选取扬麦13作为供试小麦品种,在拔节期、孕穗期和灌浆期设置4种不同遮阴持续时间(0、5、10、 15 d)的盆栽试验,模拟不同遮阴处理对小麦产量及其构成的影响.2年的盆栽试验结果表明,在不同生育期遮阴均导致小麦产量下降,且随遮阴时间延长,不同生育期小麦产量均显著下降.下降幅度呈现为拔节期>孕穗期>灌浆期,遮阴每增加1 d,产量分别下降1.53、1.30、1.16 g/株,表明拔节期是小麦阴害胁迫的敏感期.遮阴在不同生育期对小麦产量构成的影响存在显著差异.拔节期和孕穗期遮阴主要通过降低小麦穗粒数造成产量下降;灌浆期遮阴对小麦穗数和穗粒数的影响较小,主要通过降低千粒质量造成小麦产量下降.说明遮阴造成小麦减产与小麦所处生育阶段密切相关,在评估阴害对小麦产量的影响时,应考虑小麦所处生育期对产量及其构成影响的差异.

The objectives of this paper were to examine cultivar differences of aboveground rice plant morphology and structure at the early tillering stage, and analyze its potential correlations with yield. The 4 rice cultivars, Liangyou108 (hybrid long-grain rice, V1), 86You8 (hybrid round-grain rice, V2), Nanjing43 (round-grain rice, V3), and Yangtao6 (long-grain rice, V4), were grown in field. Their rice plant morphology and structure (RPMS) parameters and yield were measured. The 1st, 2nd, and 3rd leaf blade length, width, and bowstring length on the main stem, and the 1st leaf sheath length, tangent angle, and bowstring angle (between blade and main stem) among 4 cultivars at the 3th or 4th maximum blade length period on the main stem all had significant differences (P<0.01). The 1st leaf sheath length at the 4th maximum blade length period on the main stem had high positive correlation with yield, with significant (r=0.96) in 2007 at p<0.05, and the 1st maximum blade width on the main stem at the 3rd maximum blade length period on the main stem was negatively correlated with yield, with significant (r=-0.99) in 2008at p<0.05. These results provided an initial scientific basis for using morphological and structural parameters for forecasting early yield potential and selecting characters of plant morphological types for high yielding.

连阴雨是长江中下游地区小麦生产面临的主要气象灾害，渍水（渍害）和寡照（阴害）2种胁迫作用，是造成该地区小麦减产的主要原因。为定量研究渍水和寡照及其协同作用对小麦产量构成的影响，本研究选取该地区常见的小麦品种宁麦13和扬麦13号，在拔节期和灌浆期，设置对照、渍水、寡照和渍水＋寡照4种处理，以及3种不同胁迫持续时间（5、10、15 d）的盆栽试验，观测不同处理对小麦产量构成的影响。结果表明，未发生胁迫作用情况下，扬麦13号的产量高于宁麦13。拔节期宁麦13对渍水和寡照的耐受性要高于扬麦13号，灌浆期宁麦13和扬麦13号对渍水和寡照的耐受性相近。此外，拔节期不同胁迫对小麦造成的减产呈现为渍水＞渍水＋寡照＞寡照，表明寡照在拔节期对渍水胁迫存在补偿作用，而灌浆期不同胁迫对小麦造成的减产呈现为渍水＋寡照＞渍水＞寡照，表明灌浆期寡照对渍水胁迫存在叠加作用。结果表明，连阴雨造成小麦减产与小麦所处生育期有关，建立小麦生长模型渍害和阴害模块时，应考虑小麦所处生育期对其产量构成影响的差异。

春季湿渍害是江淮地区越冬作物生长的主要农业气象灾害，利用1961—2010年江苏省与安徽省155个县站逐日气象资料，分析了江淮地区春季湿渍害平均持续时间、平均发生次数、春季湿渍害过程降水量空间分布特征及不同灾害等级发生次数的时空分布规律。结果表明：春季湿渍害平均发生次数、过程的平均持续时间和累积降水量3个指标均呈随纬度降低而增加的特征；江淮地区春季湿渍害过程平均持续时间无明显的年际变化，但春季湿渍害过程发生次数呈减少趋势，而其累积降水量呈增加趋势；轻度湿害（4～6 d）发生的次数最多，约占57．3％，中度湿害（7～9 d）占25．5％，重度湿害（≥10 d）最少，约占17．2％。总体看出，江苏、安徽2省轻度湿渍灾害次数在波动减少，中度、重度湿渍灾害次数总体变化不大。

Waterlogging and shading result from continuous rain are the main meteorological disasters for wheat (Triticum aestivum L.) production. In order to evaluate the effects of waterlogging and shading on yield components of winter wheat (both independent and combined), pot experiments were conducted using two representative cultivars in local, Ningmai 13 and Yangmai 13. In total, 4 treatments, including CK (control), WA (waterlogging alone), SA (shading alone) and WS (both waterlogging and shading) were established with three duration (5, 10 and 15 d, respectively) at jointing and grain-filling stages. Results showed that, in the case of non-stressed environment, Yangmai 13 got a better production compared with Ningmai 13 (grain yield per plant was 14.25 g and 15.97 g for Ningmai 13 and Yangmai 13, respectively). However, compared with Yangmai 13, Ningmai 13 got a better yield under stresses at jointing stage, while a similar yield was observed when stresses are at grain-filling stage. By comparing wheat yield and its components, the negative effects of the stresses showed a tendency that WA > WS > SA at jointing stage, whereas WS > WA > SA at grain-filling stage. The result demonstrated that shading had a compensative effect on waterlogging at jointing stage while an addictive effect at grain-filling stage. Reduction of wheat production caused by continuous rain depended on the growth stages. Effect of growth stage on grain yield should be considered when waterlogging and shading packages of wheat growth model were established.

In order to make the RCSODS(rice cultivational simulation-optimization-decision making system) to better meet the requirements for the fine assessment of the Huaihe River Basin drought in rice-growing season,using the data of artificial control soil moisture experiment from Xinghua of Jiangsu province and Xinyang of Henan province in 2010 to 2012,the parameters of relevant sub-models in RCSODS were achieved,and the algorithms of drought stress factor in RCSODS were validated.The results showed that the correlation coefficients(R2) between simulated and measured net photosynthetic rate,root to shoot ratio,leaf area index were 0.947,0.882 and 0.950-0.988,respectively,indicating that the modification of RCSODS for drought stress was reasonable and the parameter of RCSODS was exact.The correlation coefficient(R2) between simulated and measured yields was 0.917,with the mean absolute error(MAT) being 0.551 and the root mean squared error(RMSE) being 0.623,suggesting that the new model is good in simulation and will meet the requirement of rice production in Huaihe River Basin.

Different submergence depths(completely submerged and half submerged) and submergence durations(1d,3d and 6d) treatments were conducted on two rice cultivars,Nanjing 46 and Yangliangyou 8 at the tillering stage to analyze the response of the aboveground matter distribution and yield components of different types of rice to submergence stress.The results showed that after the complete submergence treatment,significant decreases of the biomass ratios of green leaf,and stem and sheath to aboveground part and an increase in the percentage of yellow leaf biomass were found in both of the two rice types.The longer the submergence duration,the more the rice suffered,and the shower the recovery.In contrast with complete submergence treatment,the half submergence treatment showed no significant effects on the aboveground matter distribution and yield components of rice.The decrease of the panicle number per plant and the single panicle weight were responsible for the yield reduction after tillering stage submergence of rice.It was found that Yangliangyou 8 had a greater submergence tolerance than Nanjing 46,not only in the survival rate of the green leaf,stem and sheath,the recovery capability,but also in the yield formation after the submergence.

In this paper, the many indices used in validation of crop models, such as RMSE (root mean square errors), Sd (standard error of absolute difference), da (mean absolute difference), dap (ratio of da to the mean observation), r (correlation), and R2 (determination coefficient), are compared for the same rice architectural parameter model, and their advantages and disadvantages are analyzed. A new index for validation of crop models, dap between the observed and the simulated values, is proposed, with dap<5% as the suggested standard for precision of crop models. The different kinds of validation methods in crop models should be combined in the following aspects: (1) calculating da and dap; (2) calculating the RMSE or Sd; (3) calculating r and R2, at the same time, plotting 1:1 diagram.

To quantify the relationships between rice plant architecture parameters and the corresponding organ biomass, and to research on functional structural plant models of rice plant, this paper presented a biomass-based model of aboveground architectural parameters of rice (Oryza sativa L.) in the young seedling stage, designed to explain effects of cultivars and environmental conditions on rice aboveground morphogenesis at the individual leaf level. Various model variables, including biomass of blade and blade length, were parameterized for rice based on data derived from an outdoor experiment with rice cv. Liangyou 108, 86You 8, Nanjing 43, and Yangdao 6. The organ dimensions of rice aboveground were modelled taking corresponding organ biomass as an independent variable. Various variables in rice showed marked consistency in observation and simulation, suggesting possibilities for a general rice architectural model in the young seedling stage. Our descriptive model was suitable for our objective. However, they can set the stage for connection to physiological model via biomass and development of functional structural rice models (FSRM), and start with the localized production and partitioning of assimilates as affected by abiotic growth factors. The finding of biomass-based rice architectural parameter models also can be used in morphological models of blade, sheath, and tiller of the other stages in rice life.

The objectives of this study were to develop Rapeseed Cultivation Simulation-Optimization-Decision Making System (Rape-CSODS), and validate it in order to design rapeseed planting, regulate and control its growth and development, and fulfill its high yield, good quality, high benefits, and standard production eventually. The models in Rape-CSODS were developed based on field experiments with 3 cultivars, 2 plant types, 6 sowing dates, and 4 sites from 2002 to 2008 in the middle and lower valley of Yangtze river in China and relative data from references of rapeseed research, employed ideas of R/WCSODS (Rice/Wheat Cultivation Simulation-Optimization-Decision Making System), then Rape-CSODS were developed through integrating rapeseed growth models, optimization models, and expert knowledge. In that the rapeseed growth models mainly included phenophase, leaf age, population dynamic, and organ forming etc., and the optimization models for rapeseed cultivation mainly included the optimum biomass, the optimum LAI, the optimum shoot and ramification number dynamics models, and the optimum decision making models for nitrogen and water etc. As a case, the system has been applied in Shandong Province in this study, and the results showed that the scheme get by this systems can increase green leaf number before over-winter, 1.1; 1000-weight, 0.2g; and rapeseed yield, 676.5 kg ha(-1) at mature, comparing with rapeseed production in the same conditions. The Rape-CSODS with general adaptability, clear yield aim, better mechanism, and dynamic characteristic can get the optimum management sheet pre-planting, micro-adjustment sheet during growth for rapeseed under different climates, soil, cultivars, and yielding levels.

To build the dynamic model for rice root system according to the geometry parameters and to achieve visualization of the root system in combination with computer graphics,three rice cultivars,Liangyou108(hybrid long-shaped rice,Va1),86you8(hybrid round-shaped rice,Va2) and Nanjing43(round-shaped rice,Va3),were planted in pots.There are two different levels of fertilizer: fertilizer(N 3.6 g,P2O5 0.8 g,K2O 3.6 g),and no fertilizer(with loamy soil).Based on the experimental observations,the relationships between the root dry weight and root length,root volume,and root surface area were analyzed,respectively.The power function Y=axb was showed to describe above relationships.The RMSEs between the simulation values and the observation values of the root length per plant,the root volume per plant,and the root surface area per plant were 358.138 1 cm,2.089 1 cm3,and 136.325 2 cm2 in 2008 respectively,and 42.993 7 cm,0.314 9 cm3,and 14.072 8 cm2 in 2009 respectively,which revealed that the model established was reliable,accountable and universal.This research would lay a basis for developing function-structure rice models.

The objectives of this study were to quantify the relationships among morphologic characters of leaf and root system in rice during the period of elongation to booting.Three high yield rice materials of Jiangsu Academy of Agricultural Sciences,Liangyou108(indica hybrid rice,V1),86you8(japonica hybrid rice,V2),and Nangjing43(japonica rice,V3),were planted in pots with loamy soil.There were two treatments of fertilizer: fertilizer of 3.6 g N,0.8 g P2O5 and 3.6 g K2O per pot and no fertilizer.Based on the experimental observations,the multifactor linear regression(MLR) models between morphologic parameters of leaf and root system were built.The variance of three MLR models was significant,and the fitness of models was very good.The average absolute deviation between simulated value and measured value of total root length,total root surface and total root volume were-176.506 cm per plant,8.784 cm2 per plant,and-1.624 cm3 per plant(n=24),the corresponding standard error of difference per plant were 31.542 cm,15.653 cm2,and 2.117 cm3,and the corresponding correlation coefficients were 0.913**,0.904** and 0.907**,respectively.The MLR models between the characteristic of leaf and morphology of root system from elongation to booting would be help to link the research of above-ground part with that of root system,cooperate visualizing rice plants above-ground with under-ground,and select high yield plant types and cultivation practices.

The objectives of this study were to build the best-fit mathematical models for quantitative relationships between the individual plant yield and the main geometric properties of rice root system in different growth stages.Three rice cultivars,Liangyou 108(indica hybrid rice),86 You 8(japonica hybrid rice) and Nanjing 43(conventional japonica rice),were planted in pots,applying fertilizer at rates of 3.6 g N,0.8 g P2O5 and 3.6 g K2O per pot as treatment,and no fertilizer as control.Based on the experimental observations,the mathematical models were developed by regression analysis.Cubic model was the best-fit relationship between the geometric properties of rice root system and the individual plant yield except the best-fit quadratic equation between the yield of individual plant and the total root length per plant,and total root surface area per plant during jointing to end filling periods.Correlation analyses between yield per plant and various parameters of root system revealed that increasing total root length,total surface area and total volume per plant will be of benefit for raising rice yield per plant.

Objective The aims of this study were to build the biomass-based main geometric parameter dynamic models of rice root system according to the relationships between the total root length, the total root volume, the total root surface area per plant and the dry weight per plant of different rice cultivars under different fertilizer levels. Method Three rice cultivars, ‘Liangyou108’ (Hybrid long-shaped rice, V1), ‘86you8’ (Hybrid round-shaped rice, V2), and ‘Nanjing43’ (Round-shaped rice,V3), were planted in pots with two fertilizer levels: fertilizer (N:3.6 g/pot, P2O5:0.8 g/pot, K2O:3.6 g/pot) , and no fertilizer (with loamy soil) to obtain the root system morphological characteristics, the dry weight, etc., which were processed by SPSS v13.0 and Excel 2007. The models developed in this paper were verified using RMSE.ResultThe biomass-based dynamic models of the total root length were set up through analyzing relationships between the total root length and the biomass per plant from the experimental data in 2008, and then, the biomass-based dynamic models of the total root volume and the total root surface area per plant also were developed according to the principles of solid geometry, which were tested by using the experiment data in 2009. The results showed that the RMSE between the simulated values and the observed values of the total root length, the total root volume, and the total root surface area per plant were 490.660 cm/plant, 131.859 cm2/plant, and 2. 532 cm3/plant, respectively. Conclusion The models developed in this paper were reliable with mechanics, general adaptability, and interpretative characteristics in a certain extent, which would lay a basis for developing Functional-Structure Rice Models.

【Objective】 To develop a geometrical attribute model of leaf blade based on dry weight(DW),which is the output of growth model.【Method】 Firstly,the model was deduced from the relationship between biomass and geometrical parameters of leaf theoretically.And then,based on observations on leaf length and leaf width of different leaf positions(LP) on main stems under different nitrogen rates and densities with four cultivars(LY108,86Y8,NJ43,YD 6),the change patterns of specific leaf weight(SLW) with leaf position and the relationship between leaf length and width of rice on main stem were analyzed.【Result】 The results indicated that the relationship between leaf blade length and width could be expressed in power function,and SLW changed with the leaf position(LP) in quadratic equation.【Conclusion】 The validation of the model with independent experimental data under different nitrogen rates and densities with four cultivars showed that the RMSE of dynamic leaf length and leaf width on main stem were 2.55 cm and 0.06 cm,respectively,and the model could be used in simulating the geometrical characteristics of rice leaf blade on main stem under different growth conditions.

The objectives of developing optimum leaf area index dynamic models for rape (OLAIDM) was to develop Rape Cultivation Simulation-Optimization- Decision Making System(Rape-CSODS) , to design its planting , to regulate and control its growth and development, and to fulfill its high yield, good quality, high benefits and standard production eventually. The OLAIDM were developed based on field experiments with 3 cultivars, 6 sowing dates, 2 types of plant pattern and 4 sites from 2002 to 2007 in middle and lower valley of Yangtze river in China and relative data from references of rape researches, employed ideas of R/WCSODS (Rice/Wheat Cultivation Simulation- Optimization-Decision Making System), and in the same time, the OLAIMR and its parameters also were assessed, calibrated and tested. The average absolute deviation(de), correlation coefficients(r) and the standard errors of their absolute deviation(Sde) of between the observed and simulated values for LAI of two cultivars in Wuhan and Nanjing were -0.03~0.1533, 0.9707~0.9997 and0.1332~0.4032, respectively. 1:1 line of them were in Fig. 1 to 4. Multi-factors such as the ramification types, cultivars, and light et al. were taken into account in this study, therefore, the OLAIDM with general adaptability, clear yield aim, mechanism, and dynamic characteristic can simulate optimum LAI dynamic for rape under different sites, cultivars and ramification types, and yielding levels.

为了集成作物生长模型、决策支持系统与Web技术,以实现作物栽培数字决策技术的网络化.以武运粳7号、扬稻6号、粤优948及南粳41为供试材料,2005年在江苏省农科院试验区进行了不同基肥田间试验,采集和更新了品种参数,采用QLServer2000建立水稻品种、土壤及气象资料数据库,采用VS.Net进行了水稻生长模拟、决策支持系统与Web的集成和页而设计,包括登录、系统主界面、品种参数管理、地点气象数据管理、参数调整以及水稻栽培决策等功能.基于Web与模拟模型的水稻栽培数字化设计符合TCP/IP协议,可在服务器上(IIS5.0)安装、运行和在任何Intemet浏览器下调用,它继承了水稻栽培模拟优化决策系统的机理性、普适性和实川性,将水稻生长模型与'Web技术相结合,形成了水稻模拟优化决策的网络平台,使水稻土产区农业技术人员通过Intemet因地、因种、因时制定具有动态、目标和数字化特征、可下载的播前优化方案,提供当年调控建议,对作物生长模型的网络化具有重要意义.

In order to integrate web with crop growth simulation and decision-making support system, the field experiment of different basal levels was carried out in experiment area of Jiangsu Academy of Agricultural Sciences in 2005 adopting 4 cultivars such as "Wuyungeng 7", "Yangdao 6", "Yueyou 948" and "Nangeng 41", which mainly were used in collecting cultivar parameters and updating database of them. The database of rice cultivars, soil and weather data were developed using SQL Server 2000. The pages of digital design system of rice cultivation based on web and simulation model (DDSRCBWSM) were designed using Visual Studio.Net, which included register, the main page, cultivar parameter management, site data management, parameter adjustment, decision making for rice cultivation, and so on. The DDSRCBWSM accorded with TCP/IP agreements, which could be installed and run in server (IIS5.0), and be browsed on internet, it inherited mechanism, universal adaptability and utility of rice cultivation simulation-optimization-decision making system (RCSODS), combined web techniques with research of rice growth models, set up web system of RCSODS, made agricultural technicians of main rice production area of china gain pre-sowing optimization cases and rice management suggestion of the current year with dynamic, goal and digital characteristic in accordance with soil, cultivar and weather conditions online, and to fulfill technical direction through many kind manner such as paper, internet, email, television, wall newspaper, and so on, eventually.

将自主研制的水稻栽培计算机模拟优化决策系统(RCSODS)改造为适于全球气候变化影响评价的水稻效应模型(RCCMOD);利用中国南方9个有代表性样点的水稻区域试验资料与同期同地的天气资料,对RCCMOD进行可靠性检验;将RCCMOD分别在研究区域基准气候(Baseline)和3种基于大气环流模型(GCMs)的(2×CO2)气候变化情景下运行,并将模拟结果与同样条件下CERES-Rice模型的模拟结果进行了比较,两者基本一致.

利用CERES-Rice做模拟试验,分析了施氮量和氮肥运筹对水稻产量形成的影响.结果表明:当施氮水平较低时,CERES-Rice能较好地模拟施氮量对水稻产量形成的影响,但在施氮水平较高的情况下,CERES-Rice的模拟结果不太理想.模型能反映不同N肥运筹下模拟水稻产量的差异,但不能反映高产栽培的施氮特征.结合试验资料进一步分析认为,误差产生的原因是模型中仅用氮素丰缺因子对光合效率进行订正,而没有考虑氮素对物质分配和转移的作用.