Biodiversity and Ecosystem Function in Tropical Agriculture (BEFTA) Project Working with the industry to investigate biodiversity management in oil palm Jake L Snaddon1, Edgar C Turner2*, William A Foster2 1Biodiversity Institute, University of Oxford, Oxford, OX1 3PS 2University Museum of Zoology, Downing Site, Cambridge, CB2 3EJ *Contact: ect23@cam.ac.uk, www.oilpalmbiodiversity.com

Introduction

It is clear that forest conversion to oil palm has severe impacts on biodiversity1,2. However, little research has focussed on methods that can be employed to maximise biodiversity within plantations, nor the role of this biodiversity in ecosystem functioning and crop production2,3. This project aims to quantify the effect of habitat complexity within oil palm plantations on biodiversity and the role of this biodiversity in ecosystem functioning and productivity. By working closely with industry partners, Sinar Mas Agro Resources and Technology Corporation Research Institute (SMARTRI), the project is able to experimentally manipulate the oil palm landscape and link recorded changes in biodiversity to ecosystem functions and crop production. Project objectives: • quantify the effect of habitat complexity in maintaining biodiversity, ecosystem function and

ecosystem services within oil palm • develop experimental approaches for partitioning the effects of habitat structural complexity

and aspects of biodiversity on ecosystem functioning • predict and model optimal cover of understory and epiphyte vegetation in oil palm plantations

so as to maximize biodiversity and economic profitability through ecosystem services.

Results

First year of pre-treatment data is already demonstrating how variable insect communities can be within oil palm plantations. For example, composition analyses showed that both butterflies and dragonflies have different communities in plantation road and plantation interior areas (Fig. 3). Other results have demonstrated the potential importance of oil palm invertebrates for ecosystem processes such as decomposition. For example, mass loss of litter after 60 days was higher for mesh bags that allowed free access of invertebrates than those that did not (Fig. 4).

Study Site

Fieldwork is all based in in Riau, Sumatra in established plantations, owned and managed by SMARTRI (Fig 1).

Figure 2. One replicate, consisting of the three different treatment plots: Reduced complexity, removal of all understory and epiphyte trunk vegetation; Normal complexity: standard industry practice Enhanced complexity: reduced management input, reduced herbicide input and understory cutting.

Experimental set up

The experimental set-up includes three levels of habitat complexity, each replicated six times across the landscape:

Reduced complexity: This represents highly destructive management and will include the removal of all understory and epiphyte trunk vegetation using herbicides Normal complexity: This represents standard industry practice Enhanced complexity: This represents reduced-input management/conservation treatment and will include reduced herbicide input and understory cutting.

The treatment plots are located immediately adjacent to the road and in adjacent management blocks (Fig. 1). Each plot is 150m x 150m (2.25 ha) in area, with a central 50m x 50m (0.25 ha) sampling area.

Data Collection Data are being collected on a wide variety of environmental factors, taxonomic groups, ecosystem functions, and crop productivity and management in each of the plots for a year before the experimental treatments are implemented and for three years afterwards. By collecting intensive data on multiple factors in the same areas both before and after the experimental treatment, analyses can take existing habitat variability into account. Data collection framework includes:

Conclusion

The BEFTA Project has already collected comprehensive data that describes the habitat structure, biodiversity, ecosystem functioning and yield of oil palm plantation areas. This will allow future work on the project, following experimental manipulation, to assess the impacts of habitat management on oil palm ecosystems. Although data collection and analysis is still in its early stages, it is already clear from initial results that there are distinct differences in the community composition of insects in different areas of the plantation (the road or interior) and that plantation invertebrates can facilitate processes such as litter decomposition and therefore nutrient cycling within the plantations.

Literature cited (1) Fitzherbert, E.B., et al. (2008) Trends in Ecology and Evolution, 23, 538-545. (2) Foster, W.A. et al. (2011) Philosophical Transactions of the Royal Society of London B, 366, 3277-3291. (3) Turner, E.C. et al. (2011) Book Chapter in Environmental Impact of Biofuels, InTech open access publisher (http://www.intechweb.org/), Edited by Marco Aurélio dos Santos Bernardes, ISBN 978-953-307-479-5.

Acknowledgements

The BEFTA Project is co-funded by the Isaac Newton Trust, Cambridge and Sinar Mas Agro Resources and Technology Corporation Research Institute (SMARTRI). We would like to thank the staff and researches of SMARTRI for their support on the project and help with data collection and RISTEK Indonesia for granting research permission.

0

10

20

30

40

50

60

70

80

90

100

0 10 20 30 40 50 60 70 80 90 100 110 120

Fine

2mm

1cm

% M

ass

Rem

ain

ing

Number of days

% M

ass

Rem

ain

ing

Time (days)

Figure 3. Plot based on a Detrended Correspondence Analysis for dragonflies in plantation interior and road areas. The butterfly data produced similar results (not shown). Photographs show some representative morpho-species from the different habitat types.

Figure 4. Mean % mass loss of oil palm litter from bags of different mesh sizes (<0.5mm, 2mm, and 1 cm) as a result of decomposition, March-July 2013. N=54 litter bags of each size at each time point, error bars represent standard deviation.

-800

-600

-400

-200

0

200

400

600

-100 0 100 200 300 400 500 600

Plantation interior

Plantation road

DCA Axis 1

DC

A A

xis

2

Ecosystem function data • Herbivory measures • Litter decomposition (bags of 3 mesh) • Parasitization & predation rates on oil palm-feeding

caterpillars

Habitat structure & Environmental data • Topography, slope and aspect data in each plot • Temperature recorded at three hourly intervals within

each plot • Vegetation structure and canopy openness at set points

within each plot

Biodiversity data • Invertebrate abundance and biomass using integrated

flight intercept and pitfall traps (see image right) • Transect surveys for butterflies and dragonflies in the

middle and road edge of each plot • Targeted searches for aquatic invertebrates • Point counts for birds

Figure 1. Locations of the BEFTA plots (yellow squares) within SMARTRI plantations, Riau, Sumatra.

Reduced Normal Enhanced

150m 150m

50m 50m

150m

50m

Ecosystem services and industry data • Palm fruit bunch weight and quantity • Worker hours in plot • Quantity and cost of chemical applications