Studying the grain growth process of wheat varieties under drought stress conditions using mathematical models

Drought stress is one of the most important environmental stresses in different regions of the world, which causes the instability of crop production. More than 20% of the world's agricultural lands are affected by moderate to severe drought stress. The seed filling stage is the last stage in the plant development and the most important stage in the accumulation of dry matter in the seed. Seed filling rate and period and the effect of various environmental and agronomical factors on these two important parameters affecting seed weight can be predicted using the mathematical models. Quadratic, polynomial cubic, and logistic models are among the mathematical models that have been used efficiently to predict the grain growth process. The purpose of this research was to investigate the grain growth process and the accumulation of photosynthetic substances in the grain of different wheat cultivars under non-stress (control) and late season drought stress conditions using mathematical models and to investigate the effect of grain growth parameters on grain yield.

This experiment was conducted as split plots based on randomized complete block design with four replications in the research field of Faculty of Agriculture, University of Zanjan, Zanjan, Iran, in 2018-2019. Normal irrigation (control) and drought stress after flowering were considered as main factor and four wheat cultivars (Shiraz, Marvdasht, Azar2 and Roshan) as sub-factor. Drought stress was applied by interrupt irrigation until the soil water potential reached about -2 MPa. To investigate the grain growth process, samples were taken from the main spikes after flowering every week. After drying the spikes, the grains were separated from the spike, the weight of single-grain was calculated, and the grain growth process was investigated using different mathematical models. To determine the best model, various statistical indices including root mean square error (RMSE), corrected Akaike (AICc) and ∆i were used, and the model with the lowest values of these indices was selected as the best model. Then, maximum and average grain filling rate, and grain filling period was estimated using the best selected model. The traits of plant height, main spike length, grain yield, biological yield and harvest index were also measured in addition to the grain growth process.

Comparison of different models using RMSE, AICc and i∆ indices showed that Darroch and Baker model (model number one) was the best model to evaluate seeds growth process in this research. Investigating the growth of grains with this model showed that the final grain weight decreased in all cultivars under drought stress conditions, and according to the prediction of this model, the lowest and highest final grain weight was belonged to Azar2 and Shiraz, respectively. Grain filling period decreased by 9.3% under drought stress conditions, and Shiraz (42.3 days) and Azar2 (34.4 days) had the highest and lowest grain filling period, respectively. Drought stress also reduced photosynthetic rate, leaf area durability, and number of grains per spike, but average and maximum grain filling rate were not affected by drought stress, which is probably due to the increase in the remobilization of photosynthetic materials under drought stress conditions. A positive and significant correlation was observed between grain filling period and grain yield under drought stress (r=0.375) and control (r=0.634) conditions. It can be concluded that the grain filling period was more important than the grain filling rate in this research. Drought stress decreased 1000-grain weight of the studied varieties by 18.3%, and Azar2 (47.70 g) and Marvdasht (33.50 g) had the highest and lowest 1000-grain weight, respectively. Grain yield and harvest index of the studied varieties also decreased by 40.8% and 22.4% under drought stress conditions, respectively, and Shiraz (4747.4 kg) and Azar2 (3179 kg) varieties produced the highest and lowest grain yield. Decrease in the harvest index indicated that drought stress reduced the grain yield more than the biological yield, which is probably due to the reducing of the grain filling period as well as disturbance in the grain filling process.

The results of this experiment showed that late season drought stress significantly reduced the photosynthetic rate, leaf area durability, number of grains per spike, 1000-grain weight and grain filling period, and finally led to a decrease in grain yield of the studied cultivars.