基于SVR的长江经济带水环境承载力评价

基于DPSIRM模型构建区域水环境承载力评价指标体系,并基于SVR模型构建了区域水环境承载力评价模型,利用交叉验证法对SVR模型参数进行优化选择,进一步提高模型预测精度.运用该模型研究了长江经济带2009~2018年的水环境承载力演变趋势及空间差异,结果表明：长江经济带水环境承载力等级整体呈现升高趋势,其中上游城市群承载等级由II级（超载）提升到了IV级（弱可承载）;中游城市群承载等级由II级（超载）提升到了IV级（弱可承载）;下游长三角区域承载等级由I级（重超载）提升到了IV级（弱可承载）.将评价结果与熵权-TOPSIS法的评价结果相比较,相同率达到91.7%,说明SVR模型评价区域水环境承载力可行,评价结果可靠.以下游区域为例,分别对其6个子系统的承载力进行剖析,并运用单因素轮换OAT法对各子系统内的评价指标进行敏感性分析,便于决策者识别指标敏感性.     

基于维纳过程的MEMS陀螺仪贮存寿命评估

目的 针对MEMS陀螺仪在步进应力加速试验条件下获取的性能退化数据,提出基于维纳过程的贮存寿命评估方法及其模型准确度检验方法。方法 首先,确定温度为影响MEMS陀螺仪性能退化的主要环境因素,采用步进温度应力加速试验的方式获取其性能退化数据。其次,分析各项性能参数的演变规律,确定标度因数为表征产品性能退化的特征性能参数。最后,采用漂移维纳过程对标度因数退化轨迹进行建模,并外推得到常温条件下的贮存寿命。结果 采用留一法对模型精度进行验证,模型准确度最低为86.44%。可靠度水平为0.95时,常温贮存(25 ℃)条件下的寿命评估结果为50.02 a。结论 基于维纳过程建立的性能退化模型的准确度在85%以上,该模型可应用于指定贮存条件下MEMS陀螺仪的性能退化预测及贮存寿命评估。     

400 YEARS OF CENTRAL CALIFORNIA PRECIPITATION VARIABILITY RECONSTRUCTED FROM TREE-RINGS1

ABSTRACT: Coastal central California is a region that has never been the subject of tree-ring studies. New tree-ring chronologies developed from cores of big cone spruce (Pseudotusuga macrocarpa (Torr.) Mayr.) growing in the Transverse Ranges of central Santa Barbara county were used to reconstruct precipitation fluctuations for this region. To verify the new reconstructions, calibration with recorded rainfall using cross-validation, comparison with other reconstructions, and documentary evidence from historical sources were utilized. The precipitation reconstructions show that there have not been fluctuations in mean precipitation on time scales longer than 30 years, but there have been major fluctuations in precipitation variability including changes in the frequency of extremes and rare events that have not occurred in the modern record.     

遗传算法对QSAR研究中变量选择的应用

将遗传算法引入定量结构活性关系（ＱＳＡＲ）研究中，对变量进行造反选择，可同时建立几种比较好的ＱＳＡＲ模型，并以交互验证的决定系数作为适应函数，以保证模型质量的可靠性。将其分别应用于氯代酚和单取代苯系列化合物，均得到较好的结果。同时，在这些应用中也反映了遗传算法在变量选择中存在的局限性，限不能保证选择的所有变量在模型中都有显著贡献。     

A COMPARISON OF GEOSTATISTICAL METHODOLOGIES USED TO ESTIMATE SNOW WATER EQUIVALENT1

ABSTRACT: The need to monitor and forecast water resources accurately, particularly in the western United States, is becoming increasingly critical as the demand for water continues to escalate. Consequently, the National Weather Service (NWS) has developed a geostatistical model that is used to obtain areal estimates of snow water equivalent (the thtal water content in all phases of the snowpack), a major source of water in the West. The areal snow water equivalent estimates are used to update the hydrologic simulation models maintained by the NWS and designed to produce extended streamflow forecasts for river systems throughout the United States. An alternative geostatistical technique has been proposed to estimate snow water equivalent. In this research, we describe the two methodologies and compare the accuracy of the estimates produced by each technique. We illustrate their application and compare their estimation accuracy using snow data collected in the North Fork Clearwater River basin in Idaho.     

Bayesian estimation of European sulphur emissions using monitoring data and an acid deposition model

A statistical method for estimating national emissions of acidifying air pollutants in Europe is presented. The method uses an acid deposition model to match official emissions data from European countries and measured depositions from a monitoring network. An application to 1990 sulphate data demonstrates the method and suggests some tendencies in the reported emissions. The proposed framework may prove useful for verifying national compliance with emissions standards, and the method should be applicable also to other substances than sulphur dioxide. The problem of designing an optimal monitoring network may also be assessed within the proposed statistical framework.     

长江流域0.1°网格逐日降水数据集及其精度

利用气象部门在长江流域及其附近地区的国家基本站和15省市的一般观测站1971~2005年期间的逐日降水资料,根据降水资料特点,提出使用Barnes插值和最近台站降水频率相结合的混合插值方案,得到长江流域分辨率为0.1经纬度的网格化降水逐日资料集,并通过交叉检验确定了其误差估计。结果表明,该混合插值方案对降水的模拟不但绝对误差和均方根误差较小,相关系数大,而且对降水的方差和频率模拟与观测接近,为降水资料在气象和其它领域的应用提供了逼真、连续和格点化的数据。     

Modeling Habitat Suitability for Complex Species Distributions by Environmental-Distance Geometric Mean

This paper presents a new habitat suitability modeling method whose main properties are as follows: (1) It is based on the density of observation points in the environmental space, which enables it to fit complex distributions (e.g. nongaussian, bimodal, asymmetrical, etc.). (2) This density is modeled by computing the geometric mean to all observation points, which we show to be a good trade-off between goodness of fit and prediction power. (3) It does not need any absence information, which is generally difficult to collect and of dubious reliability. (4) The environmental space is represented either by an expert-selection of standardized variables or the axes of a factor analysis [in this paper we used the Ecological Niche Factor Analysis (ENFA)].We first explain the details of the geometric mean algorithm and then we apply it to the bearded vulture (Gypaetus barbatus) habitat in the Swiss Alps. The results are compared to those obtained by the median algorithm and tested by jack-knife cross-validation. We also discuss other related algorithms (BIOCLIM, HABITAT, and DOMAIN). All these analyses were implemented into and performed with the ecology-oriented GIS software BIOMAPPER 2.0.The results show the geometric mean to perform better than the median algorithm, as it produces a tighter fit to the bimodal distribution of the bearded vulture in the environmental space. However, the median algorithm being quicker, it could be preferred when modeling more usual distribution.