Simulation Of Winter Wheat Growth In Hebei Piedmont Plain Based On Aqua Crop Model

Piedmont of Taihang Mountain is the main grain producing area in Hebei Province,and the main crops here are winter wheat and maize.There is rich in light and heat resources,but is short in water resources,Piedmont of Taihang Mountain is the water and heat resources restricted area.In this paper,Piedmont of Taihang Mountain is the research region,and winter wheat is the research object,the dynamic change law of the soil water content,soil nitrate nitrogen content,and the growth index of winter wheat in different experimental treatment which included three irrigation levels(W3,W2,W1)and three fertilization levels(F3,F2,F1)were analyzed.Based on most recent three years meteorological data and and the winter wheat growth index in each experimental treatment,the Aqua Crop model and BP neural network model were constructed based on the climate factors and plant growth index,and the predicted results of the two models were compared with the measured values in this paper.The research results can provide scientific basis for the study of reasonable irrigation,fertilization and yield prediction of winter wheat in the study area.The main results are as follows:1.Through the dynamic analysis of soil water content in different growth stages of winter wheat,it can be concluded that the soil water content in 0～60cm soil layer at jointing stage changes violently,and the change range is the largest.The change of soil moisture content in 40～60cm was mainly due to the water consumption of crop root growth;During heading stage,there was no significant difference between the surface soil moisture content and the bottom soil moisture content due to the influence of rainfall,irrigation and low temperature.Through the dynamic analysis of soil nitrate nitrogen content in different growth stages of winter wheat,it can be concluded that:when winter wheat is harvested,the nitrate nitrogen content in soil layer 100cm below the ground(the maximum soil depth in the experiment)of each treatment is determined,and the nitrate nitrogen concentration in W1F2 treatment is the lowest,with the value of 15.66mg·kg-1,It can be seen that W1F2 treatment is more suitable to be popularized in the field experiment in the perspective of non-point source pollution control.2.Based on the meteorological data in recent three years and the data of irrigation and fertilization in each experimental treatment,the parameters of Aqua Crop model were calibrated,and the winter wheat yield prediction model suitable for the region was constructed,and the winter wheat yield under each treatment was predicted by using the model.The results showed that the root mean square error(RMSE)between the measured and simulated yield of Winter Wheat in W3,W2 and W1 treatments were 0.038～0.069,0.03～0.05 and 0.01～0.07,respectively;The results showed that the Aqua Crop winter wheat yield prediction model had high precision,and the model was suitable for winter wheat yield prediction in Taihang piedmont plain.3.There are 9 test treatments designed in this test,and each treatment samples are repeated 3 times.Among 27 groups of experimental data,21 groups of experimental data were randomly selected for neural network simulation training,3 groups of experimental data were used for model validation,and 3 groups of experimental data were used for winter wheat yield prediction.The R2 between simulated and predicted values of random results is 0.906.The results show that the MAPE and MAE of Aqua Crop model and BP neural network are 0.04,0.02 and 0.053,0.47 respectively.Through the analysis,the Aqua Crop winter wheat yield model constructed under the climate background and the BP neural network winter wheat yield prediction model constructed under the background of plant growth index can be found that the two models can be applied to the winter wheat yield prediction in the water and heat limited area in front of Taihang Mountain in Hebei Province.