Combining modelling and experimental approaches to assess the feasibility of developing rice-oil palm agroforestry system

Climatic hazards affecting the main rice producing regions of Indonesia increase the risk of annual production loss and encourage the development of innovative strategies to maintain stable production. Conversion of oil palm monocultures to rice-based intercropping systems is a strategy to be considered, but relies on the existence of suitable planting management that optimizes both palm productivity while providing enough light for undergrowth rice varieties tolerant to shady conditions. This paper proposes to couple a model of light interception on virtual canopies with indoor experiments to evaluate the feasibility of developing rice-oil palm agroforestry systems. We first selected a planting design that optimized the transmitted light available for rice using a functional-structural plant model (FSPM) of oil palm. Secondly, we reproduced the light regime simulated with specific changes in the intensity and the daily fluctuation of light in controlled conditions. Three light treatments were designed to test independently the effect of daily light quantity and the effect of diurnal fluctuation on contrasted rice subpopulations. Light quantity was the main factor driving changes in plant morphology and architecture, while light fluctuation only appeared to explain variations in yield components and phenology. This study highlighted the importance of light fluctuation in the grain filling process and resource reallocation. The conservation of relative change among varieties between treatments suggests that varietal responses to low light are likely to be heritable, and that varietal screening under full light can provide clue on varietal behavior under low light. However, the identification of specific traits such as a limited expansion of leaf area and a conservation of leaf senescence under shade and high light fluctuation paves the way for selecting varieties dedicated to agroforestry systems. Further investigations including light quality and larger genotypic population to screen are discussed. ### Competing Interest Statement The authors have declared no competing interest.