自组织网络与广义回归网络耦合的副热带高压指数预测

利用亚洲夏季风系统中各成员变化活动与西太平洋副高存在的不同程度的时延相关性,从1995~ 2004年NCEP/NCAR逐日再分析资料中,提取了亚洲夏季风系统各成员变化活动的特征指标及其对应的超前三候的西太平洋副高(简称副高)面积和脊线指数.在此基础之上,建立了自组织网络与径向基函数网络串级耦合的副高指数预测模型.该模型首先用自组织网络对各指标样本按其自身相似原则进行无监督分类,随后用广义回归网络分别对分类出的各指数样本子集进行有监督的训练建模和预测.模型的预测试验结果表明：副高指数的预测结果与其实际值之间的相关系数达到0.89,明显优于单一的神经网络模型预测效果.     

大厚度粗颗粒含水层井损值的确定

依据实际勘探资料,深刻分析了大厚度粗粒含水层井损值的影响因素。认为井损值主要受抽水井井壁管降深和井的最大出水量两个因素的影响和控制,三者之间高度相关,并建立了回归方程,井损值计算结果与实际观测值基本一致,应用效果较为理想。而采用现成的理论方法计算的井损值,由于其加坡系数在野外的不确定性,计算结果与实测观测值往往有着较大的差别。     

Short- and medium-term response of Atriplex halimus L. to repeated seasonal grazing in south-eastern Spain

Atriplex halimus L. is a valuable fodder shrub that has been widely used in arid Mediterranean regions. However, its response to several cumulative years of grazing management remains uncertain. Accordingly, we conducted a three-year experiment in south-eastern Spain to analyze the influence of repeated seasonal (winter-spring) sheep browsing on young A. halimus plants. Shrub growth, fodder and wood production, plant compactness and shoot moisture content were assessed through regular monitoring of the size of shrubs, sampling of summer shoots, and destructive procedures at the end of the experiment. Shrubs were on two closely located sites but where soil quality and water availability were different. At the better location, browsed shrubs grew less and were significantly smaller (P < 0.001) than the non-browsed saltbushes. In contrast, differences between groups were minimal where the growth conditions were poorer. In all cases, shrubs resprouted vigorously and maintained fodder production throughout the experiment. Browsing affected shrub structure, resulting in smaller but more compact plants (2.20 vs. 1.24 g dm⁻³, P < 0.001), but their proportion of fodder remained similar. The spring grazing increased shrub-shoot moisture content in the summer (62.5% vs. 56.0%, P < 0.001). Overall, this species responded very positively to repeated seasonal grazing.     

秦皇岛市气温与近地面臭氧浓度的关系分析

利用2013年1月至2014年12月和2017年1月至2019年6月秦皇岛市近地面臭氧（O₃）浓度数据和气象资料,采用广义相加模型（GAM）,运用回归分析方法和基于R语言的统计分析软件,控制气压、相对湿度、日照时数、总云量等要素的混杂效应及时间变化趋势,分析春季、夏季、秋季、冬季气温与O₃浓度的关系。结果表明：秦皇岛市O₃浓度夏季最高、春季次之,冬季最低,与气温变化趋势基本一致,呈现明显的季节变化。各季气温与O₃浓度呈非线性相关关系,拟合曲线存在拐点,拐点两侧相关效应存在明显差异,主要表现为春季日平均气温高于15.0℃时,气温每升高1℃,O₃浓度增加7.6 μg·m^-3,增长速率是气温低于15.0℃时的4.0倍;夏季日平均气温高于27.2℃时,气温每升高1℃,O₃浓度增加13.9 μg·m^-3,增长速率是气温低于27.2℃时的11.6倍;秋季日平均气温高于21.4℃时,气温每升高1℃,O₃浓度增加47.5 μg·m^-3,增长速率是气温低于21.4℃时的19.1倍;冬季O₃浓度偏低且变化较为平稳,气温对O₃浓度的变化影响不大。由于春夏两季O₃浓度基础值偏高,因此,夏季和春季气温偏高时O₃浓度快速增加现象应引起高度重视。     

无气象观测地区的电线覆冰厚度推算

利用湖北省1962-2008气象数据与电线积冰数据,建立了各种气象要素与电线覆冰厚度的回归模型,该回归模型充分考虑了形成电线覆冰需具备相应的气候条件.通过动力降尺度计算了灾情发生地的气象数据,结合已建立的回归模型,推算出电线覆冰厚度.与实测电线覆冰厚度相比,推算准确率在62.8%～75.9%.表明依据建立的回归方程并结...     

基于PSInSAR技术的海原断裂带地壳形变初步研究

常规差分干涉测量(DInSAR)受时间、空间失相干的严重制约和和大气延迟等相位误差的影响,难以实现对长期累积微小地壳形变场的有效探测.PSInSAR技术克服了常规DInSAR的局限性,能够高精度监测微小地壳形变.本文首先介绍了PSInSAR技术的算法模型和处理方法.该方法通过二维线性相位模型,对时序干涉图象上相干点目标...     

北京地区粘性土分类指标与力学指标关系回归分析

通过统计北京地区7 602组粘性土土工试验数据,研究了粘土分类指标与力学指标各参数之间的相关关系,并给出了相应的一元、多元回归方程式和相关系数。分类指标回归分析结果表明:含水量与孔隙比、饱和度、液限和塑限指标相关性较好;液限与比重、孔隙比、塑限、塑性指数相关性较好;液性指数与饱和度、塑限与孔隙比、孔隙比与密度、饱和度与液性指数及塑性指数与比重相关性都较好。力学指标回归分析表明:压缩模量E2与E1相关性好;E2与E1、初始应力相关性好;E2与粘聚力、摩擦角、E1、初始应力、孔隙比相关性好;E2与E1、粘聚力、摩擦角相关性好。并不建议运用相关方程式确定指标,但对于资料欠缺的地区可以借鉴。     

Landslide susceptibility assessment using SVM machine learning algorithm

This paper introduces the current machine learning approach to solving spatial modeling problems in the domain of landslide susceptibility assessment. The latter is introduced as a classification problem, having multiple (geological, morphological, environmental etc.) attributes and one referent landslide inventory map from which to devise the classification rules. Three different machine learning algorithms were compared: Support Vector Machines, Decision Trees and Logistic Regression. A specific area of the Fruška Gora Mountain (Serbia) was selected to perform the entire modeling procedure, from attribute and referent data preparation/processing, through the classifiers' implementation to the evaluation, carried out in terms of the model's performance and agreement with the referent data. The experiments showed that Support Vector Machines outperformed the other proposed methods, and hence this algorithm was selected as the model of choice to be compared with a common knowledge-driven method – the Analytical Hierarchy Process – to create a landslide susceptibility map of the relevant area. The SVM classifier outperformed the AHP approach in all evaluation metrics (κ index, area under ROC curve and false positive rate in stable ground class).     

A laboratory batch study on arsenic sorption and desorption on guava orchard soils of Baruipur,West Bengal,India

Groundwater contaminated with arsenic (As), when extensively used for irrigation, causes potentially long term detrimental effects to surface soils. Such contamination can also directly affect human health when irrigated crops, such as rice, vegetable and fruits, are used for human consumption. Therefore, an understanding of the sorption and desorption behavior of As in surface soils is of high importance, because these processes regulate the bioavailability of As in the soil environment. In this study, we have collected soils from guava orchards of Baruipur, West Bengal, and characterized soil chemistry and batch sorption and desorption behavior in the laboratory. The sorption and desorption behavior of As in the soils were examined using the Langmuir and Freundlich sorption equation. Regression analysis of the soil chemical characteristics and sorption equation parameters were also performed. The results suggest that the sorption behavior of arsenate is highly dependent on soil characteristics, specifically organic carbon, clay and Al₂O₃ content of the soils. Whereas desorption behavior is critically influenced by the presence of high concentrations of amorphous and/or crystalline Fe₂O₃ in the soils. Retention of the significant portion of As in the soils (~ 84% of the total) suggests that As in the orchard soils may not be highly bioavailable to plants for uptake. However, more detailed studies will be required to ascertain the role of individual soil components on the As sorption and desorption processes.     

水库地震震级预测的讨论

作者于１９８７年提出用水库“综合影响参数Ｅ”预测水库地震最大震级Ｍ的方法以来,为许多工程和科研人员引用或讨论．后来,除Ｅ外我们又增加了最大库深Ｈｍａｘ,并把它们都视为具有随机统计偏差的变量进行回归．为使关心的读者更多了解本法,本文讨论了回归变量的物理意义及其随机性、回归方法、几种回归结果的比较,以及实际应用时的修正问题．