基于WebGIS与知识模型的小麦病虫草害管理决策支持系统研究

为给小麦病虫草害管理提供远程预测和咨询服务,通过分析和提炼小麦病虫草害管理方面的最新研究结果和信息,根据小麦病虫草害的发生规律,结合小麦栽培管理知识模型以及经济信息,构建了小麦病虫草害管理模型;并运用软件工程的思想,以WebGIS为空间信息管理平台,结合MS-SQL Server数据库技术,建立了基于B/S结构的网络化小麦病虫草害管理决策支持系统.该系统实现了用户管理、地图操作、病虫草害管理、智能学习、农药查询以及帮助等功能.以江苏省为案例区,对系统进行了测试与检验,结果表明,该系统能根据不同地域内小麦病虫草害发生的实际情况,预测其对小麦的危害程度并生成适宜的防治措施,从而为小麦病虫草害管理提供了远程预测和决策咨询的服务工具.     

棉花生产数字化设计与管理技术的示范初报

棉花生产数字化设计与管理技术的示范应用项目旨在将信息农业技术应用于棉花生产管理的优化设计和决策,以网络化农情信息管理系统,具有广泛时空适应性的棉花生产方案及管理指标设计的动态知识模型,基于知识模型的棉花生产数字化设计与管理系统,以及基于知识模型的便携式数字化管理仪等为示范推广的重点技术支撑,以大丰市为试验示范基地,通过逐级培训以及系统的示范应用,提高棉花生产系统的科学管理水平和综合生产力,促进全省数字棉作和数字农业的发展,以农业的信息化带动农业的现代化.     

基于模型和GIS的水稻生产管理决策支持系统构建与应用

为实现水稻生产过程的精确化管理,以水稻精确管理知识模型为管理决策的智能依托,以GIS为空间数据管理平台,应用系统工程思想和软构件技术,设计实现了基于模型和GIS的水稻生产管理决策支持系统。该系统具有基本地图操作、信息管理、栽培方案设计、肥水运筹、病虫草害管理、适宜生长指标设计、因苗实时调控、专家咨询以及系统帮助等功能。系统在江苏和浙江部分水稻生产区的应用结果表明,精确管理田块平均产量达10464.8kg/hm2,较常规管理田块增产12·53%,减少氮肥投入31.5kg/hm2。系统的建立为水稻生产全过程变量管理处方的生成提供了模型化和系统化的决策工具,也为其他作物管理决策支持系统的构建提供了基础框架。     

玉米栽培管理知识模型系统的设计与实现

在广泛搜集整理玉米栽培管理知识、数据和专家经验的基础上,对玉米生育指标和调控技术的地域性和季节性变异规律进行系统分析和量化表示,构建了基于作物-环境关系的、可适用于不同时空环境的玉米栽培管理知识模型系统.系统功能主要包括播前技术方案设计和产中动态调控指标预测两大部分.播前方案设计按照玉米栽培管理流程和生育进程来设计,具体包括:产量目标确定、品种选择和品质目标、播期确定、密度和播种量设计以及水肥管理等;产中动态调控生育指标主要考虑叶龄指数、叶面积指数、干物质积累和粒叶比.该系统克服了传统专家系统区域性、分散     

基于遥感的作物生长监测与调控系统研究

为了利用遥感技术快速无损实时地监测作物的生长、产量和品质状况,达到精确作物管理的目的,在集成作物光谱无损监测模型与作物管理知识模型的基础上,利用系统设计原理和组件化程序设计思想,构建了基于遥感的作物生长监测与调控系统,可实现从小尺度单点到大尺度农区的作物生长实时监测与调控.该系统具有文件管理、图像处理、地物分类与识别、植被指数计算、生长指标监测、生理指标监测、产量与品质预测,实时诊断与调控、工具管理以及系统帮助等功能.以江苏省泰州市的小麦为例对该系统进行实例分析,结果表明,该系统能有效实现作物生长监测与调控功能,从而为作物监测与管理的动态化和数字化奠定了基础.     

玉米生产管理专家咨询系统的研究

本研究经3年多点试验,建立了玉米生产管理专家咨询系统.该系统首次把人工智能和专家系统与模似模型结合起来,以解决有关玉米生产管理问题,该系统由天气模型、CERES—玉米模似模型、专家系统及其知识模型和经济模型组成;该系统能够对不同的栽培管理措施组合(地区、年份、品种、密度、播期、施肥等)的玉米生长发育状况、产量高低和经济效益等作出预测、评价和决策,从而使用户得到了与自己愿望相符的玉米生产管理方案.试验结果表明,该系统的可靠性好,预测性强,决策功能完善,是进行现代化玉米生产管理的理想工具.1991～1992年的示范推广取得了明显的经济效益.     

"棉花生产管理专家系统"及其应用效果

"棉花生产管理专家系统"是农业部重点项目"八五"优势棉基地县科技服务85农-12专题,经八年的时间研制开发而成,是湖北省第一个用于指导棉农进行棉花生产管理的专家系统,该系统运用人工智能,知识工程和专家系统等技术,总结了湖北省品种、肥料、植保、化控、栽培等领域专家多年从事棉花生产和研究积累起来的知识、经验和关键技术,建立了包括品种选择、周年管理、全程化调、施肥推荐、诊断防治等五个主要功能子系统的专家系统.该系统能为棉农从事棉花生产管理的全过程提供详细周到的服务,达到人类活专家的水平.     

灌水方式对不同小麦品种水分利用效率和产量的影响

为给冬小麦生产中合理灌溉提供依据,以河农859和豫麦49为材料,设置三种灌水方式,即S1(冻水 拔节水)、S2(拔节水)和S3(拔节水 灌浆水),主要研究了不同水分处理下2个品种各生育时期土壤含水量的变化动态、水分利用效率和产量性状.结果表明,在拔节期、灌浆前期和生育后期,S1、S2处理0～80 cm土层分别处于不同程度的水分胁迫条件下,S3处理0～40 cm处于不同程度胁迫.豫麦49 的水分利用效率以S3处理最高,且显著高于S1和S2;河农859的水分利用效率以S2、S3较高,且显著高于S1,同时S2与S3的产量差异不显著,说明适度水分胁迫在不影响河农859产量的同时还可以提高水分利用效率.河农859籽粒产量以浇2水的S3较高,但与浇1水的S2产量差异不显著;而豫麦49在S3处理下的产量显著高于S2和S1.表明河农859抗旱性较强,只浇拔节水即可以在基本满足产量形成需要前提下提高水分利用效率;而豫麦49只有在供水充足条件下才能发挥其高产潜力.     

大小麦两种土传病毒病害的轮茬发病观察

在我县沿海地带的冈东、上冈、草埝的乡镇及沿海辐射地域里下河地区,从2 0世纪80年代以来因大小麦的品种更新和免耕麦、套寄麦的麦作耕作制度的变革,随之发生大麦黄花叶病(Barleyyellowmosaicdisease)和小麦发生梭条斑花叶病(WSSMV) 〔1〕,这两种病毒病害严重影响大小麦经济产量和品质,一般减产30 %左右;而这两种病毒病害又都是土壤传播,以土壤中的禾谷多粘菌(PolrmyxaqraminisL)为介体的真菌传毒,真菌本身不致病,也不引起任何症状〔2〕。我们多年来对大小麦这两种病害的轮茬种植进行试验观察,探索发病规律和栽培防治。1　材料与方法1 1…     

基于生长模型的稻麦轮作生产管理决策支持系统研究

为建立一种具有时空适应性、适用于稻麦轮作生产的管理决策支持系统,运用系统分析原理和数学建模技术,构建了具有较强机理性和预测性的稻麦轮作生长模拟模型,并进一步结合气象生成模型与策略分析评价模型,以C#.net为编程语言,应用面向对象的程序设计与软件技术,建立了综合性的稻麦轮作生产管理决策支持系统.系统具有农田信息管理、专家咨询、动态模拟、方案评估、因苗预测和系统维护等功能,实现了模拟预测、决策支持和人机交互技术的有机耦合.系统的建立为稻麦轮作生产管理和策略制定提供了定量化的辅助决策工具.     

A Knowledge-Based Model for Nitrogen Management in Rice and Wheat

Abstract

Excessive nitrogen fertilization results in low nitrogen-use efficiency. To improve nitrogen management for high yield and high nitrogen efficiency in rice and wheat, we developed a knowledge-based nitrogen fertilization model by integrating the quantitative relationship between N fertilization and yield target with respect to N supply and demand balance. The total amount of nitrogen and ratio of basal to top dressing nitrogen could be determined by this nitrogen fertilization model, and the desirable nitrogen fertilizer strategies could be made under the conditions of different climates, soil types and managements. Furthermore, the function of dynamic regulation of pre-designed N dressing rate could be determined by using the nitrogen fertilization model on the basis of actual growth status under a specific production system. The nitrogen fertilization model is evaluated using the data from field experiments of rice and wheat at Nanjing, and the results on crop growth pattern and N use showed that the grain yield and N recovery were markedly improved by the N fertilization plan given by the model. The nitrogen fertilization model can be used as guidance for quantifying N fertilization strategy in cereal crop production.     

The effect of tillage,crop rotation and residue management on maize and wheat growth and development evaluated with an optical sensor

Crop growth and development as well as yield are the result of the efficiency of the chosen agricultural management system within the boundaries of the agro-ecological environment. End-of-season yield results do not permit the evaluation of within-season management interactions with the production environment and do not allow for full understanding of the management practice applied. Crop growth and development were measured during the 2004, 2006 and 2008 crop cycles with an optical handheld NDVI sensor for all plots of the different management treatments of a long-term (since 1991) sustainability trial in the highlands of Mexico. Cropping systems varying in (1) tillage (conventional vs. zero tillage); (2) residue management (retention vs. removal); (3) rotation (monocropping vs. a maize [Zea mays L.]/wheat [Triticum aestivum L.] rotation) were compared. The NDVI-handheld sensor was evaluated as a tool to monitor crop growth and development and was found to be an excellent tool for this purpose. There was a strong relation between NDVI and biomass accumulation of maize and wheat. The measurement with the handheld sensor was non-destructive and fast so that a representative plot area could be measured easily and time-efficiently. Zero tillage induced different crop growth dynamics over time compared to conventional tillage. Zero tillage with residue retention is characterized by a slower initial crop growth, compensated for by an increased growth in the later stages, positively influencing final grain yield. Also crop rotation influenced early crop growth, with lower NDVI values for crops sown after wheat than crops after maize. Zero tillage with residue removal had low NDVI values throughout the growing season. Zero tillage with retention of crop residues results in time efficient use of resources, as opposed to conventional tillage, regardless of residue management, and zero tillage with residue removal. The results indicated that different tillage, rotation and residue management practices influence crop growth and development. It is important to monitor and understand crop growth under different management systems to select the right varieties and adjust timing and practice of input supply (fertilizer, irrigation etc.) in a holistic way in each cropping system.     

基于无人机可见光遥感的冬小麦株高估算

株高是作物生长过程中重要的生长指标。为探索快速准确获取作物株高的方法,利用无人机可见光图像采集系统,获取冬小麦拔节期至成熟期的高清数码图像,建立冬小麦拔节期、抽穗期、灌浆期及成熟期的作物数字表面模型(digital surface models,DSM)及作物高度模型(crop height model,CHM),并对模型进行验证。结果表明,冬小麦株高各生育时期CHM提取值与地面实测值极显著相关(P<0.01),误差为-0.10~0.09m,相对误差为17.64%~19.60%。株高预测值与实测值拟合性较高(R^2=0.82,RMSE=4.31cm)。这说明用无人机拍摄的高清数码影像可快速估算冬小麦的株高。     

Water availability at sowing and nitrogen management of durum wheat: a seasonal analysis with the CERES-Wheat model

In water-limited environments soil water content at sowing is important in determining durum wheat germination, emergence and plant establishment. Soil water content interacts greatly with soil nitrogen content, affecting nitrogen uptake and crop productivity. Simulation models can be used to confirm the optimal strategy by testing several crop management scenarios.The CERES-Wheat model, previously calibrated and validated in southern Italy, has been used in a seasonal analysis to optimise nitrogen fertilisation of durum wheat at different levels of crop available water (CAW) at planting date in southern Italy. The simulation was carried out for a 48-year period with measured daily climatic data. The 99 simulated scenarios derived from the combinations of different CAW levels at sowing, nitrogen fertiliser rates and application times.The results obtained from the simulation indicated that the effect of CAW at sowing was relevant for durum wheat production at lowest and highest values, while the optimal sowing time to maximise yield and profit can be considered when CAW is 40–60%. In the case study optimal N fertiliser amount was estimated to be 100±20 kg ha⁻¹, from a productive, environmental and economic point of view. The nitrogen split application—half at sowing and half at stem extension stage—resulted in the best management practice.This application of the CERES-Wheat model confirmed the capability of the model to compare several crop management strategies in a typical durum wheat cropping area.     

中国小麦茎基腐病的发生现状及防治策略

小麦茎基腐病是近十年来危害中国小麦生产的主要病害之一。本文综述了小麦茎基腐病在中国的发生现状、发病过程和致病机理,分析了发病的田间诱因以及耕作、管理、栽培、生物等不同技术对茎基腐病的防控效果,提出了在不同的农业生态区,需根据当地的气候条件和生产状况,通过耕作制度优化、多样化种植、高抗品种选育及高效生防菌群筛选等多措并举,实行综合防治,以期为提高小麦茎基腐病的防治效果及保障中国小麦的安全生产提供参考信息。     

基于遥感信息和WOFOST模型参数同化的冬小麦单产估算方法研究

为探讨遥感信息和作物生长模型在作物估产方面的优势互补特性,选取河北省藁城市冬小麦作为研究对象,采集多个关键生育时期的生理生化、农田环境、气象等数据,并获取准同步的环境减灾小卫星HJ-CCD影像数据,采用植被指数反演冬小麦叶面积指数（LAI）,基于扩展傅里叶振幅灵敏度检验法(EFAST)对WOFOST作物模型的26个初始参数进行全局敏感性分析,筛选敏感性参数,调整WOFOST模型的核心参数,利用查找表优化算法构建基于WOFOST模型和遥感LAI数据同化的区域尺度冬小麦单产预测模型,并定量预测区域冬小麦单产水平。结果表明,增强型植被指数（EVI）是遥感反演LAI的最佳植被指数(开花期建模r=0.913,RMSE=0.410,灌浆期建模r=0.798,RMSE=0.470),预测能力最强(开花期r=0.858,RMSE=0.531,灌浆期r=0.861,RMSE=0.428);筛选出6个待优化参数,即TSUM1、SLATB1、SLATB2、SPAN、EFFTB3和TMPF4;产量预测精度r=0.914,RMSE=253.67 kg·hm_-2,找到了待优化参数的最佳取值,最终完成了单产模拟。     

Assessing crop management options with crop simulation models based on generated weather data

The efficient use of crop simulation models is an effective complement to experimental research. Long-term weather data obtained at a specific site are normally required for the application of these crop simulation models to help determine alternate management practices and associated decisions. Stochastic weather generators sometimes are used to complement or substitute historical weather data. The objective of this study was to evaluate the suitability of weather data generated by the weather generators WGEN and SIMMETEO as input for crop simulation models in order to determine the best option(s) among a number of different crop management practices. Five locations across Iran representing different climates were selected. The wheat, maize, and soybean models of the Decision Support System for Agrotechnology Transfer (DSSAT) were applied in this study, using 30 years of observed weather data and 90 years of weather data generated by WGEN and SIMMETEO. Simulated grain yield using either observed weather data or weather data generated by WGEN and SIMMETEO in response to various ‘experimental’ factors, e.g., cultivar selection, planting date, planting density, irrigation threshold, and change in precipitation under irrigated and rainfed conditions were compared. The statistical evaluation was based on t, F, and Kolomogrov-Smirnov (K-S) tests. The average of the percentage rejected tests was 20% and the parameter estimation method had no impact on the number of rejected tests. Irrespective of some significant differences between simulated yield based on observed weather data and those based on weather data generated by WGEN and SIMMETEO, a similar conclusion could be drawn about the best cultivar, planting date, plant density and irrigation threshold and response to changes in the amount of precipitation. Based on the results of this study it can be concluded that for many crop model applications where only relative estimates or determination of the best management option(s) rather than absolute values are required, weather data generated by either WGEN and SIMMETEO are accurate and sufficient.     

美国玉米生产技术特点与启示

从玉米品种熟期选择、农事操作及玉米生育进程、土壤耕作类型与整地、灌溉与施肥、绿色覆盖作物及精准农业技术等方面介绍美国玉米生产的技术特点,并比较分析中美玉米生产的气候条件、生育进程差异,提出高产与高效协同发展、简化农事操作、推进子粒收获等建议。