| Crop models for horticultural crops are an important link between the growth status of peppers and environmental conditions.They can not only predict the growth and yield of crops under different environmental conditions,but also predict the impact of future environmental changes in facilities on crop growth,which facilitates growers to adjust their production methods in time and improve their horticultural production potential.Crop growth models are usually modeled based on local environmental conditions,soil characteristics,crop varieties,and other factors,and cannot be directly applied to another area.Therefore,in this study,one experiment（T1）were conducted in asymmetrically insulated plastic greenhouses and two experiments（T2,and T3）were conducted in solar greenhouse in the Guan Zhong area,and local widely cultivated pepper varieties Donglang（S1）and Huanle（S2）were selected for modeling and validation.Among them,T2S2 was used to establish the model,and T1S1,T1S2,T2S1,T3S1,and T3S2 were used to test the model,and finally a model of pepper growth and development suitable for the facility environment in Guan Zhong,China was developed.The model consisted of a developmental stage model,a leaf area index model,and a dry matter accumulation,distribution and yield model.After the modeling was completed,we used the above model to simulate the cultivation length and yield of pepper during the nutritional growth period of the 24 planting nodes and simulate the economic returns of the 24 planting nodes by combining the price and planting cost of pepper.The main results of this study are as follows:On the developmental phase model:the study constructed and validated a developmental stage model for pepper in a facility using four methods:effective cumulative temperature method,linear segmentation function,exponential function,and power finger function,and two calculation steps（day and hour）.The results showed that the power finger function model with h as the calculation step and the linear segmented function model performed better,the exponential function was the second best,and the GDD method was the worst.The mean error days between the actual and predicted days for the three developmental stages of pepper bud emergence,flowering and commercial fruit ripening were 3.4 d,2.2 d and 5.2 d.The root mean square error between the simulated days and the actual observed days of the power function model ranged from 0.28 to 8.52 d,with the mean absolute error ranging from 3%to 17%and the mean relative error ranging from 0.67 to 8.00 d.On the leaf area index model:the study simulated single leaf area using elliptical model,geometric model and multiple linear regression model（MLR）（of which MLR established four expressions）,which were embedded in the thermal effectiveness and PAR to construct a leaf area index model of peppers.The best simulation result was the model with MLR4 embedded in the thermal effectiveness and PAR,and the simulation results showed that the R2 between the simulated and actual values of LA and RMSE and MAE of pepper single leaf area established by this method were 0.884 6,226.03 cm2 and 171.89 cm2;the R2 between the simulated and actual values of leaf area index LAI was 0.900 7,and RMSE was 0.06 m2,MAE was 0.05 m2/m2.On the dry matter accumulation,distribution and yield model:the coefficients of determination R2 and MRSE between the simulated and measured values of the dry matter accumulation model of pepper established in this study were 0.812 4 and 23.25 g/m2,respectively,which had good predictability in different greenhouses,different crop and different pepper varieties.The R2 and RMSE of the simulated and measured values of above-ground and below-ground dry matter masses of plants from the dry matter allocation model established in this study were 0.818 2 and 16.27 g,0.730 8 and 0.97 g,respectively;the R2 and RMSE of the simulated and measured values of dry matter masses of stems,leaves and fruits were 0.793 3 and 2.48 g/m2,0.827 1 and 1.57 g/m2 The prediction results of the yield model established in this study showed that the R2 and RMSE of actual and predicted yield of pepper were 0.818 5 and 0.33 g/m2.In terms of model application:the cultivation duration and yield of nutrient growth period were simulated for 24 planting nodes,and the economic efficiency of pepper was calculated for 24 planting nodes by combining planting cost and pepper price.In terms of cultivation duration,the cultivation duration of S2 nutritional growth period was lower than that of S1,and the cultivation duration of both was shorter when planted from mid-March to early September,with 61 d for S1 and S2,respectively;in terms of price,the peak of wholesale price during harvesting period occurred from mid-August to mid-October,and the highest wholesale price during harvesting period was 6.43 yuan/kg and 6.37 yuan/kg for S1 and S2,respectively;in terms of The highest simulated yield of S1 and S2 was 90.2%and 71.4%higher than the lowest simulated yield;in terms of planting cost,the cost of S1 and S2 was lower and fluctuated smoothly when planted from March to September,and the lowest planting cost was 76.1%and76.8%of the highest planting cost;in terms of net In terms of net income,the peak period occurred from mid-September to early November when planting was done,and the highest net income of S1 was 7 261.26 yuan/mu,1.84 times of the lowest net income;the highest net income of S2 was 10 336.28 yuan/mu,2.84 times of the lowest net income. |
PDF Full Text Request