BlogPost #038

Abstract: Oil palm (Elaeis guinensis) is a controversial crop. To assess its sustainability, we analysed the contribution of different types of plantations (smallholder, industrial and unproductive) towards meeting six Sustainable Development Goals.

Using spatial econometric methods and data from 25,067 villages in Sumatra, Indonesia, we revealed that unproductive plantations are associated with more cases of malnutrition, worsened school access, more air pollution and increased criminality. We also proposed a strategy for sustainable palm oil expansion based on replanting unproductive plantations with either industrial or smallholder palm oil. Smallholder replanting was beneficial for five Goals (Zero poverty, Good health, Quality Education, Environmental preservation and Crime reduction). In comparison, the same intervention only improved two Goals in the industrial case (Zero poverty and Quality Education). Our appraisal is relevant to policymakers aiming towards the 2030 Agenda, organisations planning oil palm expansion, and retailers or consumers concerned about the sustainability of oil consumption.

Abstract: Sometimes it is challenging to translate the nuances of scientific research to non-expert audiences. I feel this is true for all fields of research but even more for modelling studies, since communicating the main assumptions of our models is often as important as communicating the results.

As scientists, we are very used to the fact that correlation is not causality, or that simulations are highly dependent on our underlying assumptions. However, while presenting the results to policymakers or other non-expert audiences it isn't easy to ensure everyone is on the same page. This is even more important when the results from our studies are being used to guide policymaking, for example as it was seen during the Covid-19 pandemic.

How could we get more prepared to present and communicate our results? Do you have any tricks?