Evidence brief: Using Sentinel-1 satellite imagery to quantify oil palm cultivation: A case study from Cameroon

Abstract

Tropical deforestation is increasing globally, putting forest ecosystems and the ecosystem services they provide at risk. The widespread use of palm oil in fast food, cosmetics, household cleaners, and other products has led to a high demand for oil palm trees, Elaeis guineensis, in tropical regions due to their favorable growth conditions and low production costs. Thus, rapid agricultural expansion poses a significant threat to regions reliant on palm expansion for economic growth, such as the African tropical nation of Cameroon. This study aims to quantify the change in oil palm extent from 2015 to 2025 in Cameroon, using Sentinel-1 SAR satellite data and the machine learning algorithm Random Forest. Our findings reveal an increase in oil palm cultivation ranging from 13-55% in the focus areas, with the South Region showing the greatest change in forest cover as a result of oil palm expansion. We also discuss the use of Sentinel-1 satellite data and machine learning for similar studies monitoring oil palm deforestation in tropical regions.

Introduction

Tropical rainforests are biodiversity hotspots [1,2] that provide critical ecosystem services such as food provision, raw materials (i.e. timber), carbon sequestration, air and water purification, and recreation/tourism [3–6]. Tropical rainforests are especially valuable for the global carbon budget as they store approximately 195 petagrams of carbon (β25% of global carbon), and are therefore effective sinks for atmospheric carbon, mitigating climate change [3,7–9]. Tropical rainforests are under threat largely due to agricultural and urban expansion focused on increasing urban development [10,11]. Understanding and monitoring agricultural-driven deforestation in tropical regions is critical to assessing and mitigating the impacts to ecosystem services [12].

Rapid deforestation is occurring in tropical forests throughout Southeast Asia, Latin America, and Africa, in part due to oil palm (Elaeis guineensis) cultivation [6,13–15]. Oil palm cultivation is expanding rapidly due to high production yields and low costs relative to other vegetable oils, enabling its widespread use in biofuels, fast foods, cosmetic products, and household cleaners [6,13–17]. In fact, oil palm is now the most commonly consumed vegetable oil source globally and accounts for up to 45% of agricultural deforestation [13]. Efforts to understand the impact of oil palm on forest loss are mainly concentrated in Southeast Asia, while the expansion of the oil palm industry in African nations remains largely unexplored [15]. The Congo Basin Rainforest is a suitable oil palm growth region that contains up to 30% of the world’s tropical forest carbon stock [15]. Little is known about the rate at which oil palm plantations are replacing tropical forests in Africa, and particularly in Cameroon which currently has one of the highest deforestation rates of any country [18,19]. Local socio-political factors within Cameroon, such as the Anglophone crisis, influence land use decisions and introduce additional sustainable agriculture challenges [20]. It is therefore difficult to assess the impact of the oil palm industry on biodiversity and valuable ecosystem services, and Cameroon would benefit from accessible methods to track palm oil cultivation over time.