Over the past century, oil palm has developed from a sustenance crop in West Africa to the world’s most traded vegetable oil. Although the oil palm industry contributes to local and national economies across the tropics, its cultivation has also been linked to severe negative impacts on biodiversity, ecosystem functioning, and the wellbeing of local communities. While there is a growing awareness of the need to manage such agricultural landscapes more sustainably, there remains a lack of interdisciplinary research focused on tractable solutions to do so. In this thesis, I conducted social, ecological, and interdisciplinary studies of both smallholder and industrial plantations in two key oil palm countries, aiming to provide a more holistic understanding of the social factors influencing plantation management, and the ecological effects of alternative management options.

My first chapter reviews the global relevance of agriculture and acts as an introduction to this thesis. In my second chapter I carried out a systematic mapping exercise to quantify social, ecological, and interdisciplinary research on oil palm cultivation, assess trends in publication focus, and identify priority knowledge gaps in the literature. I found a global increase in oil palm research over the past three decades, with over 70% of research focused on ecological outcomes, but only 20% on social and less than 10% on interdisciplinary. The majority of studies were conducted within industrial plantations and compared oil palm to non-modified habitats, such as forests. The most pressing knowledge gaps included a lack of studies on the effects of plantation inputs on pollination and herbivory, the relationship between ecological factors and human health and wellbeing, and comparisons of different management practices within oil palm plantations. I argue that these gaps should become the focus of future research attention, as they lie in identified priority research areas and their outcomes are critical for informing the development of more sustainable palm oil production. Through the rest of my thesis, I present research which begins to address these topics.

In my third chapter, I conducted social surveys in collaboration with smallholder farmers in Riau, Indonesia, and Perak, Malaysia. These farmers varied in the extent to which they followed management-assistance programs led by an industrial palm oil company (in Indonesia) and a conservation focused NGO (in Malaysia). These groups acted as examples of the range of farmer socio-demographics, attitudes, and management decisions that exist, and allowed me to assess the alignment between the intentions of partnership programs and the current realities of smallholder plantations on the ground. Field work was conducted remotely with partner organisations in Malaysia and Indonesia, owing to the impacts of COVID. I found that farmers most closely partnered with the private sector were highly variable in their socio-demographics and attitudes, but showed little variation in management inputs. In contrast, farmers most closely partnered with an NGO were highly varied in all survey sections, and deviated from the intended management regime. Income, herbicide usage, and vegetation management were consistently among the most important factors determining differences between groups, and could be a fruitful focus of further study on smallholder management. My findings demonstrate the wide variety of farmers found within the most productive oil palm regions in the world, indicating that a ‘one-size-fits-all’ approach to sustainability is unlikely to be effective in a smallholder context.

In my fourth chapter, I conducted social and ecological surveys with independent smallholders in Peninsular Malaysia to investigate which social factors affected smallholder farmer decisions to replant oil palm as a mono versus polyculture crop, and how such systems differed in management inputs and environmental features. I found no significant differences in socio-demographics or attitudes between farmers deciding to replant with a monoculture or a polyculture, suggesting there is no “typical” polyculture farmer. I also found no significant differences in management inputs, such as herbicide or fertiliser application, between plantation types, suggesting that farmers are not tailoring management strategies towards their unique plantation needs. However, plantations did differ in environmental features, suggesting that intercropping influences the ecosystem, regardless of inputs. In particular, soil conditions, understory vegetation, and key pests present differed between plantation types, with more abundant understory vegetation and higher herbivory found in polyculture plantations.

In my fifth chapter, I investigated the impact of landscape scale plantation heterogeneity on biodiversity in industrial oil palm plantations in Sumatra, Indonesia. I assessed whether density and behaviour of day-flying Lepidoptera varied between different habitats within plantations (Edge habitats bordered by plantation roads on one side, and Core habitats in the centre of oil palm planting blocks) and across seasons (March and September). I found significantly higher total density of Lepidoptera in Edge than in Core habitats. There was an interaction between season and habitat, with density increasing more markedly in Edge than Core areas in September. There was also a significant effect of habitat and season on behaviour, with more active behaviours, such as foraging and mating, recorded more frequently in Edge than Core habitats, and more commonly in September than March. My findings indicate that Lepidoptera abundance is affected by habitat characteristics and can be influenced by plantation management. In particular, increasing non-crop vegetation in plantations through reduced clearing practices or planting of flowering plants could foster more abundant and active butterfly communities within oil palm. In the following chapter I investigated this hypothesis.

In my sixth chapter and final data chapter, I assessed the long- and short-term effects of varying vegetation management regimes on the abundance, richness, and diversity of day-flying Lepidoptera, again based in mature industrial oil palm plantations in Sumatra, Indonesia. Over the long-term, less intensive understory vegetation clearing resulted in significantly more abundant and diverse Lepidoptera communities, with limited differences between plots with restricted herbicide application and plots where no herbicide was used. In contrast, Lepidoptera communities were little affected by vegetation management over the short-term, suggesting that manual removal of vegetation may be equally as damaging to butterfly and moth communities as removal by intermediate levels of herbicide spraying. My findings substantiate calls to limit understory clearing in oil palm and maintain habitat heterogeneity, while also indicating that a hard ‘no-spray’ guideline may not be the only option for developing more butterfly-friendly plantations. My findings relate directly to key areas of research (namely understory vegetation treatment and herbicide application) identified in my social and interdisciplinary data chapters (Chapters 2, 3 & 4). My seventh chapter is a discussion of my thesis findings as a whole, and its implications for socio-ecological research and the oil palm industry.

Results of my thesis identified priority areas for oil palm research and management in both industrial and smallholder contexts, and have begun to close gaps in knowledge on tractable and effective methods to manage oil palm plantations more sustainably. I have demonstrated the linkage between social and ecological factors, and shown that social factors affect management inputs, which directly affect biodiversity and ecosystem functioning. Altogether, my thesis demonstrates that both small- and large-scale oil palm plantations can, and must, develop tailored management strategies to support biodiversity, ecosystem functioning, yield, and farmer socioeconomics, in order to preserve tropical habitats and livelihoods while meeting growing global demand.