Research And Application Of Crop Models In Tobacco Production In Xuchan

Tobacco is one of the most important economic crops in China,which is related to the stability and development of China’s fiscal revenue.Today,when advocating technology to empower production and promote high-quality development in all walks of life,digital agriculture and smart agriculture have penetrated into all aspects of agricultural production,The crop model that can dynamically describe the process of crop production and development has become one of the powerful tools to promote the development of digital agriculture and smart agriculture.However,most of the current researches on crops using crop models are aimed at grain crops,and few of them are actually used to guide practice.This paper combines field experiments with crop models to explore the simulation of tobacco growth in Xuchang,and analyzes the feasibility of applying the models to serve and guide actual production,with a view to providing technical support for Xuchang tobacco to carry out meteorological science and technology services.The main conclusions of this study are as follows:(1)By using the data collected from field experiment to calibrate and test the WOFOST model and Aqua Crop model,both models can accurately simulate the growth and development of tobacco in Xuchang,but the simulation effect of WOFOST model is better than that of Aqua Crop model.The two models were calibrated by using the field test data of Xiangcheng in 2021(Xuchang Xiangcheng County for short),and verified by using the field test data of Xuchang in 2021(Xuchang City for short),Xiangcheng in 2022 and Xuchang in 2022.The WOFOST model was used to verify the leaf dry weight,aboveground biomass and leaf area index.The average NRMSE and R~2of the simulated and measured values were 15.11-16.66%and 0.83-0.98respectively;The Aqua Crop model is also used to verify the three.The average NRMSE and R~2of the simulated and measured values were 18.16-24.88%and 0.71-0.97 respectively.Both models can be used for the simulation of tobacco in Xuchang,but the WOFOST model is more accurate and simulates the dynamic change process of tobacco leaf growth,so the WOFOST model is selected for subsequent application research.(2)The precipitation during the important period of tobacco leaf formation in the past 30 years is divided into typical hydrological years using Pearson-III frequency curves,and the irrigation strategy in the water-scarce year is optimized based on the WOFOST model,so as to providing a rough reference value for irrigation in the water shortage year in the future.In most water shortage years,90 mm of water for simulated irrigation in 150-160 days(prosperous period)can meet the water shortage demand,and the irrigation effect is better.Secondary irrigation is required in the year of extreme water shortage in history,with a single irrigation volume of 50 mm and the first irrigation time is 110-130 days(transplanting-rooting period);In case of extreme water shortage in the early stage,the second irrigation should be conducted at 150-160 days(flourishing period),otherwise,it should be conducted at 170-200days(flourishing period or early stage of maturing period).For the second irrigation in 2014,which was the most serious water shortage in the past 30 years,it still could not meet the growth needs of tobacco plants,and three times of irrigation were needed.Namely at 120-130days(transplanting-rooting period),170-180 days(flourishing period)and 190-200 days(early stage of maturing period)irrigation once with 50 mm of water,can significantly improve the yield and avoid the impact of water stress.(3)Tobacco irrigation service,yield forecast and growth status evaluation based on WOFOST model.(1)Provide real-time irrigation service for Xuchang Tobacco in 2022.When the"Dry"index in the model shows three consecutive days of drought,a service opinion of 50 mm irrigation is given,which has achieved good results in actual production.(2)Using numerical prediction products to obtain meteorological forecast data on a 10 day scale for yield prediction,the accuracy of this method can reach over 96.5%.The multi-year historical meteorological data with four time scales of 30 years,20 years,10 years and 5 years were used to replace the meteorological data in the forecast period,and the average value of leaf dry weight was taken as the yield forecast value.It was found that this forecast method could reach more than 92.5%in the three reporting days.Among them,the time scale of 30years has the highest prediction accuracy(up to 96.66%),which could better reflect the climate average state of tobacco plant growth,reaching 96.66%.The similar year type method was used to replace the forecast meteorological data of Xuchang in 2022,and the forecast accuracy was 92.07%,slightly worse than the previous two methods.(3)The growth status of Xuchang tobacco in 2022 was evaluated in order to give targeted production and management suggestions.The study found that the light and temperature conditions in this year were excellent,but the precipitation conditions were poor.Synchronization and proper coordination of rain and heat are the necessary conditions to promote the high yield of tobacco leaves.