-1127/Ó 2016 Elsevier B.V. All rights reserved.

(RSPO Principles and Criteria Review, 2011/2012). However, rapid expansion of oil palm plantations in recent decades has been widely criticized for deforestation in the tropics, especially in Southeast Asia (Singh and Bhagwat, 2013), one of the world’s most biodiverse regions (Sodhi et al., 2010). As reviewed by Fitzherbert et al. (2008) and Savilaakso et al. (2014), conversion of forests, whether intact or disturbed, to oil palm monocultures reduces species richness across a wide range of animal taxa. In view of this, much effort has to be put on the research that promotes balance between cultivation of oil palm and biodiversity conservation

64

M.-K. Wong et al. / Forest Ecology and Management 381 (2016) 63–73

(Turner et al., 2011). At the same time, comparable efforts should be made to evaluate the consequences of reduced biodiversity for ecosystem functions. Although there is almost no study that seeks to relate biodiversity and ecosystem services in oil palm plantations (Foster et al., 2011; Savilaakso et al., 2014), the reduction in species richness of the above-ground animals is expected to impoverish those ecosystem functions, such as pollination, seed dispersal, and biological pest control, that strongly depend on species-to-species relationships. However, it is not necessarily obvious whether decomposition, an ecosystem function that regulates nutrient cycling and carbon sequestration, is adversely affected to a similar extent by reduced richness of soil invertebrate species, since there is a high level of species redundancy among litter decomposers (Andren et al., 1995), and decomposition is considered to be less species specific (Kurokawa and Nakashizuka, 2008) than the functions mentioned above. For example, Foster et al. (2011) did not find any significant difference in litter mass loss in litterbags between a primary forest, logged forest, and oil palm plantation, owing to the activities of a single species of the widespread termite Macrotermes gilvus in the plantation. In contrast to the general declining trend in faunal diversity with increasing simplification of the flora, abundance and/or biomass of animals is not always depressed in disturbed systems including oil palm plantations (Foster et al., 2011; Savilaakso et al., 2014) often due to colonization by a few generalist- or cosmopolitan-species with high environmental tolerances and high potential of population growth. In this context, despite confining ourselves only to comparing the soil macro-invertebrates in primary forests and tree plantations in humid tropical areas, we find sufficient evidence demonstrating that the abundance and/or biomass in the tree plantations is more than that in primary forests: earthworm abundance and biomass in Peru (peach palm, Bactris gasipaes; Lavelle and Pashanasi, 1989), abundance of beetle larvae, isopods, moth larvae, and cockroaches in Brazil (Acacia mangium; Pellens and Garay, 1999), biomass of earthworms, harvestmen, isopods, and whole soil macro-invertebrates in Sarawak (Acacia mangium; Tsukamoto and Sabang, 2005), isopod abundance in Sabah (oil palm, Elaeis guineensis; Hassall et al., 2006), and so on. Therefore, it is possible that the functional role of soil macro-invertebrates is not necessarily lower in tree plantations than in primary forests in the tropics. One of the most noticeable environmental characteristics that litter and soil animals experience in oil palm plantations is the alternating pattern of frond heaps and bare ground, brought about by localized input of fresh fronds pruned at the time of fruit harvesting, rather than the continuous litter layer formed by homogeneous input of senescent leaves. In an oil palm plantation, the distribution of soil macro-invertebrates was reported to be significantly affected by this habitat heterogeneity (Carron et al., 2015). Therefore, an estimate of the abundance and/or biomass of soil macro-invertebrates and an evaluation of their functions in oil palm plantations should be performed after taking into account this habitat heterogeneity. Several recent studies have quantitatively investigated soil macro-invertebrate communities in oil palm plantations in relation to soil quality (Lavelle et al., 2014; Carron et al., 2015) and the resulting energy flux through such communities (Barnes et al., 2014). However, no studies have fulfilled all of the following conditions needed to evaluate the functional role of soil macroinvertebrates in oil palm plantations: macro-invertebrate collection in both the surface litter and the underlying soil; stratified sampling taking habitat heterogeneity into consideration; and measurement of both the abundance and biomass of individual functional or taxonomic groups. We studied the abundance and biomass of soil macroinvertebrates in the litter and mineral soil in two oil palm planta-

tions of different planting ages in Tekam, Malaysia, taking into account three different microhabitats caused by management practices: weeded circle, harvest path, and frond heap. A neighboring secondary forest with a closed canopy was selected as a control forest site that was disturbed but with a continuous layer of litter. Our goals were (1) to describe soil macro-invertebrate communities in oil palm plantations compared with secondary forest using measures that, unlike previous studies, allow for an assessment of the functional role of these communities, and (2) to use this assessment to determine the most plausible evaluation how the conversion of forests to oil palm plantations might affect the decomposition of frond piles on the forest floor. 2. Material and methods 2.1. Study site The study was carried out in 2012 at an oil palm plantation in Tun Razak Agricultural Research Centre in Jengka Triangle, 1750 ha in area, situated on undulating terrain at an altitude of 45–75 m a.s.l. along the Pahang River; in the state of Pahang, Malaysia. The annual mean temperature is 27.4 °C, and the average annual precipitation is 2097 mm without a distinctive dry season (Tekam Meteorological station, 10 years average from 2003 to 2012). According to the Department of Mineral and Geoscience Malaysia, the Jengka Triangle exhibits largely Triassic-Jurassic geology, but older rocks from the Permian were also found. Most of the rocks originated from the volcanic rock consisting of prominent lava flows, tuffs and agglomerates of andesitic and rhyolitic composition. At the study site, the soils have derived from quartz diorite or quartz andesite (Paramananthan, 2000). Two oil palm plantations of different planting ages, 5 and 18 years after conversion from cocoa plantations, 33.2 and 60.7 ha in area, respectively (N3°540 17.600 E102°310 28.600 ), were selected from the areas of almost flat terrain. The operational procedures in both of the conversion from forests to cocoa plantations and from cocoa plantations to oil palm plantations included logging, bulldozing, and burning. A neighboring secondary forest with a closed canopy, 8.8 ha in area (N3°520 16.200 E102°300 59.600 ), was selected as a control forest site. They were given habitat-codes of OP5, OP18, and SF, respectively. Soil fertility status in the study sites has been discussed by Yusuyin et al. (2015, 2016) and Tan et al. (2014, 2015) in relation to the same microhabitats as those in the present study: weeded circle, harvest path, and frond heap. 2.2. Microhabitats Commercial oil palm trees (Deli dura  Yangambi pisifera) were planted in equilateral triangular arrangement with a distance of 9.1 m between each tree and at a density of 136 trees ha1. The following three microhabitats were distinguished and designated by a microhabitat-code, WC, HP, and FH. WC (weeded circle) is a circle area 1.5 m in radius from each tree trunk. This microhabitat receives herbicides and inorganic fertilizers periodically and has exposed ground surface except when a pile of fruit bunches is sometimes formed during the harvesting operation. The area between two adjacent rows of planted trees is alternately used for passage and as a depositing place for fronds that are pruned at the time of fruit harvesting. HP (harvest path) is the passage area where no frond litter is supplied and lacks the surface litter layer, although the ground surface is covered with soft grass. FH (frond heap) is the area stacked with oil palm fronds. In a 20-year-old oil palm plantation with a density of 143 palms ha1, planted in an equilateral triangular arrangement at a distance of 9 m, the

M.-K. Wong et al. / Forest Ecology and Management 381 (2016) 63–73

percentage share of the total area of each microhabitat was estimated to be 10% WC, 65% HP, and 25% FH (Lamade et al., 1996). In the present study, we used values of 10%, 60%, and 30% for the respective microhabitats based on the results of a field survey conducted in a 20 m  20 m quadrat at both plantation sites. Since the control habitat SF was characterized by a uniform microhabitat with homogeneous stand structure and forest floor, it was also used as a control for microhabitat in relation to the microhabitats WC, HP, and FH. That is, in the present study, we examine comparisons between three habitats (OP5, OP18, and SF) and between seven microhabitats (OP5 WC, OP5 HP, OP5 FH, OP18 WC, OP18 HP, OP18 FH, and SF). 2.3. Soil animal sampling

65

Kruscal-Wallis test was made with SPSS version 17 (IBM Corp. Armonk, NY, USA). PCA and Monte-Carlo Test on a Between-Class Analysis were performed using the R environment and ade4TkGUI package (Thioulouse and Dray, 2007; R Development Core Team, 2004). 2.6. Identification of species of earthworms and termite soldiers Species identification was made under a binocular stereo microscope based on morphological characters of preserved specimens following Blakemore (2002) for earthworms and Tho (1992) for termite soldiers. 3. Results

Six quadrats of 30 cm  30 cm were randomly placed in SF and in each microhabitat WC, HP, and FH in both of OP5 and OP18. The six quadrats in SF, 18 in OP5, and 18 in OP18 were at least 5 m apart from each other and within a circular area 50 m in diameter. The A0-layer and the soil layers at 0–10, 10–20 and 20–30 cm depth were taken from the quadrat and spread over a vinyl sheet separately in the field. The soil macro-invertebrates, which are broadly defined as the animals larger than 2 mm in body length, were collected by hand sorting using a forceps and an aspirator. In the case of five major taxa of soil macro-invertebrates—ants, termites, spiders, cockroaches, and woodlice—, the individuals less than 2 mm in body length were also collected as far as possible. On the other hand, mesofauna individuals of mites, springtails, symphilids, diplurans, enchytraeids, etc. were not sampled even when larger than 2 mm. Collected animals were preserved in 75% ethanol except the earthworms which were preserved in 10% formalin. In the laboratory, wet weights of the preserved animals were determined individually to 0.1 mg accuracy using an analytical balance after carefully placing on filter paper. 2.4. Data processing The soil macrofauna biomass per unit area and biomass-based taxonomic composition in the studied oil palm plantations were compared with those in the secondary forest (SF). For this, weighted mean biomass was calculated for the plantation sites based on the percentage of each microhabitat to the total area:

Weighted mean biomass ¼ ðmean biomass in WC  0:1Þ þ ðmean biomass in HP  0:6Þ þ ðmean biomass in FH  0:3Þ 2.5. Statistical analysis The differences in the abundance and biomass of each taxonomic group of soil macro-invertebrates among the microhabitats were tested by Kruscal-Wallis test followed by Tukey-Kramer test for non-parametric variables. Characterization of each microhabitat in terms of the soil macro-invertebrates was made by principal component analysis (PCA). Biomass of the whole community (total) and biomasses of the following six taxonomic groups were used as variables: 1. Oligochaeta (earthworms); 2. Isoptera (termites); 3. Surface–living litter transformers (‘‘SLLT” = Isopoda [isopods] + Blattariae [cockroaches] + Orthoptera [crickets and grasshoppers, referred below as crickets]); 4. Surface–living predators (‘‘SLP” = Araneae [spiders] + Carabidae [carabid beetles] + Staphylinidae [staphylinid beetles]); 5. Formicidae (ants); and 6. Dermaptera (earwigs). The significance of the ordination of microhabitats on the PCA coordinates was tested with Monte-Carlo test based on 999 replicates.

3.1. Vertical distribution of biomass in the soil profile Fig. 1 illustrates the vertical distribution of total biomass of the whole community (Fig. 1(H)) and biomass of major taxa (Fig. 1(A)–(G)) that contributed more than 5% to the total biomass in at least one of the three habitats (OP5, OP18, and SF). Weighted mean biomass was used for OP5 and OP18. In SF, more than 10% of biomass of earthworms (Fig. 1(C)), ants (Fig. 1(G)), and total fauna (Fig. 1(H)) was contributed from the deepest sampled soil layer (20–30 cm). Therefore, a small fraction of the soil fauna in SF might have occurred in the soil deeper than 30 cm and was missed by the sampling. In OP5 and OP18, however, more than 95% of the biomass of the total fauna and of each of the major taxa was collected from the A0-layer and layers in the upper 20 cm of the soil. Thus, the sampling depth of 30 cm is considered sufficient to collect almost entire soil macro-invertebrates faunas in OP5 and OP18. In relation to functional role of earthworms and termites, the ‘‘ecosystem engineers”, which are able to modify the soil environment through their mechanical activities (Lavelle, 1997), it is worth noting that, while little or no earthworms was found in the A0-layer, a substantial part of termite biomass (Fig. 1(E)) in oil palm plantations, especially in OP18, was contributed from the A0-layer regardless of the localized distribution of the frond litter. 3.2. Diversity of soil macrofauna Table 1 shows the mean abundance of soil macro-invertebrates in each of the seven microhabitats and the total number of individuals collected from the 42 samples. Animals were classified into 19 broad taxonomic groups, including miscellaneous groups, ‘‘Others 1–3.” The total number of collected animals was largest for termites, second largest for ants, and third largest for earthworms. Each of these dominant taxa tended to occur at substantial densities in all of the seven microhabitats. Snails (Gastropoda), flatworms (Turbellaria), cicada larvae (Cicadidae L.), and scarabaeid larvae (Scarabaeidae L.) were poorly represented in the studied sites, each with a total number of five individuals or fewer. All of these scarce taxa were found in SF but were not collected from WC and HP in both OP5 and OP18. WC and HP, both devoid of a litter layer, were poorly represented by the surface– living litter transformers (isopods, cockroaches, and crickets) and millipedes (Diplopoda), as compared with FH and SF microhabitats with a litter layer. Thus, in comparison to SF, the soil macroinvertebrate community was simplified in WC and HP microhabitats, which cover about 70% of the total area of the oil palm plantations. Table 2(a) shows the species composition of earthworms in three habitats, measured as the total number of individuals from different species. Pontoscolex corethrurus and Dichogaster sp. were

66

M.-K. Wong et al. / Forest Ecology and Management 381 (2016) 63–73

% : SF

% : OP5

: OP18

Fig. 1. Percentage vertical distribution of biomass of major taxonomic groups of soil macro-invertebrates. SF = secondary forest, OP5 = 5-years-old oil palm plantation, OP18 = 18-years-old oil palm plantation.

common in all three habitats. OP18 was almost completely dominated by P. corethrurus, while Dichogaster sp. was present in very small numbers. In OP5, Dichogaster sp. was the most dominant species, but the population of P. corethrurus was also substantial. SF was more evenly occupied by these two species and other unidentified species. Thus, evenness, a component of species diversity, was highest in SF followed by OP5, and lowest in OP18. In terms of functional diversity, both oil palm plantation sites were highly simplified; almost 100% of the total biomass consisted of endogeic species (Table 2(b)). The species composition of termite soldiers, based on total number of individuals collected, and Simpson’s species diversity

index (D) are shown in Table 3. Soldiers from nine species were collected from the three habitats, with two wood (litter)–feeding termites, Microtermes sp. and Odontotermes sp. 1, being common to all three habitats. While Microtermes sp. was almost exclusively dominant among soldiers in OP18, the soldier fauna in OP5 was evenly comprised of Microtermes sp. and Odontotermes sp. 1, resulting in a lower D value in OP5 than in OP18. The most dominant and the second most dominant soldier populations in SF were Microtermes sp. and Odontotermes sp. 1, respectively, but the number of Subritermes sp., a soil–feeding species, was also considerable. Thus, although D value was similar for OP5 and SF, the termite fauna was more functionally diverse in SF than OP5.

67

M.-K. Wong et al. / Forest Ecology and Management 381 (2016) 63–73

Table 1 Abundance of soil macro-invertebrates (N m2) (mean ± SE; n = 6) and total number of individuals collected from 42 sampling pits in oil palm plantations and secondary forest. OP5

OP18

SF

p-value in Kruscal-Wallis test

Total number of individuals collected

WC

HP

FH

WC

HP

FH

Saprophage Gastropoda Oligochaeta Isopoda Diplopoda Blattariae Isoptera Orthoptera Others 1A

0 102 ± 48 0a 0 0 504ab ± 212 0a 7±5

0 69 ± 14 2ab ± 2 9±7 0 328a ± 272 0a 0

0 224 ± 71 46bc ± 16 48 ± 31 20 ± 7 911ab ± 261 4ab ± 2 17 ± 10

0 46 ± 27 6ab ± 4 9±9 0 807ab ± 573 2ab ± 2 0

0 98 ± 31 2ab ± 2 2±2 4±4 89a ± 53 2ab ± 2 0

7±4 78 ± 17 41abc ± 11 17 ± 7 22 ± 14 2293b ± 445 11ab ± 6 4±4

2±2 69 ± 21 85c ± 24 20 ± 8 29 ± 9 1206ab ± 414 35b ± 8 6±6

0.023 0.192 <0.001 0.017 0.002 0.002 <0.001 0.048

5 370 98 57 38 3314 29 18

Zoophage Turbellaria Araneae Chilopoda Carabidae Staphylinidae Formicidae Others 2B

0 17 ± 5 24 ± 6 0 17 ± 13 369 ± 277 2±2

0 26 ± 13 7±4 6±6 7±5 248 ± 104 2±2

0 43 ± 13 30 ± 7 9±7 7±5 603 ± 272 4±2

0 31 ± 11 20 ± 12 0 0 417 ± 156 0

0 37 ± 19 17 ± 4 2±2 2±2 191 ± 90 2±2

2±2 72 ± 14 19 ± 11 2±2 7±4 852 ± 249 11 ± 5

7±4 91 ± 17 39 ± 14 9±7 15 ± 6 1098 ± 399 26 ± 10

0.023 0.017 0.304 0.559 0.309 0.132 0.023

5 171 84 15 30 2036 26

Polyphage Dermaptera

59 ± 50

2±2

98 ± 20

2±2

6±4

11 ± 7

6±4

0.003

99

Rhizophage Cicadidae L. Scarabaeidae L.

0 0

0 0

0 0

0 0

0 0

0 0

4±4 6±4

0.423 0.056

2 3

Others 3C

0a

63b ± 17

<0.001

64

2814ab ± 717

0.003

6464

Total

1100

7ab ± 4 ab

± 451

15ab ± 8

a

ab

713 ± 354

2079

± 388

4a ± 4 ab

1346

4a ± 2 ± 569

26ab ± 7

a

456 ± 143

b

3474 ± 678

OP5 = 5-years-old oil palm plantation; OP18 = 18-years-old oil palm plantation; SF = secondary forest. WC = weeded circle microhabitat; HP = harvest path microhabitat; FH = frond heap microhabitat. a < b < c (p < 0.05; Tukey-Kramer test). A Mainly Lepidoptera larvae and Teneblionidae larvae. B Mainly Opilliones and Schizomida. C Unidentified.

Table 2 Species composition of earthworms based on (a) total number and (b) total preserved wet weight biomass of individuals collected in three habitats. Species

Ecological type

OP5

a

Endogeic Endogeica Unidentified

OP18

SF

No/18pits

%

No/18pits

%

No/6pits

%

46 153 14

21.6 71.8 6.6

110 5 5

91.7 4.2 4.2

12 10 15

32.4 27.0 40.5

(a)

Pontoscolex corethrurus Dichogaster sp. Others

(b)

Species

Ecological type

mg/18pits

%

mg/18pits

%

mg/6pits

%

Pontoscolex corethrurus Dichogaster sp. Others

Endogeica Endogeica Unidentified

5953 2825 177

66.5 31.5 2.0

15,541 43 64

99.3 0.3 0.4

509 192 173

58.3 21.9 19.8

OP5 = 5-years-old oil palm plantation; OP18 = 18-years-old oil palm plantation; SF = secondary forest. a Endogeic species lives in mineral soil and feeds on soil organic matter (Bouché, 1977).

Table 3 Species composition of termite soldiers and their total number collected in three habitats. Subfamily

Species

Feeding habit

Macrotermitinae

Microtermes sp. Odontotermes sp. 1 Odontotermes sp. 2 Macrotermes gilvus Macrotermes carbonarius Inditermined genus sp.

Wood Wood Wood Wood Wood Wood

Termitinae

Pericapritermes sp. 1 Pericapritermes sp. 2 Subritermes sp.

Soil Soil Soil

Nasutitermitinae

(litter) (litter) (litter) (litter) (litter) (litter)

OP5

OP18

SF

No

%

No

%

No

%

91 109 11 3 0 1

41.4 49.5 5.0 1.4 0.0 0.5

282 9 3 0 0 0

95.6 3.1 1.0 0.0 0.0 0.0

59 15 0 0 2 0

64.8 16.5 0.0 0.0 2.2 0.0

5 0 0

2.3 0.0 0.0

1 0 0

0.3 0.0 0.0

0 3 12

0.0 3.3 13.2

Total number

220/18pits

295/18pits

91/6pits

Simpson’s Da

0.42

0.91

0.46

OP5 = 5-years-old oil palm plantation; OP18 = 18-years-old oil palm plantation; SF = secondary forest. a D = (Rni ⁄ (ni  1))/(N/(N  1)), ni, number of individuals of species i collected; N, total number of individuals collected.

68

M.-K. Wong et al. / Forest Ecology and Management 381 (2016) 63–73

Table 4 Biomass (preserved wet weight) of soil macro-invertebrates (mg m2) (mean ± SE; n = 6). OP5

OP18

SF

p-value in Kruscal- Wallis test

WC

HP

FH

WC

HP

FH

Saprophage Gastropoda Oligochaeta Isopoda Diplopoda Blattariae Isoptera Orthoptera Others 1A

0 8508ab ± 3016 0a 0 0 984ab ± 537 0a 44 ± 43

0 3110ab ± 780 6abc ± 6 7±5 0 856a ± 617 0a 0

0 4966ab ± 1678 204bcd ± 108 58 ± 40 294 ± 204 2327ab ± 666 1ab ± 1 40 ± 30

0 2780ab ± 1319 3abcd ± 2 5±5 0 1475ab ± 1001 4ab ± 4 0

0 13931b ± 3864 1ab ± 1 1±1 49 ± 42 174a ± 101 4ab ± 4 0

2920 ± 2885 12267ab ± 3220 132cd ± 25 21 ± 9 101 ± 67 6905b ± 1525 193ab ± 100 8±8

9±9 1617 a ± 597 322d ± 134 22 ± 11 374 ± 195 2350ab ± 747 332b ± 114 87 ± 87

0.020 0.014 <0.001 0.014 0.003 0.001 0.001 0.053

Zoophage Turbellaria Araneae Chilopoda Carabidae Staphylinidae Formicidae Others 2B

0 33ab ± 22 154 ± 58 0 4±4 250 ± 102 1±1

0 24ab ± 8 41 ± 30 1±1 1±1 251 ± 105 1±1

0 28ab ± 8 210 ± 102 61 ± 60 6±4 398 ± 151 4±4

0 20a ± 10 87 ± 46 0 0 316 ± 104 0

0 98ab ± 56 86 ± 28 0.2 ± 0.2 1±1 167 ± 30 26 ± 26

40 ± 40 177b ± 34 120 ± 98 3±3 12 ± 9 1073 ± 329 18 ± 9

787 ± 715 694ab ± 507 150 ± 59 164 ± 127 58 ± 24 1866 ± 708 97 ± 65

0.023 0.004 0.294 0.510 0.116 0.022 0.020

Polyphage Dermaptera

298ab ± 245

4a ± 4

1064b ± 404

2a ± 2

43ab ± 30

113ab ± 66

66ab ± 61

0.003

Rhizophage Cicadidae L. Scarabaeidae L.

0 0

0 0

0 0

0 0

0 0

0 0

72 ± 72 62 ± 54

0.423 0.056

Others 3C

0a

397b ± 162

0.001

9521ab ± 2233

0.010

Total

32ab ± 20 ab

10275

± 3740

a

4332 ± 880

6ab ± 3

6a ± 6

ab

a

9662

± 2342

4699 ± 1318

22ab ± 19 14604

ab

124ab ± 87

± 3970

b

24230 ± 4007

OP 5 = 5-years-old oil palm plantation; OP18 = 18-years-old oil palm plntation; SF = secondary forest. WC = weeded circle microhabitat; HP = harvest path microhabitat; FH = frond heap microhabitat. a < b < c < d (p < 0.05; Tukey-Kramer test). A Mainly Lepidoptera larvae and Teneblionidae larvae. B Mainly Opilliones and Schizomida. C Unidentified.

3.3. Distribution of biomass of broad taxonomic groups

60

Table 4 shows the mean biomass of soil macro-invertebrates in each of the seven microhabitats. The distribution of all taxa of saprophages, excepting the miscellaneous group ‘‘Others 1”, was significantly affected by the factor ‘‘microhabitat” and/or ‘‘habitat”. Three groups of surface–living litter transformers—isopods, cockroaches, and crickets—showed a common pattern of distribution; their largest biomass was found in SF but were almost absent from the microhabitats WC and HP in both OP5 and OP18. The termite biomass also tended to be smaller in WC and HP devoid of litter than in FH and SF which had litter. However, the termite biomass was not the largest in SF. Fig. 2 illustrates the percentage frequency distribution of termite abundance in the A0-layer in microhabitats with litter layer, SF, OP5 FH, and OP18 FH. The distributions in the three sites overlapped at the lower classes of 0–50 m2. However, the distribution in OP18 FH extended most toward the higher classes, and in SF was confined most to the lower classes. In the three samples from OP18 FH yielding more than 2000 m2, numerous individuals of Microtermes sp. were collected from the inside of thick rachises of the more or less freshly deposited frond litter. The largest abundance and biomass of termites in OP18 FH (Tables 1 and 4) was caused by this hot spot of wood (litter)–feeding termites. The distribution of earthworm biomass was different from those of the taxa mentioned above, with the biomass being smallest in SF, and not necessarily smaller in WC and HP compared to FH in both OP5 and OP18. The biomass of zoophages tended to vary according to habitat rather than microhabitat. It appears that the biomass of all taxa,

50

SF OP5 FH OP18 FH

40

% 30 20 10 0 0

- 10

- 50

- 100

- 500 - 1000 - 2000 - 3000 - 4000

Abundance of termites (N m-2) Fig. 2. Percentage frequency distribution of abundance of termites in the A0-layer. SF = secondary forest, OP5 = 5-years-old oil palm plantation, OP18 = 18-years-old oil palm plantation, FH = frond heap microhabitat.

excluding centipedes, was higher in SF than any of the microhabitats in the oil palm plantations. Omnivorous earwigs differed from all the other groups in that they had the highest biomass in OP5 FH, and were poorly represented in OP18 and SF. Biomass of earwigs in relation to that of termites in OP5 and OP18 is illustrated in Fig. 3. There were no samples yielding large biomasses of both taxa, suggesting they may have different habitat preferences. Thus, the biomass of soil macro-invertebrates was unevenly distributed among the microhabitats as well as the habitats, and the pattern of distribution differed from taxon to taxon.

Biomass of earwigs mg (30 cm)- 2

M.-K. Wong et al. / Forest Ecology and Management 381 (2016) 63–73

0

200

400

600

800

Biomass of termites mg (30

1000

1200

cm)-2

Fig. 3. Relationship between biomass of termites and earwigs in OP5 and OP18. OP5 = 5-years-old oil palm plantation, OP18 = 18-years-old oil palm plantation.

3.4. Characteristics of different microhabitats in relation to the biomass of soil macro-invertebrates Faunal characterization of the seven microhabitats was made using PCA. The results are shown in Fig. 4. PC1 and PC2 explained 59.8% of the total variance. Both PC1 and PC2 were significantly correlated with the biomass of the all taxonomic groups considered except for earwigs (Fig. 4(A)). The biomass of earwigs hardly contributed to PC1 and PC2; the factor loading was 0.241 (P = 0.124) and -0.020 (P = 0.898) on PC1 and PC2, respectively. PC3, which explained an additional 15.9% of the total variance, was almost exclusively loaded with the biomass of earwigs (factor loading: 0.920, P < 0.001). The center of the sample points of OP18 HP ( ), OP5 WC ( ), OP18 WC ( ), and OP5 HP ( ) fell within the positive region on the PC1 axis (Fig. 4(B)). This was mainly because of the absence

(A)

69

of a litter layer in these microhabitats, resulting in poor populations of SLLT and SLP (Table 4). The arrangement of these four microhabitats from upper left (OP18 HP) to lower right (OP5 HP and OP18 WC) was largely brought about by the difference in the mean biomass of earthworms. Positioning of the center of OP18 FH ( ) in the second quadrant was driven by the large biomasses of earthworms and termites and the resulting largest biomass of the whole animals. The center of SF ( ) fell in the third quadrant due to the large biomasses of ants, SLLT, and SLP, together with the small biomass of earthworms. The position of the center of OP5 FH ( ) close to the origin was explained by moderate biomasses of most of the taxonomic groups. If we added the PC3 axis into the ordination, OP5 FH would shift upward along this axis due to the even larger biomass of earwigs than that of any other microhabitats. The above-mentioned ordination of the microhabitats on the PC1-PC2 coordinate plane was statistically significant (P < 0.001, Monte-Carlo test). Thus, the soil macro-invertebrate community differed not only between the secondary forest and the oil palm plantations, but also between the microhabitats within the oil palm plantations. This heterogeneous distribution should be taken into account when comparing the biomass of soil macro-invertebrates in oil palm plantations with those of other ecosystems. 3.5. Comparison of biomass of soil macro-invertebrates among habitats Table 5 shows the weighted mean biomass of relatively wellrepresented 12 taxa and the whole community in the oil palm plantations and the corresponding mean biomass in the secondary forest. Percentage contribution of each taxon to the whole community is also shown. The results of the comparison among habitats varied with taxa. All three taxa from the surface–living litter transformers—isopods, cockroaches, and crickets—were poorly

(B) PC2: 25.7%

PC1: 34.1%

Fig. 4. Results of ordination of microhabitats by a PCA analysis using biomasses of broad taxonomic groups as variables (A) correlation circle and (B) projection of sample points in the plane defined by principal components 1 and 2. W = OP18 WC, w = OP5 WC, H = OP18 HP, h = OP5 HP, F = OP18 FH, f = OP5 FH, S = SF. OP5 = 5-years-old oil palm plantation, OP18 = 18-years-old oil palm plantation, SF = secondary forest, WC = weeded circle microhabitat, HP = harvest path microhabitat, FH = frond heap microhabitat. Ellipses show envelope for 75% of the sample points from a given microhabitat (Monte-Carlo test of zones, 999 replicates, P < 0.001).

70

M.-K. Wong et al. / Forest Ecology and Management 381 (2016) 63–73

Table 5 Biomass of soil macro-invertebrates (preserved wet weight mg m2) in three habitats. OP5 Biomass

OP18 a

SF a

%

Biomass

%

Biomass

%

Saprophage

Oligochaeta Isopoda Diplopoda Isoptera Blattariae Orthoptera

4206 65 22 1310 88 0

64.5 1.0 0.3 20.1 1.4 0.0

12,317 40 8 2324 60 61

74.6 0.2 0.0 14.1 0.4 0.4

1617 322 22 2350 374 332

17.0 3.4 0.2 24.7 3.9 3.5

Zoophage

Araneae Chilopoda Carabidae Staphylinidae Formicidae

26 103 19 3 295

0.4 1.6 0.3 0.0 4.5

114 97 1 4 454

0.7 0.6 0.0 0.0 2.7

694 150 164 58 1866

7.3 1.6 1.7 0.6 19.6

Polyphage

Dermaptera

351

5.4

60

0.4

66

0.7

SLLT

153

2.4

161

1.0

1028

10.8

SLP

124

1.9

102

0.6

372

3.9

Total of 12 taxa

6488

99.4

15,538

94.2

8015

84.2

Whole community

6525

16,501

9521

OP5 = 5-years-old oil palm plantation; OP18 = 18-years-old oil palm plantation; SF = secondary forest; SLLT = surface living litter transformers (Iopoda + Blattariae + Orthoptera). SLP = surface living predators (Cilopda + Crabidae + Staphylinidae). a Indicates weighted mean biomass.

represented in the oil palm plantations, and their biomasses were much smaller than those in the secondary forest. The total biomass of the three surface–living litter transformers (SLLT) did not differ between OP5 and OP18. The termite biomass in OP18 was comparable to that in the SF. While OP5 hosted a substantial population of termites, their biomass was around half that in OP18. The earthworm biomass in OP5 was more than twice that of SF, but approximately three-fold smaller than that in OP18. The biomasses of all of the three taxa of predators (spiders, carabid beetles, and staphylinid beetles) preying mainly on surface–living animals were clearly larger in the secondary forest than in the oil palm plantations. The total biomass of them (SLP) was similar between the two oil palm plantations. The ant biomass exhibited a similar trend. Earwigs had the third largest biomass among all taxa in OP5, but they were poorly represented in OP18 and the SF. The total biomass of the whole community was lowest in OP5, higher in SF, and highest in OP18. In the oil palm plantations, earthworms comprised the most dominant taxon followed by termites. The two taxa together accounted for 85% and 89% of the total biomass in OP5 and OP18, respectively. In SF, termites, ants, and earthworms were co-dominant, accounting together for 61% of the total biomass. As a result, in the secondary forest, the biomass was more evenly distributed among the 12 taxa than in the oil palm plantations. 4. Discussion We present the results of a comparison of the soil macroinvertebrate community in a secondary forest and that in oil palm plantations converted from cocoa plantations. In the studied area, however, destructive operational procedures, including logging, bulldozing, and burning, have been used until recently to convert forests to tree plantations as well as to convert tree plantations of one species to those of another. Therefore, whatever the type of the previous tree vegetation is, the soil macro-invertebrate communities in tree plantations are considered to develop mainly as a result of being brought in with tree seedlings and via colonization from surrounding areas. Thus, the differences between the soil macro-invertebrate communities of the secondary forest and the oil palm plantations observed in this study may be regarded as

resulting from the aforementioned destructive operational procedures as well as from the differences between their present environments, which were caused by the conversion. 4.1. Does conversion of forests to oil palm plantations reduce the biomass of soil macro-invertebrates? A major outcome of the research project on soil macroinvertebrates in the TSBF (Tropical Soil Biology and Fertility Programme) was the identification of three major functional groups, each containing termites, earthworms, and litter arthropods. These functional groups tend to react independently according to ecosystem type and land use practices (Lavelle et al., 1994). Therefore, the question posed in the heading above is examined separately for each of the three functional groups. 4.1.1. Surface–living litter transformers (SLLT) In the oil palm plantations, the biomass of SLLT was smaller in the microhabitats without a litter layer (weeded circle (WC) and harvest path (HP)) than in the microhabitat with a litter layer (frond heap (FH)) (Table 4). Carron et al. (2015), who studied soil macrofauna in a Sumatran oil palm plantation, also reported that the abundance of ‘‘other groups”, which was mainly comprised of litter invertebrates, was significantly lower in the zone without frond heaps than in the zone with them. This localized relationship between frond heaps and SLLT biomass is to be expected given that the litter layer provides food and shelter for these invertebrates. An extensive study on the abundance of soil macroinvertebrates in Colombia recorded the lowest abundance of litter invertebrates in oil palm plantations compared with four other types of land use including natural savannas and rubber plantations (Lavelle et al., 2014). A significantly lower mean abundance and biomass of litter arthropods in oil palm plantations than in primary forests and logged forests was reported in Sabah, Malaysia (Turner and Foster, 2009). Similarly, in this study, the weighted mean biomass of SLLT in the oil palm plantations was more than six times lower than the mean biomass of SLLT in the secondary forest (Table 5). ‘‘Litter transformers” feed on purely organic material by developing an external mutualism with microflora (Lavelle, 1997). As a result, the bulky, upper parts of the large frond piles

M.-K. Wong et al. / Forest Ecology and Management 381 (2016) 63–73

that are isolated from the ground surface, and are relatively dry, may not provide litter transformers with sufficient food. In other words, in oil palm plantations, even though there may be a large potential food resource, the amount of food actually available at any point in time may be limited compared with forests, where a continuous litter layer covers most of the ground surface. Therefore, this study concludes that the biomass of litter transformers is lower in oil palm plantations than in forests because of the management practice that places fronds in piles at the time of fruit harvest. 4.1.2. Termites Along with earthworms, termites are the most important component of the ‘‘ecosystem engineers” (Lavelle, 1997; Jouquet et al., 2016). In contrast to litter transformers, wood (litter)–feeding termites are capable of feeding on fresh and dry litter by developing a symbiosis with microorganisms in their gut, enabling them to utilize the whole of the frond pile. Among five types of land use surveyed by Lavelle et al. (2014), oil palm plantations yielded the second highest mean abundance of termites, exceeding that of rubber plantations and even that of natural savannas, one of the major habitats for termites. In the present study, the weighted mean biomass of termites in OP18 was similar to the mean biomass of termites in the secondary forest (Table 5). OP5 also accommodated a sizable population of termites, although the weighted mean biomass of this population was about half that of OP18. These substantial populations of termites in OP5 and OP18 are considered to consist largely of wood (litter)–feeding species of the Macrotermitinae, since the soldier community in these sites was almost completely dominated by species from this subfamily (Table 3) and the soldier/worker ratio of Macrotermitinae species is not necessarily higher than those of the other subfamilies found in the present study (Termitinae and Nastitermitinae) (Haverty, 1977). In general, wood–feeding termites, which have larger and highly sclerotized bodies, are less vulnerable to forest disturbance or canopy opening than smaller and more fragile soil–feeding termites (De Souza and Brown, 1994; Eggleton et al., 1996; Bandeira and Vasconcellos, 2002; Jones et al., 2003). Eggleton et al. (1996) found a significant positive correlation between the abundance of wood–feeders and the amount of available large logs: wood–feeders were more abundant in a young plantation with a larger volume of dead wood comprised of the cut or poisoned trees that previously grew on the site, than in an old secondary forest and a nearby primary forest. The increased number of encounters of wood (litter)–feeding termites with the increase in the amount of available food was also observed in teak plantations (Attington et al., 2005) and fragmented forests (Davies, 2002) compared with undisturbed forests, notwithstanding the harsher micro-climatic conditions in the former. Thus, a plentiful food supply in the form of the thick rachises of fresh fronds pruned at the time of fruit harvesting would be enough to sustain populations of wood (litter)–feeding termites no smaller than those found in forests. Frond inputs in immature oil palm plantations increase linearly up to eight to ten years of planting age (Haron et al., 1998). Therefore, the amount of food supply for wood (litter)–feeding termites might have been smaller in OP5 than in OP18. A parallel difference was found in termite biomass between the two sites (Table 5). However, the difference in termite biomass between OP5 and OP18 was caused by the presence or absence of the frond heap samples with extremely great number of individuals, the so-called ‘‘termite hot spots” (Fig. 2). The largest three samples from OP18 FH were due to the thick and more or less fresh rachises colonized by an army of individuals of Microtermes sp. as Nastitermes sp. observed by Kon et al. (2012). On the other hand, among the samples with poor termite populations, aged pieces of

71

rachis packed with mineral soil from the termite activities were often observed in both sites. Thus, it is probable that in oil palm plantations with an intermittent and concentrated supply of nonsenescent fronds, the abundance of wood (litter)–feeding termites greatly fluctuates with the age of the frond litter. In this context, partly incompatible occurrence of termites and earwigs (Fig. 3) may be worth noting. While the biomass of termites was three fold larger in OP18 FH than in OP5 FH, the biomass of earwigs was nearly ten times larger in OP5 FH than in OP18 FH (Table 4). Earwigs are gregarious and feed on fully decayed plant materials, carcasses and fungal hyphae (Aoki, 1973). Therefore, incompatible occurrence of wood (litter)–feeding termites and earwigs could be caused by segregation in food use rather than by an exclusive interaction between them. Therefore, this study concludes that wood (litter)–feeding termite populations can be largely maintained in oil palm plantations under the standard management system due to the large input of fronds, even though termite populations may vary with the age of the trees and the frond piles. In contrast, as observed in OP5 and OP18 (Table 3), populations of vulnerable soil–feeding termites may become depauperate in oil palm plantations, like those in teak plantations (Attington et al., 2005) and in fragmented forests (Davies, 2002) compared with those in undisturbed forests, as a consequence of site destruction during conversion from forests, and remained depauperate due to unfavorable micro-climatic conditions in oil palm plantations compared with those in forests (Luskin and Potts, 2011).

4.1.3. Earthworms Biomass data is indispensable for evaluating the functional role of soil macro-invertebrates, especially earthworms, which vary greatly in absolute body weight among species. For example, in the present study, the preserved wet weight of the individuals of Pontoscolex corethrurus ranged from 15.2 to 535.4 mg, whereas the range for Dichogaster sp. was 1.5–54.3 mg. As a result, while the abundance of earthworms (No. m1) for the three habitats decreased in the following order from highest to lowest: OP5 (119), OP18 (87), and SF (69), biomass (mg m1) followed a different pattern across the three habitats, decreasing in the following order from highest to lowest: OP18 (12,317), OP5 (4206), and SF (1617). Despite the importance of biomass measures, in the two most recent publications that provide quantitative data on soil macro-invertebrates in oil palm plantations (Lavelle et al., 2014; Carron et al., 2015), the biomass of earthworms is not available. Available data on macrofaunal communities from tropical rainforests, grasslands, and cropland was compiled in the TSBF (Lavelle et al., 1994). The results showed that the mean biomass of earthworms was clearly larger in tree plantations such as palm and cocoa than in natural forests due to the disturbance and/or establishment of an herbaceous stratum which allows some exotic species to colonize. In this study, the biomass of earthworms was also larger in the oil palm plantations than the secondary forest (Table 5). The two dominant species, P. corethrurus and Dichogaster sp., were both exotics (Fragoso et al., 1999a). P. corethrurus is a cosmopolitan species with a tolerance for a wide range of environmental conditions (Fragoso et al., 1999a), and it is often favored by disturbance, predominating deforested sites where the survival of native earthworms is difficult (Fragoso et al., 1997, 1999b; Marichal et al., 2010). This species was also reported to be the only species occurring in three oil palm plantations of different ages, in each of five sites with differing soil types (i.e. a total of 15 oil palm plantation sites) in Malaysia (Sabrina et al., 2009). Thus, oil palm plantations can accommodate sizable populations of earthworms that are similar, or larger, than those found in forests, owing to the colonization of exotic species that are favored by disturbance.

72

M.-K. Wong et al. / Forest Ecology and Management 381 (2016) 63–73

4.2. Implications of altering soil macro-invertebrates for decomposition It has been confirmed in many studies that the organic carbon content of the soil under frond heaps is higher than that of the soil along a harvest path (Haron et al., 1998; Law et al., 2009; Frazao et al., 2013; Tan et al., 2014; Yusuyin et al., 2015). However, based on the fact that the organic carbon concentrations of the soil under frond piles did not reflect the large cumulative inputs of organic matter provided by the frond piles, Haron et al. (1998) suggested that the decomposition of this material is largely taking place on the soil surface. This is true for OP18 in the present study. Yusuyin et al. (2015) investigated soil nutrients in the three microhabitats and nutrients contained in heaped fronds at the OP18. They found that the total carbon content of the soil at the frond heap was very low compared with the level that would be expected given the huge carbon input to the soil through the frond heaping practice. This suggests rapid decomposition of organic matter in the frond heap and the underlying soils. These studies suggest that litter decomposition is being sustained in oil palm plantations via the interaction of soil macrofauna and microorganisms, despite the uneven litter distribution and harsher microclimatic conditions at these sites. Populations of surface–living litter transformers may be reduced in oil palm plantations, which has a negative effect on mineralization of the litter by free–living microorganisms through a reduced service of litter fragmentation. However, litter fragmentation is not monopolized by surface–living litter transformers, but is efficiently executed by wood (litter)–feeding termites (Matsumoto and Abe, 1979; Collins, 1981). In the oil palm plantations studied, the termite assemblage was almost completely dominated by Macrotermitinae species. They promote mineralization of litter by free–living microorganisms not only through fragmentation but also through transportation of mineral soil into the interior of rachises. Symbiotic microorganisms in their gut and fungus comb also accelerate mineralization. It is likely that populations of wood (litter)–feeding termites in oil palm plantations are maintained by the large input of thick frond rachises, at least at a level that compensates for the reduced populations of surface–living litter transformers, as observed in the Macrotermes gilvus population studied by Foster et al. (2011). In addition, a possibility of further processing of the aged litter by earwigs, usually a minor component of the soil fauna in natural forests, was shown in OP5, although the generality of it should be examined in a more comprehensive survey. A sizable population of cosmopolitan earthworm species, such as Pontoscolex corethrurus favored by disturbance may also indirectly contribute to microbial decomposition by burying the litter under the casts it deposits on the ground surface (Topoliantz and Ponge, 2005; Sabrina et al., 2009). 5. Conclusion Soil macro-invertebrate assemblages in oil palm plantations in this study were similar with those observed in other oil palm plantation study sites. We conclude that conversion of forests to oil palm plantations may reduce the diversity of soil macroinvertebrate assemblages, increase patchiness in their distribution, and may suppress some functional groups. However, it does not appear that these changes have led to a decline in decomposition, due to colonization by other groups of decomposers that are favored by disturbance and/or the management practices applied in oil palm plantations. Considering the differences in the biomasses of earthworms and termites between the two oil palm plantations, of different ages but under the same climatic conditions and on the same soil type, diachronic studies on soil macro fauna in oil palm plantations merit further investigation.

Acknowledgements The authors sincerely grateful to Dr. Masamichi Ito, Surugadai University, for carrying out species identification of earthworms. The authors gratefully acknowledge to the financial support by the Sumitomo Foundation, Japan (No. 113051). The authors also thank Felda Agricultural Services and Felda Global Ventures Research and Development for their permission and assistance in conducting the field survey. References Andren, O., Bengtssan, J., Clarholm, M., 1995. Biodiversity and species redundancy among litter decomposers. In: Collins, H.P., Robertson, G.P., Klug, M.J. (Eds.), The Significance and Regulation of Soil Biodiversity. Kluwer Academic Publishers, Dordrecht, pp. 141–151. Aoki, J., 1973. Soil Zoology. Hokuryukan, Tokyo (in Japanese). Attington, S.E., Lachat, T., Sinsin, B., Nagel, P., Peveling, R., 2005. Termite assemblages in a West-African semi-deciduous forest and teak plantations. Agric. Ecosyst. Environ. 110, 318–326. Bandeira, A.G., Vasconcellos, A., 2002. A quantitative survey of termites in a gradient of disturbed highland forest in northeastern Brazil (Isoptera). Sociobiology 39, 429–439. Barnes, A.D., Jochum, M., Mumme, S., Haneda, N.F., Farajallah, A., Widarto, T.H., Brose, U., 2014. Consequences of tropical land use for multitrophic biodiversity and ecosystem functioning. Nat. Commun. 5, 5351. Blakemore, R.J., 2002. Cosmopolitan Earthworms. An Eco-taxonomic Guide to the Peregrine Species of the World, VermEcology, Australia. Bouché, M.B., 1977. Stratégies Lombriciennes. In: Lohm, U., Persson, T. (Eds.), Soil Organisms as Components of Ecosystems. NFR, Stockholm, pp. 122–132. Carron, M.P., Auriac, Q., Snoeck, D., Villenave, C., Blanchart, E., Ribeyre, F., Marichal, R., Darminto, M., Caliman, J.P., 2015. Spatial heterogeneity of soil quality around mature oil palms receiving mineral fertilization. Eur. J. Soil Biol. 66, 24–31. Collins, N.M., 1981. The role of termites in the decomposition of wood and leaf litter in the Southern Guinea Savanna of Nigeria. Oecologia 51, 389–399. Davies, R.G., 2002. Feeding group responses of a Neotropical termite assemblage to rain forest fragmentation. Oecologia 133, 233–242. De Souza, O.F.F., Brown, V.K., 1994. Effects of fragmentation on Amazonian termite communities. J. Trop. Ecol. 10, 197–206. Eggleton, P., Bignell, D.E., Sands, W.A., Mawdsley, N.A., Lawton, J.H., Wood, T.G., Bignell, N.C., 1996. The diversity, abundance and biomass of termites under differing levels of disturbance in the Mbalmayo Forest Reserve, southern Cameroon. Philos. Trans. R. Soc. Lond. B 351, 51–68. Fitzherbert, E.B., Struebig, M.J., Morel, A., Danielsen, F., Brühl, C.A., Donald, P.F., Hhalan, B., 2008. How will oil palm expansion affect biodiversity? Trends Ecol. Evol. 23, 538–545. Foster, W.A., Snaddon, J.L., Turner, E.C., Fayle, T.M., Cockerill, T.D., Ellwood, M.D.F., Broad, G.R., Chung, A.Y.C., Eggleton, P., Khen, C.V., Yusah, K.M., 2011. Establishing the evidence base for maintaining biodiversity and ecosystem function in the oil palm landscapes of South East Asia. Philos. Trans. R. Soc. Lond. B 366, 3277–3291. Fragoso, C., Brown, G., Partón, J.C., Blanchart, E., Lavelle, P., Pashanasi, B., Senapati, B. K., Kumar, T., 1997. Agricultural intensification, soil biodiversity and agroecosystem function in the tropics: the role of earthworms. Appl. Soil Ecol. 6, 17–35. Fragoso, C., Kanyonyo, J., Moreno, A., Senapati, B.K., Blanchart, E., Rodriguez, C., 1999a. A survey of tropical earthworms: taxonomy, biogeography and environmental plasticity. In: Lavelle, P., Brussaard, L., Hendrix, P. (Eds.), Earthworm Management in Tropical Agroecosystems. CABI Publishing, Oxon, pp. 1–26. Fragoso, C., Lavelle, P., Blanchart, E., Senapati, B.K., Jimenez, J.J., Martinez, M.A., Decaëns, T., Tondoh, J., 1999b. Earthworm communities of tropical agroecosystems: origin, structure and influence of management practices. In: Lavelle, P., Brussaard, L., Hendrix, P. (Eds.), Earthworm Management in Tropical Agroecosystems. CABI Publishing, Oxon, pp. 27–55. Frazao, L.A., Paustian, K., Cerri, C.E.P., Cerri, C.C., 2013. Soil carbon stocks and changes after oil palm introduction in the Brazilian Amazon. Glob. Change Biol. Bioenergy 5, 384–390. Haron, K., Brookes, P.C., Anderson, J.M., Zakaria, Z.Z., 1998. Microbial biomass and soil organic matter dynamics in oil palm (Elaeis guineensis Jacq.) plantations, West Malaysia. Soil Biol. Biochem. 30, 537–552. Hassall, M., Jones, D.T., Taiti, S., Latipi, Z., Sutton, S.L., Mohammed, M., 2006. Biodiversity and abundance of terrestrial isopods along a gradient of disturbance in Sabah, East Malaysia. Eur. J. Soil Biol. 42, 197–207. Haverty, M.I., 1977. The proportion of soldiers in termite colonies: a list and a bibliography (Isoptera). Sociobiology 2, 199–216. Jones, D.T., Susilo, F.X., Bignell, D.E., Hardiwinoto, S., Gillison, A.N., Eggleton, P.E., 2003. Termite assemblage collapse along a land-use intensification gradient in lowland central Sumatra, Indonesia. J. Appl. Ecol. 40, 380–391. Jouquet, P., Bottinelli, N., Shanbhag, R.R., Bourguignon, T., Traore, S., Abbasi, S.A., 2016. Termites: the neglected soil engineers of tropical soils. Soil Sci. 181, 157– 165.

M.-K. Wong et al. / Forest Ecology and Management 381 (2016) 63–73 Kon, T.W., Bong, C.F.J., King, J.H.P., Leon, C.H.S., 2012. Biodiversity of termite (Insecta: Isoptera) in tropical peat land cultivated with oil palm. Pak. J. Biol. Sci. 15, 108–120. Kurokawa, H., Nakashizuka, T., 2008. Leaf herbivory and decomposability in a Malaysian tropical rain forest. Ecology 89, 2645–2656. Lamade, E., Djegui, N., Leterme, P., 1996. Estimation of carbon allocation to the roots from soil respiration measurements of oil palm. Plant Soil 181, 329–339. Lavelle, P., 1997. Faunal activities and soil processes: adaptive strategies that determine ecosystem function. Adv. Ecol. Res. 24, 93–132. Lavelle, P., Pashanasi, B., 1989. Soil macrofauna and land management in Peruvian Amazonia (Yurimaguas, Loreto). Pedobiologia 33, 283–291. Lavelle, P., Dangerfield, M., Fragoso, C., Ecshenbrenner, V., Lopez-Hernandez, D., Pashanasi, B., Brussaard, L., 1994. The relationship between soil macrofauna and tropical soil fertility. In: Woomer, P.L., Swift, M.J. (Eds.), The Biological Management of Tropical Soil Fertility. John Wiley & Sons Ltd., West Sussex, pp. 137–169. Lavelle, P., Rodriguez, N., Arguello, O., Bernal, J., Botero, C., Chaparro, P., Gomez, Y., Guitierrez, A., Hurtado, M.P., Loaiza, C., Pullido, S.X., Rodriguez, E., Sanabria, C., Velasquez, E., Fonte, S.J., 2014. Soil ecosystem services and land use in the rapidly changing Orinoco River Basin of Colombia. Agric., Ecosyst. Environ. 185, 106–117. Law, M.C., Balasundram, S.K., Husni, M.H.A., Ahmed, O.H., Harun, M.H., 2009. Spatial variability of soil organic carbon in oil palm. Int. J. Soil Sci. 4, 93–103. Luskin, M.S., Potts, M.D., 2011. Microclimate and habitat heterogeneity through the oil palm life cycle. Basic Appl. Ecol. 12, 540–551. Marichal, R., Feijoo, M.A., Praxedes, C., Ruiz, D., Carjaval, A.F., Oszwald, J., del Pilar Hurtado, M., Brown, G.G., Grimaldi, M., Desjardins, T., Sarrazin, M., Decaens, T., Velasquez, E., Lavelle, P., 2010. Invasion of Pontoscolex corethrurus (Glossoscolecidae, Oligochaeta) in landscapes of the Amazonian deforestation arc. Appl. Soil Ecol. 46, 443–449. Matsumoto, T., Abe, T., 1979. The role of termites in an equatorial rain forest Ecosystem of West Malaysia. II. Leaf litter consumption on the forest floor. Oecologia 38, 261–274. Paramananthan, S., 2000. Soils of Malaysia: Their Characteristics and Identification, vol. 1. Academy of Sciences Malaysia, Kuala Lumpur. Pellens, R., Garay, L., 1999. Edaphic macroarthopod communities in fast-growing plantations of Eucalyptus grandis Hill ex Maid (Myrtaceae) and Acacia mangium Wild (Leguminosae) in Brazil. Eur. J. Soil Biol. 35, 77–89. R Development Core Team, 2004. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna. RSPO Principles and Criteria Review 2011/2012 – Summary From 2nd Public Consultation. .

73

Sabrina, D.T., Hanafi, M.M., Nor Azwady, A.A., Mahmud, T.M.M., 2009. Earthworm populations and cast properties in the soils of oil palm plantations. Malay. J. Soil Sci. 13, 29–42. Savilaakso, S., Garcia, C., Garcia-Ulloa, J., Ghazou, J., Groom, M., Guariguata, M.R., Laumonier, Y., Nasi, R., Petrokofsky, G., Snaddon, J., Zrust, M., 2014. Systematic review of effects on biodiversity from oil palm production. Environ. Evid. 3, 4. Singh, M., Bhagwat, S.A., 2013. Tropical agricultural production, conservation and carbon sequesteration conflicts: oil palm expansion in South East Asia. In: Zhen, F. (Ed.), Biofuels – Economy, Environment and Sustainability. Intech, Rijeka, pp. 39–71. Sodhi, N.S., Posa, M.R.C., Lee, T.M., Bickford, D., Koh, L.P., Brook, B.W., 2010. The state and conservation of Southeast Asian biodiversity. Biodivres. Conserv. 19, 317– 328. Tan, N.P., Wong, M.K., Yusuyin, Y., Abdu, A.B., Iwasaki, K., Tanaka, S., 2014. Soil characteristics in an oil palm field, Central Pahang, Malayia with special reference to micro sites under different managements and slope positions. Trop. Agric. Develop. 58, 146–154. Tan, N.P., Wong, M.K., Yusuyin, Y., Abdu, A.B., Iwasaki, K., Tanaka, S., 2015. Distribution of soil phosphorus fractions influenced by chemical fertilizer application and frond heaping practices in an oil palm plantation in Pahang, Peninsular Malaysia. Pedologist 59, 12–24. Thioulouse, J., Dray, S., 2007. Interactive multivariate data analysis in R with the ade4 and ade4TkGUI packages. J. Stat. Softw. 22, 1–14. Tho, Y.P., 1992. Termites of Peninsular Malaysia, Malayan Forest Records No. 36. Forest Research Institute Malaysia, Kuala Lumpur. Topoliantz, S., Ponge, J.F., 2005. Charcoal consumption and casting activity by Pontoscolex corethrurus (Glossoscolecidae). Appl. Soil Ecol. 28, 217–224. Tsukamoto, J., Sabang, J., 2005. Soil macro-fauna in an Acacia mangium plantation in comparison to that in a primary mixed dipterocarp forest in the lowlands of Sarawak, Malaysia. Pedobiologia 49, 69–80. Turner, E.C., Foster, W.A., 2009. The impact of forest conversion to oil palm on arthropod abundance and biomass in Sabah, Malaysia. J. Trop. Ecol. 25, 23–30. Turner, E.C., Snaddon, J.L., Ewers, R.M., Fayle, T.M., Foster, W.A., 2011. The impact of oil palm expansion on environmental change: putting conservation research in context. In: Bernardes, M.A.D.S. (Ed.), Environmental Impact of Biofuels. InTech, Rijeka, pp. 19–40. Yusuyin, Y., Tan, N.P., Wong, M.K., Abudu, A.B., Iwaski, K., Tanaka, S., 2015. Nutrient Status of frond heaps and the underlying soils at an 18-year-old oil palm field in Central Pahang, Malaysia. Trop. Agric. Develop. 59, 212–220. Yusuyin, Y., Tan, N.P., Wong, M.K., Abudu, A.B., Iwaski, K., Tanaka, S., 2016. Chemical Forms and Distribution of Soil Micronutrients at an 18-Year-Old Oil Palm Field in Central Pahang. Tropical Agriculture Development, Malaysia (in press).