Modelling The Effects Of Low Temperature Stress At Jointing And Booting Stages On Floret Development And Grain Setting In Wheat

In recent years,global warming has led to increased frequency,intensity and duration of low temperature stress events,which has had a great impact on crop production.Therefore,simulating the impacts of low temperature stress on wheat production accurately can improve the credibility of the assessment results of the impact of climate change on agricultural production,and it is of great significance for ensuring national food security.However,the current wheat growth simulation models have relatively few simulations on the process of the yield formation,especially the lack of simulation research of the differentiation and degradation of wheat florets and the spatial distribution of grain weights and grain numbers under low temperature stress that are closely related to yield formation.In this study,aiming at the shortcomings of the current wheat growth simulation models under low temperature stress,we implemented low temperature control experiments of wheat with different low temperature occurrence periods,low temperature levels and low temperature durations by using artificial climate chambers.We clarified the dynamic characteristics of wheat floret differentiation and degradation,and the spatial distribution characteristics of wheat grain weights and grain numbers under low temperature stress.Using system analysis methods and dynamic modeling techniques,we constructed a simulation algorithm for wheat floret differentiation and degradation and the spatial distribution of wheat grain weights and grain numbers under low temperature stress,based on the wheat growth simulation model Wheat Grow that our research group had.The simulation accuracy of the model under low temperature stress was improved significantly.The main results were as follows:(1)Low temperature stress at jointing and booting stages both delayed the development of young spikes,and the dynamics of wheat floret differentiation and degradation showed a trend from fast to slow.Low temperature stress at the jointing stage affected the differentiation stage of wheat florets.The total number of florets differentiated per spike showed a downward trend with decreasing temperature.Low temperature stress at the booting stage affected the degeneration stage of wheat florets.With the decrease of the temperature,the total number of florets degenerated per spike showed an increasing trend.The logistic equation and the exponential function could be used to simulate the floret differentiation and degradation,in which the equation parameters were affected by the low temperature influence factors at jointing and booting stages.Compared with the original model,the improved model significantly reduced the prediction errors of wheat floret differentiation and degradation dynamics under low temperature stress.Among them,the root mean square error(RMSE)of the wheat floret differentiation and degradation dynamics prediction were reduced by 10.09 and 24.02,respectively.The relative root mean square error(NRMSE)were reduced by 10.46%and 33.71%,respectively,and the coefficient of determination(R~2)were above 0.9.The results of the study provide effective support for the quantitative prediction and evaluation of wheat productivity under the background of climate change.(2)The grain weights and grain numbers of wheat showed an increasing trend first and then decreasing trend with the spikelet position from the base to the top.The distribution presented a mesial advantage,and the center of distribution was the 6-10 spikelets.Moreover,with increasing low temperature duration and decreasing temperature level,the grain weights and grain numbers in the same spikelet position showed a downward trend.The middle spikelet positions were more severely affected by low temperature,followed by the lower spikelet positions,and the upper spikelet positions were less affected.Based on the wheat experimental data of different varieties,different low temperature treatment periods,different low temperature levels and different low temperature durations in the artificial climate chambers and combined with the original spatial distribution algorithm of wheat grain numbers and grain weights in the model Wheat Grow,we proposed a new simulation algorithm for the spatial distribution of grain weights and grain numbers under low temperature stress at the jointing and booting stages,in which the low temperature sensitivity parameters of variety were introduced to reflect the differences in low temperature sensitivity among different varieties at different growth periods.The model was tested by the independent observation data.Compared with the original model,the improved model significantly reduced the prediction errors of the grain weights and numbers of the two varieties at different spikelet positions under low temperature stress.Among them,the root mean square error(RMSE)of the prediction of grain weights and numbers at different spikelet positions under the low temperature stress at jointing,booting and jointing-booting stages were reduced by 38.6%and 25.4%,38.3%and 36.8%,and 37.5%and 36.4%on average,respectively.The coefficients of determination(R~2)were mostly above 0.95.In addition,the improved model was used to analyze the sensitivity of the grain weights of each wheat spikelet to low temperature under different low temperature stress scenarios.The results showed that the improved Wheat Grow model was suitable for predicting the spatial distribution of wheat grain weights and grain numbers under different low temperature stress scenarios.The results of the study will deepen the understanding of the change rule in grain development in wheat spikes and lay the foundation for further quantitative regulation of wheat spike development.