Sweet sorghum (Sorghum bicolor (L.) Moench) response to supplemental irrigation in different growth stages

Abstract

Sweet sorghum (Sorghum bicolor (L.) Moench) is a high biomass and sugar-yielding crop. There is recently an interest in sweet sorghum as feedstock for ethanol production, since it is rich in sugars, and reportedly has low nutrition and water requirements. A field experiment was conducted at Hatfield Experiment farm of the University of Pretoria, South Africa in 2010/11. The aim of the study was to evaluate sweet sorghum performance under different water regimes and determine their effect on biomass and sugar yields. Four water treatments (Control, Supplemental irrigation at early vegetative stage (EVS), Supplemental irrigation at late vegetative stage (LVS) and Dry land. One sweet sorghum variety (Sugar graze) was used in the experiment. Plant height, leaf area index (LAI) and dry matter accumulation were measured periodically through growth analysis. This data was used to calibrate the Soil Water Balance (SWB) crop model for sweet sorghum. Brix and quality analysis were carried out by the ACCI laboratory at the University of KwaZulu Natal at final harvest. Total fresh biomass production (t ha-1) of sweet sorghum was significantly improved by full irrigation (Control) and supplemental irrigation at either the early vegetative stage (EVS) or late vegetative stage (LVS), compared to the Dry land treatment. This shows that with supplemental irrigation higher fresh biomass production can be attained. The Control and Supplemental irrigation at early vegetative stage (EVS) treatments gave the highest fresh stalk yield (16.6 ton ha-1 and 17.1ton ha-1 respectively) at harvest, followed by the Dry land and Supplemental irrigation at late vegetative stage (LVS) treatments. However, the three irrigated treatments did not differ significantly from each other with regard to total biomass production. It should be noted that the experiment was conducted during a wet season (total rainfall of 757mm during the growing period), which may have contributed to the limited response of the crop to applied water treatments. Water treatments had no significant effect on total dry matter yield. Although the differences among the water treatments were not significant, a slightly higher dry matter yield was obtained for EVS. Similarly, total sugar yield (t ha-1) and theoretical ethanol yield (L ha-1) were not significantly influenced by water treatments. Sugar yields ranged between 1.64 and 2.77 t ha-1 and ethanol yields between 1763 and 2984 L ha-1. The results also showed that treatments that were irrigated until late in the season (Control and LVS) had lower stalk dry matter contents (% DM) than both EVS and Dry land treatments. This probably resulted in lower sugar (t ha-1) and ethanol yields (L ha-1) for these treatments, although high fresh stalk yields were obtained. Brix (t/ha) was greatly influenced by irrigation as there were significant differences between all the water treatments. Based on these results, the main objective was achieved since it was clear that irrigating during the early stages of plant growth ensured optimum results in terms of biomass yield, sugar and ethanol yield. Model simulation results for top and harvestable dry matter and leaf area index of all the irrigation treatments were within reasonable accuracy and statistical parameters were generally acceptable. Soil water deficits were not so well simulated, especially during dry periods when simulated deficits were much higher than measured values. Nonetheless, it can be concluded that the SWB model should be a useful tool for scenario modelling in order to estimate sweet sorghum production and water use under a wide range of conditions.