###
DOI:
本文二维码信息
WOFOST模型对吉林省土壤湿度同化的响应
刘维1, 王冬妮2, 侯英雨1, 何亮1
(1.国家气象中心;2.吉林省气象科学研究所)
Responses of Assimilating Soil Moisture Data in Jilin Province Using WOFOST Model
Liu Wei1, Wang Dongni2, Hou Yingyu1, He Liang1
(1.National Meteorological Center of China;2.Jilin Institute of Meteorological Sciences)
摘要
相似文献
本文已被:浏览 38次   下载 7
投稿时间:2017-11-25    修订日期:2018-04-20
中文摘要: 本研究利用吉林省白城站实验数据进行模型参数调整,利用独立的观测资料对生育期、叶面积指数、地上部分各器官生物量进行模拟验证与评价。以白城站和榆树站代表吉林省西部玉米种植区和中部黄金玉米带参数,利用农业气象观测站发育期资料、气象资料和经过质量控制后的逐日土壤水分自动站观测数据,采用强迫法将逐日自动土壤水分站观测数据同化进WOFOST模型,以此来模拟2001-2016年春玉米穗生物量变化状况,构建玉米土壤-降水改善率PD指标,来表征降水驱动和土壤水分驱动对作物模型模拟结果的影响。结果表明:(1)模型对白城站春玉米生育期的观测值与模拟值线性回归决定系数R2为0.999,生育期的模拟精度较高。(2)通过独立资料对模型参数验证,叶面积、地上部分总生物量和叶生物量三者整体模拟性能(EF)较好;残差聚集指数(CRM)较小,表明模型对三者的模拟基本较准确;而穗生物量模拟效果欠佳。(3)2001-2016年吉林中部和西部春玉米穗生物量年际间波动较大,西部玉米穗生物量明显低于中部黄金玉米带。(4)同化土壤水分后,西部地区PD值显著提高,中部黑土地区PD值略有提高;西部地区对水分的敏感性高于中部地区。(5)从代表站白城来看,2015年降水偏少的年份土壤模拟效果明显优于降水驱动;降水量偏多的年份模拟效果接近。(6)总体来说,以盐碱土为主的地区在降水量偏少的年型下土壤驱动效果优于降水驱动;在以黑土为主的区域或降水偏多的年型下,两者模拟效果基本接近。
Abstract:Crop model evaluation is a key and essential process in its application. The experimental data include agro-meteorological station, soil moisture data and evaluated by the dataset in Baicheng was used to calibration the crop model parameters. Independent observed data with spring maize was used to verify and evaluate the adaption. By the comparison between simulated and measured values of development stage, leaf area index, total biomass, biomass for each organ and performance of WOFOST were evaluated .The crop model parameters in Baicheng and Yushu were used to substitute the parameters in the West and Middle Jilin Province. The developmental information and the meteorological data and the daily soil moisture data under optimization from automatic soil observation stations were used in the WOFOST model to simulate the biomass from 2001 to 2016.While he daily soil moisture data was instead the precipitation data in model using the method of constrained. The improvement rate of soil-precipitation (PD) was constructed to analyze the different changes by using soil moisture data and precipitation data. The result showed that: 1)The model could simulate the phenology of spring maize, and the R2 was 0.999 by measured and simulated data. 2) The regression coefficient and coefficient of determination of each biomass organ simulated well, which passed the significance test. The modeling efficiency, EF was 93.3% for above ground total biomass and 87.8% for LAI. CRM for the model were 10.5% for above ground total biomass and 0.2% for LAI. But the modeling efficiency, EF was 55.8% for stored biomass. 3) The simulated stored biomass in the West and Middle Jilin province from 2001 to 2016 was high fluctuation in different years, and the simulated stored biomass in West was far below the biomass in Middle. 4) The PD was obviously improved in West and little improved in Middle. It was much more sensitive by water content in West.5)The simulation effect by soil moisture data was better than the precipitation data in the year of less precipitation in the typical station Baicheng. While the simulation effect by soil moisture data was close to the precipitation data in the year of more precipitation.6)In general, these results showed that the WOFOST model using soil moisture data can improve the simulation effect in these places where the soil type was saline-alkali soil and in the year of less precipitation. Also model using soil moisture data can little improve the simulation effect in these places where the soil type was backland and in the year of more precipitation.
文章编号:201711250509     中图分类号:S161    文献标志码:
基金项目:公益性行业(气象)科研专项(GYHY201506001);国家自然科学基金(41705095);国家气象中心作物模型业务应用创新团队
引用文本:
刘维,王冬妮,侯英雨,何亮,0.[en_title][J].Meteor Mon,():-.
Liu Wei,Wang Dongni,Hou Yingyu,He Liang,0.Responses of Assimilating Soil Moisture Data in Jilin Province Using WOFOST Model[J].Meteor Mon,():-.