基于柑橘叶面积指数和产量的WOFOST模型关键参数校正

Key Parameter Calibration of WOFOST Model Based on Citrus Leaf Area Index and Yield

为进一步分析不同水肥管理条件对柑橘产量的影响使其更广泛的应用于不同区域,并探明WOFOST模型参数针对柑橘果树生长模拟时的参数敏感性结果和取值情况,开展2021-2022年柑橘不同生育期滴灌水肥控制试验,采用EFAST方法对WOFOST模型模拟叶面积指数(LAI)和产量所需的叶片生长、CO_(2)同化、同化物转换和叶片营养物质等25个参数进行了敏感性分析,使用PEST软件完成高敏感参数的优化,并利用田间试验观测数据对模型模拟的LAI和产量进行验证。结果表明,模型模拟LAI和产量的敏感参数相似,均包括叶面积指数的最大相对增长率、比叶面积、单叶片CO_(2)同化效率,叶片同化物转换效率,初始根长,区别为LAI模拟对茎同化物转换效率,积温等参数敏感,而产量模拟对叶片干物质最大钾浓度敏感;基于2021年数据校正后最优参数模型模拟的不同灌溉管理下LAI最大值(LAI_(m))模拟值与实测值之间R2=0.5800,NRMSE=15.10%,产量模拟值与实测值之间R^(2)=0.6471,NRMSE=6.81%;2022年数据验证表明模拟果实膨大期进行HWMF处理的果树模拟精度最高,果实膨大期各处理LAIm模拟值与实测值之间R^(2)=0.4726~0.8492,NRMSE=13.09%~15.26%,产量模拟值与实测值之间R^(2)=0.4063~0.7022,NRMSE=6.33%~11.33%。研究表明,校正后的WOFOST模型在田间尺度取得了较高的模拟精度,可以开展不同灌溉管理方式对柑橘生长影响的定量化分析。

In recent years,the citrus industry has developed into one of the agricultural and economic pillar industries in the hilly areas of Southwest China.After exploring the effects of different water and fertilizer management conditions on citrus yield,it is necessary to conduct crop model research for citrus fruit trees in order to apply them more widely to different regions.The feasibility of simulating crop growth using the WOFOST model has been widely recognized.Parameter calibration of the model based on data collected from field experiments can improve the accuracy of the model′s simulation of citrus,and thus provide better guidance for production practices.In this study,drip irrigation water and fertilizer control experiments were carried out for different growth stages of citrus from 2021 to 2022.The EFAST method was used to analyze the sensitivity of 25 parameters required for simulating leaf area index(LAI)and yield by WOFOST model,such as leaf growth,CO_(2) assimilation,assimilate conversion and leaf nutrient content.The PEST software was used to optimize the highly sensitive parameters,and the field experimental data were used to validate the LAI and yield simulated by the model.The results showed that the sensitive parameters for simulating LAI and yield by the model were similar,including maximum relative growth rate of LAI,specific leaf area,single leaf CO2 assimilation efficiency,leaf assimilate conversion efficiency,initial root length.The difference was that LAI simulation was sensitive to stem assimilate conversion efficiency,thermal time and other parameters,while yield simulation was sensitive to maximum potassium concentration of leaf dry matter.Based on the optimal parameters calibrated by 2021 data,R^(2) between simulated values and measured values of LAI_(m) under different irrigation management was 0.5800,NRMSE was 15.10%,R2 between simulated values and measured values of yield was 0.6471,NRMSE was 6.81%.The validation with 2022 data showed that HWMF treatment during fruit expansion period had the highest simulation accuracy.R^(2) between simulated values and measured values of LAI_(m) under various treatments during fruit expansion period ranged from 0.4726 to 0.8492,NRMSE ranged from 13.09%to 15.26%,R^(2) between simulated values and measured values of yield ranged from 0.4063 to 0.7022,NRMSE ranged from 6.33%to11.33%.This study showed that WOFOST model achieved high simulation accuracy at field scale after calibration,and can be used to carry out quantitative analysis of the impact of different irrigation management methods on citrus growth.