Grain quality of rice (Oryza sativa L.) genotypes grown under aerobic and rainfed production environments in Queensland, Australia

The Australian rice industry faces tremendous pressure for incremental increase in water use efficiency. Rainfed (RF) rice production could provide opportunities for diversifying the options for the rice industry to address the issue of water shortage. Rice varieties for RF production need to be developed to extend the rice industry to Australian wet-tropical regions. Achieving economic yields and matching grain quality to market standards are some of the challenges for RF rice production. Thirteen RF rice genotypes obtained from Australian Agricultural Technologies Limited (AAT) were evaluated during 2013 – 2015 in central and north Queensland locations as RF and supplementary/strategic irrigation (SI) crops, and were assessed for their yield potentials and grain quality performances. The results from varietal trials under RF and SI showed large genetic variation, primarily associated with grain types (GT), for yield and grain quality attributes. The long grain (LG) types recorded significantly greater (6.7 fold) yield response to SI (3.16 vs 0.47 t/ha) compared to the medium grain (MG) type varieties (1.8 fold yield in SI) (4.37 vs 2.49 t/ha). The RF yields under the drier tropical environment of Alton Downs (AD) were limited by a short growing season and low rainfall which impacted on late maturing LG type indica varieties. However, greater yield was recorded for early maturing, MG type japonica varieties. The effect of SI on grain quality was not as strong as the effects on the yields. Among the milling characteristics millout, was significantly greater for LG compared to MG type (0.58 vs 0.47) under the SI. Among the chemical characteristics, the SI rice showed greater gelatinisation and pasting temperature compared to RF rice, particularly in the second year, due to earlier planting, therefore, LG type varieties escaped low temperature stress. The effects of SI on physical, cooking and nutritional characteristics were not significantly. Given the short window of the warm, wet season (January – May) at AD, the performance of LG types was significantly favoured by SI. The LG type were generally late maturing, and therefore, flowering and grain filling coincided with cold spells, particularly for late planted crops. Hence, a RF cropping option at AD is favourable for short duration MG type varieties. In the varietal evaluations under RF production over multiple seasons of three years, RF yield was 1.38 – 1.43 t/ha. The overall yield of MG type was significantly greater compared to LG type (2.49 vs 0.47 t/ha). In the first season (planting February 2013), none of the LG type produced yield; however, in the second and third season (planted in January and November), LG type varieties produced yields of 0.12 – 0.66 t/ha. These longer term yield data for LG types clearly demonstrated that the commercial harvest of LG as a rainfed crop in central Queensland could not be a viable option. Among the MG types the higher yielding varieties were AAT-4 and AAT-19 (2.92 t/ha), followed by AAT-6 and AAT-17. In spite of the low yield, the milling ratio/millout was significantly higher in LG types compared to MG varieties. The Year (Y) x Grain type (GT) x Variety (V) interaction for milling ratio was due to the fact that the varietal effect was significant in both grain types, as LG variety AAT-12 recorded the highest millout (~0.60), whereas MG variety AAT-4 maintained lowest mill out (0.40) in the first year. Similarly, chalkiness was significantly higher for MG compared to LG (14% vs 7%) types. The grain chemical characteristic measured as alkali spreading value was significantly greater for MG compared to the LG varieties (6.01 vs 1.60 score). The amylose content differed significantly due to YxGTxV interactions suggesting greater genotype x environmental effects. Amylose content was greater in year1 followed by year2 and year3, and was generally higher for LG compared to MG. However, the highest amylose containing LG varieties AAT-15 and AAT-16 in 2013 and 2014 showed significant reduction of amylose in 2015 but the amylose content of LG AAT-18 recorded consistently higher amylose (24%) in all 3 years. The gelatinisation temperature (GeT) of LG type lines was significantly greater than that of the MG types (74.2 °C vs 70.2 °C). In summary, the yield performance of LG type at AD was low, primarily governed by a smaller planting window for late maturing rice varieties. Some of the MG type varieties, e.g., AAT-4 recorded the highest but economically marginal yield, but the MG types suffered from poor grain quality associated with the lowest millout and higher chalkiness; both of these contributed negatively to marketable yield. Hence, a RF rice crop performance under a dry tropical environment with limited rain and a short planting window can be a significant limitation for development of the RF rice industry. The AAT rice varietal evaluations in contrasting RF environments of AD (c. 593 mm rain) and wet tropical South Johnstone (SJ) (c. 876 mm rain) showed that paddy yield in the SJ was 2.4 times greater than at AD (4.66 vs 1.92 t/ha), and the yield difference between the LG and MG in the wet tropics was not significant (4.92 vs 4.36 t/ha), demonstrating the opportunity of commercial RF rice crops for both grain types. Significant effects of the growing environment were noted on grain quality parameters. Millout was consistent across the locations; however, a greater millout was recorded for LG compared to MG type (0.56 vs 0.50). Although greater chalkiness was recorded for MG compared to LG types at AD (27% vs 13%), no difference between GT for GeT was recorded at SJ. The GeT varied between 71.3 and 80.0 °C and differed due to GT x V interaction; the greatest GeT (80.2 °C) was recorded for AAT-4 at both locations. The amylose content is related to cooking and nutritional quality, but did not vary significantly due to V and locations. The effect of growing location was significant for cooking time, suggesting that grain from AD took longer to cook (42 vs 29 minutes). Nutrition characteristics, primarily the protein, nitrogen and sulphur concentrations of the grain were higher at SJ, irrespective of the GT compared to AD where the protein, nitrogen and sulphur concentration decreased in MG compared to LG varieties, showing significant L x GT interactions. Under the environment of SJ, the yield and grain quality of LG varieties were enhanced when compared to AD. The results showed that AAT RF rice varieties in SJ could produce a feasible crop, with acceptable level of milling recovery, and grain quality traits comparable to those of the commercially recommended varieties such as Topaz and Doongara in tropical Australian north Queensland.