One thrust in increasing food security in Jamaica is expansion of cassava production. The multiple shoot removal technique (MSRT) for rapid propagation of cassava can help address limitations in planting material. Shoots sprouting from cuttings of hardwood stem are severed in such a way as to induce further sprouting, and then put to root for subsequent transfer to the field. The effects of age and fertilization of parent plants and nodal age of stems were studied. Six Colombian varieties were planted in fertilized and unfertilized field plots with similar growing conditions to provide stems for MSRT propagation. Volume of two-node cuttings increased from apical to basal nodal age, but cutting density was a better predictor of shoot production. On average, three to six viable shoots were produced per cutting over 3 months in a greenhouse. All nodal ages of stems from parent plants aged 6, 7 and 9 months were suitable if the quality of the planting stakes producing parent plants was adequate. If stake quality is uncertain, it is recommended that apical pieces are not used from parents younger than 9 months. The variety CM 6119-5 consistently produced most shoots, suggesting a strong genotypic effect, but other varieties, particularly CM 849, were less consistent, indicating the role of environmental interactions. The physiological status of cuttings as influenced by stem maturity, parent plant age, nutrition and growing conditions of both grandparent and parent stems was as important as genotypic characteristics in determining shoot production from two-node cuttings of cassava stem.

Cassava (Manihot esculenta Crantz) is an important food crop, especially in developing countries, because of its resilience and ability to grow in conditions generally inhospitable for other crops. However, tropical crops like cassava are not as frequently modeled compared with crops from temperate locations. The objective of this research was to calibrate the CSM-MANIHOT-Cassava model of the Decision Support System for Agrotechnology Transfer, DSSAT beta v4.8 and use the model to evaluate the potential benefits of irrigation on yield. We established two field trials with two water treatments (rainfed and irrigated) and four cultivars that had not been studied previously. We simulated in-season biomass and end-of-season yield, evaluating the model performance with different statistical measures. There was good agreement between simulated and measured values; the best results showed a deviation of 9.7%, normalized RMSE of 18%, and d-index of 0.98 for biomass, with corresponding values of 11, 24, and 0.98, respectively, for yield. Good simulations of yield correlated with accurate simulations for leaf area index and harvest index. The varieties showed differential responses to irrigation, suggesting that there are diverse levels of drought tolerance even within the same environmental conditions.