Neural network classification of tan spot and stagonospora blotch infection periods in a wheat field environment

ABSTRACT Tan spot and Stagonospora blotch of hard red spring wheat served as a model system for evaluating disease forecasts by artificial neural networks. Pathogen infection periods on susceptible wheat plants were measured in the field from 1993 to 1998, and incidence data were merged with 24-h summaries of accumulated growing degree days, temperature, relative humidity, precipitation, and leaf wetness duration. The resulting data set of 202 discrete periods was randomly assigned to 10 modeldevelopment or -validation (n = 50) data sets. Backpropagation neural networks, general regression neural networks, logistic regression, and parametric and nonparametric methods of discriminant analysis were chosen for comparison. Mean validation classification of tan spot incidence was between 71% for logistic regression and 76% for backpropagation models. No significant difference was found between methods of modeling tan spot infection periods. Mean validation prediction accuracy of Stagonospora blotch incidence was 86 and 81% for backpropagation and logistic regression, respectively. Prediction accuracies of other modeling methods were 相似文献   

Neural networks that distinguish infection periods of wheat tan spot in an outdoor environment

ABSTRACT Tan spot of wheat, caused by Pyrenophora tritici-repentis, provided a model system for testing disease forecasts based on an artificial neural network. Infection periods for P. tritici-repentis on susceptible wheat cultivars were identified from a bioassay system that correlated tan spot incidence with crop growth stage and 24-h summaries of environmental data, including temperature, relative humidity, wind speed, wind direction, solar radiation, precipitation, and flat-plate resistance-type wetness sensors. The resulting data set consisted of 97 discrete periods, of which 32 were reserved for validation analysis. Neural networks with zero to nine processing elements were evaluated 20 times each to identify the model that most accurately predicted an infection event. The 200 models averaged 74 to 77% accuracy, depending on the number of processing elements and random initialization of coefficients. The most accurate model had five processing elements and correctly predicted 87% of the infection periods in the validation set. In comparison, stepwise logistic regression correctly predicted 69% of the validation cases, and multivariate discriminant analysis distinguished 50% of the validation cases. When wetness-sensor inputs were withheld from the models, both the neural network and logistic regression models declined 6% in prediction accuracy. Thus, neural networks were more accurate than statistical procedures, both with and without wetness-sensor inputs. These results demonstrate the applicability of neural networks to plant disease forecasting.     

田间飞翔蚜虫传染大豆花叶病毒(SMV)的模型预测

 本文提出一个飞翔蚜虫非持久性传染植物病毒的侵染概率模型。设置绿色诱盘捕捉着落大豆株冠的蚜虫估计飞翔翅蚜日平均着落株冠次数。采用诱饵植株测定田间一定传染性病株率和飞翔蚜量条件SMV的日侵染率,对迁飞着落蚜虫群体传染SMV的效率进行估计,实现模型对SMV日侵染率的预测。结合显症率预测,对两年三个区的SMV病株率发展动态用模型拟合回测,平均准确度为90.18%,对两年非建模的两区病株率发展动态拟合预测,平均准确度为88.41%。     

基于水分驱动的AquaCrop模型及其研究进展

介绍了AquaCrop模型的原理及基本参数,从模型的校验与应用两方面阐述了该模型的研究进展。指出目前仍缺乏实测数据验证AquaCrop模型对蒸发及蒸腾的模拟效果;AquaCrop模型在严重水分及盐分胁迫下模拟结果精度较差;已开展的模拟研究地域范围窄;由于缺少更复杂的生理子模块,AquaCrop模型不能很好解释水分胁迫对光合产物向籽粒运输分配过程的影响。为了提高模型的模拟精度并进一步延伸模型的应用范围,应完善模型水分及盐胁迫模块,并在较广范围内获取丰富的实测数据对模型开展进一步的校验研究。     

Influence of crop management on take-all development and disease cycles on winter wheat

ABSTRACT Wheat was assessed at four crop growth stages for take-all (Gaeumannomyces graminis var. tritici) in a series of field trials that studied the effects of five wheat management practices: sowing date, plant density, nitrogen fertilizer dose and form, and removal/burial of cereal straw. An equation expressing disease level as a function of degree days was fitted to the observed disease levels. This equation was based on take-all epidemiology and depended on two parameters reflecting the importance of the primary and secondary infection cycles, respectively. Early sowing always increased disease frequency via primary infection cycle; its influence on the secondary cycle was variable. Primary infection and earliness of disease onset were increased by high density; however, at mid-season take-all was positively correlated to the root number per plant, which was itself negatively correlated to plant density. At late stages of development, neither plant density nor root number per plant had any influence on disease. A high nitrogen dose increased both take-all on seminal roots and severity of primary infection cycle but decreased take-all on nodal roots and secondary infection cycle. Ammonium (versus ammonium nitrate) fertilizer always decreased disease levels and infection cycles, whereas straw treatment (burial versus removal of straw from the previous cereal crop) had no influence.     

Host frequency and density effects on powdery mildew and yield in mixtures of barley cultivars

The effects of frequency and density of susceptible plants on barley powdery mildew epidemics were studied in a combined set of addition and replacement series of field trials. In the addition series, plant densities in pure stands of three cultivars, Rambo, Rodos and Grosso (susceptible, moderately resistant and immune, respectively) were varied six-fold. In the replacement series, the three possible two-way mixtures were analysed at different frequencies but at a density corresponding to the maximum pure stand density. Disease and yield were assessed on a per-plant basis. In the pure stands, tillering reduced the range of densities from six-fold to between three- and four-fold, while in the mixtures, frequencies changed only slightly over time, indicating that competitive interactions among the cultivars were roughly equal. Yield per plant decreased logarithmically with increasing density as expected. However, yield per seed head was not correlated with the final number of heads per plot, indicating low competition among heads even at the highest density. Disease in susceptible pure stands increased strongly with decreasing density in 1994 and to a lesser degree in 1995. These differences could have been caused by differences in plant nutritional status and consequent epidemiological effects. Disease reduction on the susceptible cultivars in mixtures varied between 33% and 71% among years. Depending on the length and strength of the epidemic, the effects of host density and frequency on disease severity varied substantially among years.     

基于Sentinel-2影像的河套灌区作物种植结构提取

作物识别是提取作物种植结构的基础,利用遥感技术对作物进行监测识别,对优化生产布局、调整农业生产模式有着重要意义。文中选取河套灌区杭锦后旗为研究区域,基于2019年覆盖生长周期的Sentinel-2号卫星影像数据,构建NDVI时间序列数据集,利用Savitzky-Golay(S-G)滤波对NDVI时间序列数据集进行平滑,分析不同作物不同发育期的光谱曲线特征,计算各主要作物识别关键期的光谱阈值,构建基于决策树分层分类的农作物种植面积提取模型,并用验证样本对分类结果进行精度验证。结果表明:利用整个生育期内的NDVI最大合成影像确定植被地表覆盖,NDVI曲线变化区别林地与耕地,逐层提取地物,简便易行;采用S-G滤波重构高质量的NDVI时间序列曲线,研究证明重构后曲线更加平滑符合作物生长趋势;基于Sentinel-2号遥感数据和整个生育期NDVI时序数据,构建分层分类决策树模型,作物分类总体精度达92.1%,Kapppa系数精度达0.857。本研究采用的方法满足遥感观测应用化需求,也为县级区域农作物分类提供重要参考价值。     

Spatial pattern of Cercospora leaf spot of sugar beet in fields in long- and recently-established areas

Spatial disease pattern of Cercospora beticola was characterised during natural epidemics of Cercospora leaf spot (CLS) in sugar beet. We applied linear regression and geostatistical analyses to characterise CLS spatial patterns in three field trials, in long-established and recently-established CLS-areas, during two consecutive years. Linear regression showed a positive influence of average disease severity of within-row neighbouring plants (0.38 < < 0.88). Semi-variograms modelled the spatial dependence of disease severity for two directions per week in both years. Disease severity displayed strong spatial dependence over time. The within-row spatial dependence was the largest, but across-row dependence was irregular and weaker. Both long- and recently established areas showed strong spatial dependence of disease severity within row, decrease in variability between years and within the second trial year and a relation between and the relative nugget. Observed differences were more field than area specific. These spatial and temporal analyses indicated that disease severities of adjacent plants were dependent; hence, we concluded that C. beticola is dispersed mainly over short distances from plant to plant.     

鞘腐病发生程度与玉米倒伏及产量损失间的相关性分析

为深入探讨鞘腐病的发生对玉米倒伏及产量的影响,通过温室接种法测定了玉米不同发育阶段鞘腐病的发病程度及相关防御酶活性,以确定玉米鞘腐病的易感时期;并通过田间接种不同浓度的层出镰孢菌获得不同发病级别的玉米鞘腐病病株,于乳熟期调查病害发生程度,利用YYD-1B数显植物茎秆强度检测仪测定每株玉米茎秆的抗倒伏能力,收获后测定其产量。结果显示,玉米鞘腐病的易感时期为开花期,郑单958和浚单20在此时期鞘腐病的发病率分别为64.36%和40.22%;病情指数分别达42.73和19.58,均高于其它时期;玉米自交系OH43Ht1、郑58和杂交品种郑单958的茎秆抗倒伏能力均随着玉米鞘腐病发病级别的升高而降低;郑58和郑单958的产量随玉米鞘腐病发病级别的升高而降低,每公顷产量损失郑58从13.84%增加到29.53%、郑单958从3.99%增加到16.72%。表明玉米鞘腐病严重发生时能够降低玉米的抗倒伏能力和产量,且对自交系的影响大于杂交品种,生产中应引起高度重视。     

Is the parasitization rate of<Emphasis Type="Italic">Diaeretiella rapae</Emphasis> influenced when<Emphasis Type="Italic">Brevicoryne brassicae</Emphasis> feeds on<Emphasis Type="Italic">brassica</Emphasis> plants?

The development time and parasitization rate ofDiaeretiella rapae (M’Intosh) onBrevicoryne brassicae (L.) feeding on differentBrassica cultivars was studied in the laboratory at 20°C. The shortest development time from egg to adult parasitoid was 11.6 days on cabbage cv. ‘Yalova 1’ and the longest was 12.1 days on turnip cv. ‘Antep’ and rapeseed cv. local variety. Females lived significantly longer than males on the host plants used in the study. Females and males had the shortest longevity on rapeseed at 11.1 and 5.1 days, respectively. The highest percent parasitism ofB. brassicae byD. rapae was found on cabbage (40.20%), and the lowest was recorded on turnip (32.64%). Our results demonstrate that parasitism rate could be influenced by the plant quality, probably due to the nutritional status of the aphids or to toxic compounds ingested through the plant. Cabbage, cauliflower and broccoli were found to be suitable plants for the parasitoid, considering the development time of pre-adults, and the parasitization rate ofD. rapae onB. brassicae. http://www.phytoparasitica.org posting Jan. 23, 2007.