Plant and bird diversity in rubber agroforests in the lowlands of Sumatra, Indonesia

Terrestrial pteridophytes as indicators of a forest-like environment in rubber production systems in the lowlands of Jambi, Sumatra

Protists, abundant but enigmatic single-celled eukaryotes, are important soil microbiota providing numerous ecosystem functions. We employed high-throughput sequencing of environmental DNA, targeting the V4 region of the 18S rRNA gene, to characterize changes in their abundance, species richness, and community structure with conversion of lowland rainforest into rubber agroforest (jungle rubber), and rubber and oil palm plantations; typical agricultural systems in Sumatra, Indonesia. We identified 5,204 operational taxonomic units (OTUs) at 97% identity threshold of protists from 32 sites. Protists species richness was similar in rainforest, jungle rubber and oil palm plantations but significantly lower in rubber plantations. After standardization, 4,219 OTUs were assigned to five trophic groups, and inspected for effects of land-use change, and potential biotic and abiotic driving factors. The most abundant trophic group was phagotrophs (52%), followed by animal parasites (29%), photoautotrophs (12%), plant parasites (1%), and symbionts (<1%). However, the relative abundance and OTU richness of phagotrophs and photoautotrophs increased significantly with increasing land-use intensity. This was similar, but less pronounced, for the relative abundance of symbionts. Animal and plant parasites decreased significantly in abundance and species richness with increasing land-use intensity. Community compositions and factors affecting the structure of individual trophic groups differed between land-use systems. Parasites were presumably mainly driven by the abundance and species richness of their hosts, while phagotrophs by changes in soil pH and increase in Gram-positive bacteria, and photoautotrophs by light availability. Overall, the results show that relative species richness, relative abundance, and community composition of individual trophic groups of protists in tropical lowland rainforest significantly differ from that in converted ecosystems. This is likely associated with changes in ecosystem functioning. The study provides novel insight into protist communities and their changes with land-use intensity in tropical lowland ecosystems. We show, that trophic groups of protists are powerful indicators reflecting changes in the functioning of ecosystems with conversion of rainforest into monoculture plantations.

Rapid land-use change in the tropics causes dramatic losses in biodiversity and associated functions. In Sumatra, Indonesia, lowland rainforest has mainly been transformed by smallholders into oil palm (Elaeis guineensis) and rubber (Hevea brasiliensis) monocultures, interspersed with jungle rubber (rubber agroforests) and a few forest remnants. In two regions of the Jambi province, we conducted point counts in 32 plots of four different land-use types (lowland rainforest, jungle rubber, rubber plantation and oil palm plantation) as well as in 16 nearby homegardens, representing a small-scale, traditional agricultural system. We analysed total bird abundance and bird abundance in feeding guilds, as well as species richness per point count visit, per plot, and per land-use system, to unveil the conservation importance and functional responses of birds in the different land-use types. In total, we identified 71 species from 24 families. Across the different land-use types, abundance did not significantly differ, but both species richness per visit and per plot were reduced in plantations. Feeding guild abundances between land-use types were variable, but homegardens were dominated by omnivores and granivores, and frugivorous birds were absent from monoculture rubber and oil palm. Jungle rubber played an important role in harbouring forest bird species and frugivores. Homegardens turned out to be of minor importance for conserving birds due to their low sizes, although collectively, they are used by many bird species. Changes in functional composition with land-use conversion may affect important ecosystem functions such as biological pest control, pollination, and seed dispersal. In conclusion, maintaining forest cover, including degraded forest and jungle rubber, is of utmost importance to the conservation of functional and taxonomic bird diversity.

BackgroundIn the last decades, Southeast Asia has experienced massive conversion of rainforest into rubber and oil palm monoculture plantations. The effects of this land-use change on canopy arthropods are still largely unknown. Arboreal Collembola are among the most abundant canopy arthropods in tropical forests, potentially forming a major component of the canopy food web by contributing to the decomposition of arboreal litter and being an important prey for canopy arthropod predators. We investigated abundance, richness, and community composition of, as well as the influence of a series of environmental factors on, canopy Collembola communities in four land-use systems in Jambi Province, Sumatra, Indonesia: (1) lowland rainforest, (2) jungle rubber (rubber agroforest), and monoculture plantations of (3) rubber and (4) oil palm.ResultsUsing canopy fogging in 32 research plots in both the dry and rainy seasons in 2013, we collected 77,104 specimens belonging to 68 (morpho) species. Generally, Collembola communities were dominated by few species including two species of the genus Salina (Paronellidae; 34% of total individuals) and two species of Lepidocyrtinae (Entomobryidae; 20%). The abundance of Collembola in lowland rainforest (53.4 ± 30.7 ind. m−2) was more than five times higher than in rubber plantations, and more than ten times higher than in oil palm plantations; abundances in jungle rubber were intermediate. Collembola species richness was highest in rainforest (18.06 ± 3.60 species) and jungle rubber (16.88 ± 2.33 species), more than twice that in rubber or oil palm. Collembola community composition was similar in rainforest and jungle rubber, but different from monoculture plantations which had similar Collembola community composition to each other. The environmental factors governing community composition differed between the land-use systems and varied between seasons.ConclusionsOverall, this is the first in-depth report on the structure of arboreal Collembola communities in lowland rainforest and agricultural replacement systems in Southeast Asia. The results highlight the potentially major consequences of land-use change for the functioning of arboreal arthropod food webs.

Tropical landscapes are not only rapidly transformed by ongoing land-use change, but are additionally confronted by increasing seasonal climate variation. There is an increasing demand for studies analyzing the effects and feedbacks on ecosystem functioning of large-scale conversions of tropical natural forest into intensively managed cash crop agriculture. We analyzed the seasonality of aboveground litterfall, fine root litter production, and aboveground woody biomass production (ANPP(woody)) in natural lowland forests, rubber agroforests under natural tree cover ("jungle rubber"), rubber and oil palm monocultures along a forest-to-agriculture transformation gradient in Sumatra. We hypothesized that the temporal fluctuation of litter production increases with increasing land-use intensity, while the associated nutrient fluxes and nutrient use efficiency (NUE) decrease. Indeed, the seasonal variation of aboveground litter production and ANPP(woody) increased from the natural forest to the plantations, while aboveground litterfall generally decreased. Nutrient return through aboveground litter was mostly highest in the natural forest; however, it was significantly lower only in rubber plantations. NUE of N, P and K was lowest in the oil palm plantations, with natural forest and the rubber systems showing comparably high values. Root litter production was generally lower than leaf litter production in all systems, while the root-to-leaf ratio of litter C flux increased along the land-use intensity gradient. Our results suggest that nutrient and C cycles are more directly affected by climate seasonality in species-poor agricultural systems than in species-rich forests, and therefore might be more susceptible to inter-annual climate fluctuation and climate change.

Land cover has changed dramatically in Sumatra Island, Indonesia over the last decades. Rampant deforestation has drawn a lot of attention due to the potential global impact of the associated carbon stock loss on climate warming and the erosion of biodiversity. The various land uses which replace natural forest are not equally benign to the environment. Rubber agroforests (jungle rubber) are extensive traditional cropping systems. They have been singled out by previous studies as the best land use option for biodiversity conservation once forest is cleared, while allowing farmers to make a living from the deforested land. But how sustainable are complex agroforestry systems themselves? Are they not just a transient stage in the overall process of land use intensification? We studied land cover change in the Bungo district, in Jambi, Sumatra (Indonesia), a 4,550 km2 area. Large forest tracks have been cleared since the early seventies and replaced by rubber plantations, oil palm plantations and other agricultural land-uses. Landsat images taken between 1973 and 2005 were used to quantify the trends of land cover changes in the area. during that period forest cover fell from more than 75% to 30%. Simultaneously monoculture plantations increased from 3% to over 40%, while rubber agroforests, decreased from 15% to 11%. Strikingly most of the rubber agroforests present in 2005 where absent in 1973 while most of the rubber agroforests present in 1973 had been replaced by more intensive agricultural systems by 2005. Rubber agroforests are now the ultimate reservoir of the original lowland forest biodiversity since natural forest has almost completely disappeared from the peneplain. They are however under growing pressure themselves and have incurred an accelerated conversion rate to more intensive agricultural systems in the period 2002–2005.

Natural forests in South-East Asia have been extensively converted into other land-use systems in the past decades and still show high deforestation rates. Historically, lowland forests have been converted into rubber forests, but more recently, the dominant conversion is into oil palm plantations. While it is expected that the large-scale conversion has strong effects on the carbon cycle, detailed studies quantifying carbon pools and total net primary production (NPPtotal ) in above- and belowground tree biomass in land-use systems replacing rainforest (incl. oil palm plantations) are rare so far. We measured above- and belowground carbon pools in tree biomass together with NPPtotal in natural old-growth forests, 'jungle rubber' agroforests under natural tree cover, and rubber and oil palm monocultures in Sumatra. In total, 32 stands (eight plot replicates per land-use system) were studied in two different regions. Total tree biomass in the natural forest (mean: 384Mgha(-1) ) was more than two times higher than in jungle rubber stands (147Mgha(-1) ) and >four times higher than in monoculture rubber and oil palm plantations (78 and 50Mgha(-1) ). NPPtotal was higher in the natural forest (24Mgha(-1) yr(-1) ) than in the rubber systems (20 and 15Mgha(-1) yr(-1) ), but was highest in the oil palm system (33Mgha(-1) yr(-1) ) due to very high fruit production (15-20Mgha(-1) yr(-1) ). NPPtotal was dominated in all systems by aboveground production, but belowground productivity was significantly higher in the natural forest and jungle rubber than in plantations. We conclude that conversion of natural lowland forest into different agricultural systems leads to a strong reduction not only in the biomass carbon pool (up to 166MgCha(-1) ) but also in carbon sequestration as carbon residence time (i.e. biomass-C:NPP-C) was 3-10 times higher in the natural forest than in rubber and oil palm plantations.

&lt;p&gt;Birds play a pivotal role in ecosystems, but in disturbed areas their role may be limited due to changes in their habitat. &nbsp;In Simalungun and Asahan districts, North Sumatra, bird communities were studied in a range of habitats: natural forest, rubber agroforests (RAF), rubber monoculture plantations (RMP) and emplacement areas. The birds were observed using descriptive survey methods by implementing a quick biodiversity survey, with data collected along a 1km transect.&nbsp; In total, 142 species of birds from 42 families were found in the four habitats. Natural forests held the highest diversity of bird species, followed by rubber agroforests, emplacement areas and rubber plantations, with a Shannon-Wiener index of 4.5, 3.6, 3.6 and 3, respectively. Regarding the IUCN red list species, 12 bird species of near-threatened status and 2 species of vulnerable status were recorded.&nbsp; Based on CITES categories, one species was listed in the Appendix criteria I, 12 species were classified in Appendix II and 26 bird species were protected under Indonesian regulations. Changes in the structure and composition of vegetation in disturbed forests and cleared land determined the richness of bird species. The different tree compositions in the three habitats of the rubber estate plantations and surrounding areas influenced the number of bird species, their diversity and species composition.&lt;/p&gt;

Globally rising demand for cash crops such as oil palm, rubber, and cacao is driving the extensive conversion of natural rainforests in Southeast Asia with increasing speed. The consequences for humans and nature are immense and until today not thoroughly investigated. In the present thesis the impact of natural rainforest conversion into intensively managed rubber and oil palm monocultures, and smallholder rubber and cacao agroforests on carbon storage and sequestration, litter dynamics and hydraulic anatomic properties in Indonesia was assessed and quantified. To fulfill the objectives, carbon and nutrient dynamics in phytomass of lowland rainforest transformation systems in two regions of Jambi Province on Sumatra were studied, while estimators of tree productivity and hydraulic anatomy patterns were investigated in cacao agroforests in Central Sulawesi. The transformation of rainforest in our study area clearly resulted in significantly lower total carbon stocks in all investigated land-use types. According to our estimations, natural forests in this region store three times more carbon in biomass than even the extensively utilized jungle rubber system. The total carbon pools comprising aboveground tree biomass, dead wood debris as well as coarse and fine roots in rubber and oil palm monocultures consist of even less than 20 % of the natural forest C stocks they are replacing. Thus, compensating for the emitted carbon caused by land-use change on mineral soils may take hundreds of years or even be irreversible. While oil palm monocultures showed the lowest carbon stock, we have measured the highest total net primary production (NPP) values in oil palm plantations followed by natural forest, jungle rubber and rubber monocultures. However, the carbon sequestration potential is reduced by forest conversion in the long term, as removal of major NPP fractions namely oil palm fruits, rubber latex and wood occurs. Since we found the carbon residence time in biomass to be longer in natural forest and jungle rubber, the establishment of oil palm or rubber plantations on former forest land does not enhance carbon sequestration potential despite the high total net primary production of oil palms. With predicted climate change associated with increasing mean temperatures and drought events, land-use intensification with introduced cash-crops is also expected to have influence on seasonal growth responses in moist tropical lowlands even with only moderately dry seasons. We found a higher coefficient of variance (CV) following forest transformation for all examined components namely aboveground litterfall, fine root mortality, and aboveground woody production. In addition to changes in the annual net primary production, element return to the soil via aboveground litterfall was significantly reduced for C, N, Ca, Fe, Mg, Mn and S in monoculture plantations, particularly for rubber. Due to crop removal, a high demand for fertilizer use was caused, which likely leaded to significantly lower nutrient use efficiencies (NUE) for all major nutrients in oil palm plantations. Decomposition rate was reduced in oil palm plantations compared to natural forest which further decelerated nutrient cycling. Beside nutrient availability, a permanent water supply from roots to leaves is essential for effective plant growth. We found wood anatomical and derived hydraulic properties to be a good predictor for tree stem growth performance in the study on hydraulic architecture of the root, stem and branch wood in Theobroma cacao and five common shade tree species in agroforestry systems on Sulawesi. In contrast, neither wood density, nor empirically measured branch and root hydraulic conductivity, foliar delta 13C or foliar nitrogen content were good predictors for aboveground growth performance. A similar relationship was found between basal area increment and theoretical hydraulic conductivity in forest and rubber trees on Sumatra. Several wood anatomical patterns were observed to potentially follow drought adaptations strategies based on biogeographic origin. Hence, our results imply that future research on conceptual trade-offs of tree hydraulic architecture should consider that some of the long-established paradigms might not be uniformly applicable to all tree species. Overall, the results of our study suggest that where natural forest cannot be conserved, agroforestry is the better land-use option in terms of long-term carbon storage and nutrient sustainability compared to rubber and oil palm monocultures. On the other hand, higher yields of monocultures and thus a potentially better income are decreasing the attractiveness of agroforests for local owners. Further depletion of carbon storage potential in the tropical lowlands may only be mitigated in cooperation with political and local institutions in the long term. Therefore, scientific evidence on the ecological costs and benefits of land-use change like presented in this study is aimed to guide political decision-making towards a more sustainable path in forest conservation and land management.

Birds play a pivotal role in ecosystems, but in disturbed areas their role may be limited due to changes in their habitat. In Simalungun and Asahan districts, North Sumatra, bird communities were studied in a range of habitats: natural forest, rubber agroforests (RAF), rubber monoculture plantations (RMP) and emplacement areas. The birds were observed using descriptive survey methods by implementing a quick biodiversity survey, with data collected along a 1km transect. In total, 142 species of birds from 42 families were found in the four habitats. Natural forests held the highest diversity of bird species, followed by rubber agroforests, emplacement areas and rubber plantations, with a Shannon-Wiener index of 4.5, 3.6, 3.6 and 3, respectively. Regarding the IUCN red list species, 12 bird species of near-threatened status and 2 species of vulnerable status were recorded. Based on CITES categories, one species was listed in the Appendix criteria I, 12 species were classified in Appendix II and 26 bird species were protected under Indonesian regulations. Changes in the structure and composition of vegetation in disturbed forests and cleared land determined the richness of bird species. The different tree compositions in the three habitats of the rubber estate plantations and surrounding areas influenced the number of bird species, their diversity and species composition.

Rainforest conversion into monoculture plantations results in species loss and community shifts across animal taxa. The effect of such conversion on the role of ecophysiological properties influencing communities, and conversion effects on phylogenetic diversity and community assembly mechanisms, however, are rarely studied in the same context. Here, we compare salticid spider (Araneae: Salticidae) communities between canopies of lowland rainforest, rubber agroforest (“jungle rubber”) and monoculture plantations of rubber or oil palm, sampled in a replicated plot design in Jambi Province, Sumatra, Indonesia. Overall, we collected 912 salticid spider individuals and sorted them to 70 morphospecies from 21 genera. Salticid richness was highest in jungle rubber, followed by rainforest, oil palm and rubber, but abundance of salticids did not differ between land-use systems. Community composition was similar in jungle rubber and rainforest but different from oil palm and rubber, which in turn were different from each other. The four investigated land-use systems differed in aboveground plant biomass, canopy openness and land use intensity, which explained 12% of the observed variation in canopy salticid communities. Phylogenetic diversity based on ~850 bp 28S rDNA fragments showed similar patterns as richness, that is, highest in jungle rubber, intermediate in rainforest, and lowest in the two monoculture plantations. Additionally, we found evidence for phylogenetic clustering of salticids in oil palm, suggesting that habitat filtering is an important factor shaping salticid spider communities in monoculture plantations. Overall, our study offers a comprehensive insight into the mechanisms shaping communities of arthropod top predators in canopies of tropical forest ecosystems and plantations, combining community ecology, environmental variables and phylogenetics across a land-use gradient in tropical Asia.

Land use effects on tree species diversity and soil properties of the Awudua Forest, Ghana

In chapter 2, I investigated the canopy spider abundance, biomass, richness and community composition across a land-use gradient from tropical lowland rainforest via jungle rubber to monocultures of rubber and oil palm. Overall, I demonstrated that canopy spider communities in oil palm and rubber plantations are less abundant, contain lower biomass, and are less diverse compared to rainforest and jungle rubber systems. I also provided evidence that community composition is similar between less intensive land-use systems, i.e. rainforest and jungle rubber, and differed from more intensive land-use systems, i.e. oil palm and rubber plantations, which also differed from one other. Investigating environmental variables that shape spider communities, I demonstrated that at family level aboveground biomass, number of trees per hectare and canopy openness were major factors determining spider community composition. I also demonstrated that at species level, the most important factors were plant richness and the number of trees per hectare. My results provided detailed information on the importance of rainforests for the conservation of canopy spider communities as only a subset of the community can tolerate the more harsh environmental conditions and disturbances in monoculture plantations. Moreover, the fact that in rainforest and jungle rubber the diversity and community composition are similar suggests that the majority of spiders might tolerate a moderate degree of disturbances. Finally, disturbed agroecosystems, such as oil palm and rubber plantations, may also contribute to the total diversity in a geographic area (γ-diversity) suggesting that landscape scale perspectives are needed for the conservation of spider diversity. In chapter 3, I focused on changes in the trophic structure of canopy spider communities with the transformation of tropical rainforests into rubber and oil palm plantations by measuring bulk tissue 15 N/ 14 N and 13 C/ 12 C ratios. Additionally, I studied in more detail the response of the families Salticidae, Clubionidae, Theridiidae, and Thomisidae, which contributed most to the abundance, biomass, and diversity of the spider communities in the studied land-use systems. Overall, we found that rainforest spider communities respond differently to land-use transformation. Generally, the range in trophic niches and use of basal resources (∆ C values) did not differ between land-use systems, but maximum ∆ 15 N values suggested that some families have narrower trophic niches in plantations. Moreover, functional diversity was similar across the four land-use systems, with the exception of isotopic divergence, which was higher in rainforest and jungle rubber than in monoculture plantations.Further, there was no difference in the trophic positions between cursorial and sheet-web building spiders, but they occupied virtually the same average position. However, maximum ∆ 15 N values in Salticidae, Clubionidae and Thomisidae were higher in plantations than rainforest and jungle rubber indicating that intra-guild predation may be more widespread in free hunting than web-building spiders but only in plantation systems. Finally, we found evidence that transformation of rainforest into monocultures plantations is associated with a shift towards more herbivore prey in canopy spiders. Overall, our study provides first insight into trophic changes in canopy spider communities associated with the transformation of tropical rainforest into plantations.

Rainforest canopies, home to one of the most complex and diverse terrestrial arthropod communities, are threatened by conversion of rainforest into agricultural production systems. However, little is known about how predatory arthropod communities respond to such conversion. To address this, we compared canopy spider (Araneae) communities from lowland rainforest with those from three agricultural systems in Jambi Province, Sumatra, Indonesia, i.e., jungle rubber (rubber agroforest) and monoculture plantations of rubber and oil palm. Using canopy fogging, we collected 10,676 spider specimens belonging to 36 families and 445 morphospecies. The four most abundant families (Salticidae N = 2,043, Oonopidae N = 1,878, Theridiidae N = 1,533 and Clubionidae N = 1,188) together comprised 62.2% of total individuals, while the four most speciose families, Salticidae (S = 87), Theridiidae (S = 83), Araneidae (S = 48) and Thomisidae (S = 39), contained 57.8% of all morphospecies identified. In lowland rainforest, average abundance, biomass and species richness of canopy spiders was at least twice as high as in rubber or oil palm plantations, with jungle rubber showing similar abundances as rainforest, and intermediate biomass and richness. Community composition of spiders was similar in rainforest and jungle rubber, but differed from rubber and oil palm, which also differed from each other. Canonical Correspondence Analysis showed that canopy openness, aboveground tree biomass and tree density together explained 18.2% of the variation in spider communities at family level. On a morphospecies level, vascular plant species richness and tree density significantly affected the community composition but explained only 6.8% of the variance. While abundance, biomass and diversity of spiders declined strongly with the conversion of rainforest into monoculture plantations of rubber and oil palm, we also found that a large proportion of the rainforest spider community can thrive in extensive agroforestry systems such as jungle rubber. Despite being very different from rainforest, the canopy spider communities in rubber and oil palm plantations may still play a vital role in the biological control of canopy herbivore species, thus contributing important ecosystem services. The components of tree and palm canopy structure identified as major determinants of canopy spider communities may aid in decision-making processes toward establishing cash-crop plantation management systems which foster herbivore control by spiders.

Land use and conservation value for forest birds in Central Sulawesi (Indonesia)