关键词: 小麦, 玉米, 粒重动态, 共性特征, 模型筛选

Abstract:

Grain weight (GW) is one of important components of yield in cereal crops. Currently, there are several models on GW of cereal crops, such as wheat (Triticum aestivum L.), maize (Zea mays L.), and rice (Oryza sativa L.). However, these models are mostly applicable on a single crop.To establish a common model of GW for at least two crops with wider application under different conditions, three cultivars of winter wheat and four cultivars of summer maize were used in field experiments in four environments in 2006–2008. Each cultivar had three treatments of density. A common GW model, y= a / (1+be^-cx), was developed with normalized GWand grain filling duration for the two crops. The parameters of a, b, and c were 1.0624, 52.8653, and 6.7609 (r=0.9916, P<0.01) on the basis of the experimental data, respectively. In different crops, cultivars, and densities, the GW dynamic model kept a relative stable a value, which was around 1; however, the b and c values varied in different conditions. The b value changed slightly with density, and shift from 45.3379 to 66.9306 in different cultivars; whereas, the c value had small differences among different cultivars and densities, and varied from 6.2122 to 6.8025 in winter wheat and from 7.0199 to 7.7325 in maize. The accuracy and precision of the normalized model were tested with theGWdata of winter wheat from Jiaozuo, Henan province and summer maize from Tianan, Shandong province as well as data in this study. The normalized dynamic model could make a good estimation of GW dynamics with the accuracies of 0.9870, 1.0057, and 0.9982, 1.0131, and the precision (R²) of 0.9854, 0.9918 and 0.9772, 0.9926 for winter wheat and summer maize respectively. Compared with other GWmodels established by other researches, normalized GW dynamic model could eliminate the variance of the model parameters caused by location, year, cultivar, and density. Normalized GWdynamic model can predict the increase of GW reliably and easily, if the GW_max and grain filling duration are acquired, and the characteristics of parameter b are ascertained. This model is applicable to calculate the GW of winter wheat and summer maize at early, middle, and late stages of growth under different conditions (region, years, cultivar, and density), and the error is less than 0.2797 between the measured GW and the simulated GW.

Key words: Wheat, Maize, Grain weight dynamic, Common characters, Model selection