Simulation Evaluation And Application Of Cotton Growth In Film-free Cultivation In Southern Xinjiang Based On Cotton2K Model

Plastic film cultivation technology has a great impact on agricultural production mode and regional planting structure.However,plastic film mulching cultivation has brought adverse effects on the environment while increasing the yield of cotton.Filmless cultivation is an effective method to solve the residual film pollution in the cotton fields of southern Xinjiang,and is a sure way to ensure the green and sustainable development of cotton in China.In order to promote the application of Cotton2 K model in the simulation of cotton growth without film cultivation in South Xinjiang region,this paper took cotton 619 without film as the research object,and conducted the global sensitivity analysis of Cotton2 K model under different irrigation levels based on the cotton field trial data in South Xinjiang region,and compared the same sensitivity analysis methods under six irrigation treatment conditions using TDCC and P value.Consistency.The global sensitivity of first flowering day,yield,plant height,stem dry matter mass,and whole plant dry matter mass among the five output results of Cotton2 K model was analyzed by Morris method and EFAST method using Sim Lab software,and the analysis results of the two methods were compared,and the uncertainty analysis of the analysis results of EFAST method was performed.Then the NSGA-III multi-objective genetic algorithm was used to optimize the sensitive genetic variety parameters.The simulation accuracy of the model was then validated using cotton observation data in 2020 and 2021.Finally,the cotton irrigation system of the calibrated Cotton2 K model was optimized under different climatic conditions to provide some theoretical basis and reference for the production and management of cotton in South Xinjiang.The main research results are as follows:(1)The sensitivity analysis showed that the sensitivity analysis results of the two methods on the first flowering day of Zhongmian 619 were identical,and the analysis results on yield,plant height and dry matter quality of the whole plant were highly correlated.All TDCCs were greater than 0.85 and all P values were less than 0.05,indicating that there was consistency in the statistical significance of the parameters in terms of sensitivity ranking under six different conditions of irrigation treatments from W1 to W6.The most sensitive parameter of Cotton2 K model to the first flowering day is Var30;Var41 is the most sensitive parameter to yield;The main parameters sensitive to plant height are Var35,Var01 and Var26;The main parameters sensitive to stem dry matter quality are Var16,Var01 and Var34;Var01 is the most sensitive parameter to the dry matter quality of the whole plant.With the change of irrigation amount,the ranking of sensitive parameters to yield and dry matter quality of the whole plant will also change.Therefore,sensitivity analysis should be carried out under different irrigation conditions,and the impact of the change of parameter sensitivity ranking should be noted.The uncertainty analysis shows that the simulation results corresponding to other outputs are distributed within the 95% confidence interval except for the plant height,which indicates that the overall simulation results of the model have high reliability.Compared with low water treatment,under high water treatment conditions,the probability of model error is smaller,and the simulation of yield is more accurate.The Cotton2 K model itself has a certain degree of deviation in the simulation of stem dry matter quality.Under different irrigation conditions,the simulation results of dry matter quality of the whole plant are basically normal distribution.The simulation results show that with the increase of irrigation amount,the measured value and the median value are closer and closer,and the model simulation effect is better.(2)NSGA-Ⅲ genetic algorithm was used to optimize the sensitive variety parameters.The optimized Cotton2 K model is the best for yield simulation,with n RMSE less than 10%;The simulation results of stem dry matter took the second place,and n RSME was less than 20%.The simulation result of plant height is the worst,which may be caused by the failure of the model to fully consider plant diseases,insect pests and climate factors.For the simulation of the growth period,60% of the daily errors are relatively large,but the errors of other growth period simulation results are relatively small.The model can provide the main growth period prediction.(3)In this paper,12 climatic conditions and 17 irrigation systems are set up.The study shows that the increase of temperature will reduce the yield and water use efficiency of cotton,but a reasonable irrigation system will reduce the negative impact of climate change on the yield and water use efficiency of cotton.According to the simulation results under different scenarios,taking into account the factors such as cotton yield and water use efficiency,T16 irrigation system can make the yield change with the minimum range under the temperature change,with high water use efficiency,and has good applicability.The results of this study provide a theoretical basis for the optimization and applicability of parameters of Cotton2 K model in the simulation process of filmless cotton in southern Xinjiang.