Effect And Simulation Of Water And Nitrogen On Fruit Quality Of Muskmelon In Greenhouse

As"Xi Tian 208"as research object, this thesis presents the effects of different nitrogen elements as growth nutrients on the quality of fruits of muskmelon in green house. The N levels we applies are 0, 58.125, 116.250, 174.375, 232.500, 465 kg/hm~2. Based on the data from dynamical measurements and their corresponding fruit qualities, we established a dynamic model which reveals relationships between the growth qualities and the relevant N concentrations.Different N levels have significant influences on the quality of fruits. The inflexion, as a result of high N level which restricts the concentrations of N, P and K, is 174.375 kg/hm~2. Therefore in this study, we apply the N level between 58.125 and 174.375 kg/hm~2.An obvious positive correlation lies between the N concentration of fruit and the N level. We carried the stepwise regression analysis on fruit nutrient quality and established the optimum regression equation on the N, P, and K concentrations of fruit as the factors fruit total soluble sugar, soluble solid, and soluble protein content. The decisive coefficients R2 of regression equations are all above 0.85 and the F test's level of equations are all below 0.05, which indicates that our model could predict the growth quality relatively precisely. The experiments were carried out to study the effects of water-N level coupling on melon's fruit quality and establish the effective equation. We found that N is the dominant factor; water is the second primary factor on melon's nutrient quality. The highest soluble sugar, soluble solid and soluble protein was obtained by using Irrigating 171m3/667m~2, applying N 198.38kg/hm~2.The N levels we applied here are 0, 58.125, 116.250 and 174.375 kg/hm~2. By regression analysis of different N levels on fruit formation stage, we established the dynamic simulation model on melon in greenhouse. Fruit diameter, length and fruit thickness with the change of∑EATn fits Logistic function; fruit shape index with the change of∑TDn fits power function; fruit fresh weight, dry matter and fruit soluble solid with the change of∑TDn on fruit formation stage fits Logistic; fruit soluble total sugar with the∑TDn on fruit formation stage fits exponential function; fruit Vc, soluble protein respectively with the∑EATn and∑TDn fits quadratic function; fruit water content with the change of∑TDn fits Gompertz function. Besides, using fruit N concentration as impact factor simplified the models established under the different N levels. Except the fruit water content model established under different N levels, the other simplified models all got through F test and F test's decisive coefficients of these models were all above 0.85.Using the materials and different species in fall within one year validated the established models on the balanced fertilization level. The result showed that the models could simulate the change of fruit diameter and length, fruit thickness, fruit shape index, fruit fresh weight, soluble solid, total sugar, Vitamin C, soluble protein, water content and fruit dry matter with the their key environment factors. The RMSE of the simulated values with actual values were: 0.86cm,0.7199cm,0.1883cm,0.132,8.633g,0.965%,1.2569%,1.7163mg/100g,0.748mg/g,1.5964%,0.1368g. Therefore, from the Verification results, the models established under different N levels were all basic reliable and also realized the forecasting on fruit's evey quality characters. Because of the limitation of experimental conditions and time, the model's related parameters and results all need further optimization and inspection.