农业生态系统智能决策支持系统初探

建立了面向行政首长和农业主管部门的地、县级农业生态系统智能决策支持系统（ＩＤＳＳＡ）。系统实现“四库一体化”结构,提出全新的方法库概念和适合于大于知识库的产生式规则法、框架法、关系数据库法结合地知识表达体系。采取回旋阶梯式和缩进式结合的方法,通过推理、人机交互,较好地解决了农业决策的半结构化和非结构化问题。系统驱动为知识一数据的共轭驱动和用户驱动结合的模式。     

晚稻细菌性条斑病流行趋势预测模型的研究

本文在分析万安县１９８９～１９９４年晚稻细菌性条斑病发生的历史资料和相关气象资料的基础上,应用灰色关联度选择预报因子,根据模糊综合评判原理组建模型,预报晚稻细菌性条斑病的流行趋势,经回测符合率达１００％,并在１９９５年、１９９６年和１９９７年应用该模型指导万安县晚稻细菌性条斑病的防治,取得了显著的经济效益。     

基于双年份三季相光谱混合分解的耕地变化遥感检测方法初步研究——以甘肃省民勤县为例

为切实保护耕地,基于遥感技术对不同耕地演化类型,包括不变耕地、新增耕地、撂荒耕地和退耕耕地进行识别和检测,以民勤县为例,采用2010和2015年2个年份的春、夏、初冬地物景观三季相特点的Landsat TM/OLI数据作为主要数据源。依据不同耕地演化类型的季相变异规律和内部要素特征选取具有物理意义的端元丰度值、丰度值差值建立2年份混合决策树划分不同耕地演化类型。同时对比分析在2010和2015年各自决策树土地利用/覆被分类结果后比较提取的耕地演化类型。结果表明:民勤县大部分耕地未发生变化,新增耕地在各个区域均有一定量的分布且主要分布在下游;退耕耕地大面积分布在坝区;撂荒耕地主要分布在沙漠绿洲过渡带。研究表明通过能够与地物先验知识关联的端元丰度值、特征端元丰度的差值建立决策树可以有效实现不同耕地演化类型的检测。误差矩阵精度评价结果显示,基于混合决策树耕地演化类型提取结果在总体精度上要高于分类后变化检测的结果。     

新疆棉花高产高效施肥信息系统的建立与应用

在新的土壤养分状况调查结果基础上 ,应用棉花高产平衡施肥技术和计算机地理信息系统 (GIS)技术 ,建立了新疆县级农田棉花高产高效施肥信息系统 ,系统最小可实现条田级的平衡施肥决策 ,实现县、乡、村级数据的积累。 1999年通过 3种模式大面积应用于棉花生产高效施肥中 ,并取得良好效果。     

基于知识模型的棉花管理决策支持系统

通过对棉花栽培研究资料和专家知识的总结归纳,运用系统学方法和结构化途径,建立了基于知识模型的棉花管理决策支持系统(KMDSSCM)。它由数据库、知识模型库、专家咨询库和人机界面等部分组成;可以实现栽培方案设计、适宜生育指标确定、适时苗情调控、专家知识咨询和系统维护等功能,具有普适性强,独立性好等特点。实例分析结果表明,系统可以根据决策点的基础农情条件,设计出适宜的综合性栽培管理方案和生育指标动态。     

Properties of Edible Films Made From Anchovy By-Product Proteins and Determination of Optimum Protein and Glycerol Concentration by the TOPSIS Method

In the present study, the physicochemical and mechanical properties of films made from anchovy (Engraulis encrasicholus) by-product proteins (ABPs) were characterized as a function of protein (3, 4, and 5%) and glycerol (30, 40, 50, and 60%) concentration. In particular, the thickness, color, transparency, water vapor permeability (WVP), solubility, oxygen permeability (OP), and mechanical properties of the films were measured, and generally, these properties were affected by both protein and glycerol concentration. The thickness, redness, WVP and solubility values of films increased, while their lightness and yellowness values decreased as protein concentration increased. Furthermore, lightness, yellowness, and elongation at break (EAB) values were found to increase at higher glycerol concentrations within ABP films, whereas tensile strength (TS) and elastic modulus (EM) values decreased. ABP films became more transparent with increase of glycerol concentration and decrease of protein concentration. The OP values of all films decreased depending on protein and glycerol concentration. The techniques for order performance by similarity to ideal solution (TOPSIS) method was used to determine optimum protein and glycerol concentrations. According to this method, an edible film containing 4% protein and 40% glycerol was the best film with a C value of 0.6665.     

An Expert System for the Diagnose of peripheral Nerovus Conduction

To promote automation of diagnoses for the pherpheral nervous conduction,byartificial in telligence technique,an expert system of diagnose is formed. This expert system is madefrom the experiences of the soecialists in the field,which constitute the rcasoning rules and k nowl-edge-based of the system,and the database managemnt system.The tests show that the efficiencyand correctness are obviously improved.     

基于贝叶斯方法的东、黄海鲐资源评估及管理策略风险分析

运用基于贝叶斯的剩余产量模型,对东、黄海鲐资源进行评估,确定了当前鲐资源开发利用状态,估算了在不同收获率水平下未来5年鲐资源量和年总可捕捞量,分析了管理策略实施后鲐资源崩溃的风险。结果表明,2006年东、黄海鲐正遭受过度捕捞,但其资源量并未处于过度捕捞状态。决策分析表明,收获率为0.3是最适预防性的管理策略,在该策略下,鲐平均资源量将从2006年的451千吨将增加到2011年的871千吨,2011年资源量恢复到BMSY的概率为0.48,而过度捕捞的概率为0。     

Changing how we approach fisheries: A first attempt at an operational framework for ecosystem approaches to fisheries management

The increasing need to account for the many factors that influence fish population dynamics, particularly those external to the population, has led to repeated calls for an ecosystem approach to fisheries management (EAFM). Yet systematically and clearly addressing these factors, and hence implementing EAFM, has suffered from a lack of clear operational guidance. Here, we propose 13 main factors (shift in location, migration route or timing, overfishing (three types), decrease in physiology, increase in predation, increase in competition, decrease in prey availability, increase in disease or parasites and a decline in habitat quality or habitat quantity) that can negatively influence fish populations via mechanisms readily observable in ~20 population features. Using these features as part of a diagnostic framework, we develop flow charts that link probable mechanism(s) underlying population change to the most judicious management actions. We then apply the framework for example case studies that have well‐known and documented population dynamics. To our knowledge, this is the first attempt to provide a clearly defined matrix of all the probable responses to the most common factors influencing fish populations, and to examine possible diagnostics simultaneously, comparatively and relatively in an attempt to elucidate the most probable mechanisms responsible. The framework we propose aims to operationalize EAFM, thereby not only better diagnosing factors influencing fish populations, but also suggesting the most appropriate management interventions, and ultimately leading to improved fisheries. We assert the framework proposed should result in both better use of limited analytical and observational resources and more tailored and effective management actions.     

基于GF-1卫星遥感的冬小麦面积早期识别

GF-1号卫星是中国高分卫星系列首颗卫星,自2013年04月26日发射以来,提供了大量的2 m/8 m/16 m空间分辨率的卫星数据,成为中国农业遥感监测的主要数据源之一。该文以GF-1卫星携带的16 m空间分辨率的宽视场(wide field view,WFV)传感器为主要数据源,采用2013年10月2日、10月17日、11月7日和12月5日4个时相的数据,以多尺度分割后的对象为基本分类单元,采用分层决策树分类的方法对冬小麦面积进行提取,并利用地面样方数据对分类结果进行了精度验证。结果表明,北京市顺义区冬小麦面积7 095 hm2,分类总体精度达到96.7%,制图精度为90.0%,其他未分类类别精度为97.3%,Kappa系数为0.8。研究区内冬小麦的播种时间可以分为10月1-5日早播、10月6-10日中播、10月11-15日中晚播、10月16-20日晚播等4个时间段,不同播期对应着归一化植被指数(normalized difference vegetation index,NDVI)不同的变化规律,是分层的基础,结合波段反射率、波段反射率和、波段反射率比值等参数的变化规律,通过分层可以有效的剔除草坪、桃树等容易同冬小麦混淆的地物类型,GF-1/WFV提供的多时相遥感数据能够可靠的反映冬小麦发育变化的规律,是冬小麦面积准确提取的基础,在农作物面积遥感监测业务运行中具有较大的开发应用潜力。