Bayesian network modeling of early growth stages explains yam interplant yield variability and allows for agronomic improvements in West Africa

Yams (Dioscorea spp.) are important species, especially for resource-poor farmers of West Africa, where crop yields are affected by early plant size hierarchy linked with uneven emergence. Although the causes of this phenomenon are not fully known, yams, like other vegetatively propagated crops, have heavy planting material that is liable to induce such interplant variability. In addition, planting practices may mitigate this phenomenon via the selection of the seed-tuber size or state. To gain further insight into yam interplant variability, this study identified and quantified, for the first time, the direct and indirect dependency between planting practices, early growth variables and yield components of Dioscorea rotundata and Dioscorea alata, the two main food yam species. The experimental dataset came from six field trials carried out in Benin at two locations between 2007 and 2009. Additive Bayesian network modeling was used for structure discovery—its directed acyclic graph offers an ideal background for discussing complex systems when theoretical knowledge is lacking, e.g., for yams. Here we found that the emergence date was the only direct cause of plant yield variability common to both species. For D. rotundata, we observed a direct contribution of the cataphyll number to the plant tuber weight. These combined results suggest the existence of some uncontrolled latent variables (i.e., seed-tuber physiological age and reserves). For D. alata, the model did not reveal any effect of seed-tuber size, despite a strong effect noted for D. rotundata. We suggest that the transposition of traditional native D. rotundata planting practices may have led to oversized D. alata seed-tubers, resulting in wastage of planting material. This study demonstrated that traditional West African cropping systems have a serious drawback concerning the uncontrolled wide range of physiological ages and reserves in seed-tuber lots, which affect the plant size hierarchy and ultimately the marketable yield. (Résumé d'auteur