面向粮食产量空间化的多元回归分析模型中的两个关键问题探讨

为比较不同区域尺度变量对模型拟合效果的影响,在全国不分区和分区两种情况下,分别基于县级和地市级两个区域尺度上的样本构建粮食产量与水田、水浇地、旱地面积之间的多元线性回归模型,结果显示,用地市级数据作分析样本比用县级数据作分析样本好,分区建模比不分区建模效果好。因此,将全国划分为7个区,以地市级数据作为区域尺度的变量,在区域、栅格、亚栅格三个尺度上探讨变量（因变量、自变量）尺度和常数项取值这两个因素对模型应用的影响。得到以下结论：1）基于区域尺度样本构建的多元回归模型,如果常数项不为0,则不能用于空间化计算;如果常数项为0,则可以用于空间化计算;2）基于栅格尺度样本构建的多元回归模型,不论常数项是否为0,均可应用于空间化计算;3）基于亚栅格尺度样本构建的多元回归模型,不论其常数项是否为0,也均可用于空间化计算,但需要将计算结果乘以一个系数,该系数等于栅格单元面积与亚栅格单元面积的比值。上述结论对其他类型的统计数据空间化具有指导和参考价值。     

线性工程切割耕地的权属调整方法及系统实现——以保阜高速顺平段沿线土地整治项目为例

论文以高速公路切割的耕地地块为研究对象,根据权属调整的不同方向,将权属调整分为所有权置换和使用权流转两种类型,并将权属调整的方法分为面积法和价值法;在GIS服务软件SuperMap iServer支持下,建立了一个交互式权属调整应用系统,以保阜高速顺平段沿线的高于铺镇、腰山镇土地整治项目区中的切割耕地为案例,应用系统生成了项目区高速公路沿线切割耕地的权属调整方案。研究结果表明,线性工程切割耕地需进行权属调整以保障农业生产,研究采用的权属调整方法符合当地实际情况,利用权属调整系统可以自动生成推荐权属调整方案,并可以根据用户需求进行调整,同时系统可以对权属调整相关数据进行动态管理。     

基于生态服务价值的多目标水源地土地利用结构优化

为探讨首都水源地最佳水土资源优化配置模式,选取北京山区怀柔水库流域为研究对象,基于研究区5期(1990、1995、2000、2005 和2008)卫星遥感影像数据,分析了怀柔水库流域1990~2008年间土地利用时空变化特征,并基于生态服务价值理论、灰参数线性规划理论和CLUE-S模型对怀柔水库流域土地利用进行空间优化配置模式研究.结果表明:现有土地利用结构在优化配置后流域生态系统服务功能价值较2008年增加1795.4万元,增幅5.4%.其中,流域涵养水源功能价值量增加最多,为560.9万元,净化大气的价值增长幅度最大,增长率达到9.3%.研究结果可为该区域水土地资源的可持续利用、开发和管理提供科学依据.     

FDI对中国环境污染影响的实证研究

基于中国期刊全文数据库中的30篇中文社会科学引文索引(CSSCI)的期刊论文,运用Meta方法定量分析FDI(外商直接投资)对中国环境污染的影响. 结果表明:①单一或几个环境污染指标和FDI相对值的回归系数分别为-2.72和-1.83,说明采用二者会降低FDI环境效应显著性,但后者不显著;②面板数据和样本时间跨度的回归系数分别为3.19和1.52,说明采用二者会提高FDI环境效应显著性,并更有可能得到FDI加剧了环境污染的结论;③样本数据平均年限变量的回归系数为-0.90,说明采用2000年后数据会降低FDI环境效应显著性,2000年前FDI对环境污染加剧效应呈下降趋势,此后FDI改善了环境质量;④OLS法和固定效应模型的回归系数分别为2.22和0.78,说明采用二者会提高FDI环境效应显著性,但后者不显著. 综上可知,在纳入部分控制变量前提下,采用环境污染综合指数和FDI水平值,基于1985—2010年面板数据,运用OLS法和固定效应模型实证发现,FDI更有可能显著加剧了环境污染,但这种效应呈下降趋势,以至于2000年后FDI改善了环境质量.      

燃煤型砷中毒病区尿砷甲基化代谢和砷致皮肤损伤危险度的关系初探

砷中毒具有特异的皮肤损伤特征。为了研究燃煤型砷中毒病区高砷暴露、人体甲基化代谢能力与皮肤损伤患病风险之间的关系,在陕南典型燃煤型砷中毒村进行了皮肤损伤诊断和流行病学调查,采集尿样并分析总砷及形态砷含量,同时计算了用于表征人体砷甲基化代谢能力的指标包括尿中无机砷、一甲基砷和二甲基砷占总砷的百分含量(i As%、MMA%、DMA%),以及一甲基化率(PMI=MMA/i As)和二甲基化率(SMI=DMA/MMA)。Logistic回归分析结果表明:尿总砷含量(UTAs)是砷致皮肤损伤的危险因素(OR=1.038,95%CI:1.003~1.073),二甲基砷百分含量和SMI是皮肤损伤的保护因素(OR=0.883,95%CI:0.798~0.976;OR=0.724,95%CI:0.535~0.978);且砷致皮肤损伤的危险度随砷暴露水平的增高和甲基化能力的降低而增大。     

Fuzzy optimization of multi-period carbon capture and storage systems with parametric uncertainties

Carbon capture and storage (CCS) is an important technology option for reducing industrial greenhouse gas emissions. In practice, CO₂ sources are easy to characterize, while the estimation of relevant properties of storage sites, such as capacity and injection rate limit (i.e., injectivity), is subject to considerable uncertainty. Such uncertainties need to be accounted for in planning CCS deployment on a large scale for effective use of available storage sites. In particular, the uncertainty introduces technical risks that may result from overestimating the limits of given storage sites. In this work, a fuzzy mixed integer linear program (FMILP) is developed for multi-period CCS systems, accounting for the technical risk arising from uncertainties in estimates of sink parameters, while still attaining satisfactory CO₂ emissions reduction. In the model, sources are assumed to have precisely known CO₂ flow rates and operating lives, while geological sinks are characterized with imprecise fuzzy capacity and injectivity data. Three case studies are then presented to illustrate the model. Results of these examples illustrate the tradeoff inherent in planning CCS systems under parametric uncertainty.     

基于回归支持向量机的指标规范值的水质评价模型

传统的回归支持向量机的水质评价模型不具有普适性和通用性,当指标较多时,模型的学习效率和求解精度均会受到影响.若适当设定3类水体(地表水、地下水和富营养化水体)各项指标的参照值及指标值的规范变换式,使不同指标的同级标准的规范值差异不大,从而可以认为用规范值表示的不同指标皆“等效”于某个规范指标.因此,可建立用规范值表示的任意m项指标组合皆适用的水质评价的回归支持向量机模型.通过实例对模型的实用性进行了效果检验,结果表明:用基于回归支持向量机的指标规范值的水质评价模型对河桥地表水、黑龙洞泉域地下水和山仔水库富营养化水体的水质评价结果与用BP神经网络评价法、模糊综合评价法和属性识别评价法的评价结果基本一致.     

Preliminary shotgun glycoproteome analysis of N-glycosylation sites in serum proteins of ricin-intoxicated rats

In this study, the variations of serum glycoproteins after exposure to ricin toxin (RT) in Wistar albino rats are reported. For glycopeptide enrichment, Microcon YM-3 centrifugal filter device capture and micro-liquid chromatography (LC)–electrospray ionization (ESI)–tandem mass spectrometric (MS/MS) analysis were used; 74 trypsin-digested proteins were identified in the control group, and 58 in the RT-intoxication group. Additionally, 33 N-glycosides and 14 glycoproteins were identified in the control, and 50 and 21 in the RT-intoxication group.     

Detection and quantitation of trace levels of low molecular weight halocarbons (LMHs) associated with commercial sodium humate

Examination of vapor above commercial sodium humate and above 10wt% aqueous humate solution by gas chromatography (GC) revealed the presence of chloroform and trichloroethylene. Further examination of the aqueous humate by the purge and trap procedure confirmed the association of chloroform and trichloroethylene with the humate. GC/mass spectrometry analyses of methylene chloride extract of the humate established the association of LMHs with the humate. Improved headspace capillary GC technique showed that the humate contained not only chloroform, trichloroethylene but several other LMHs in the range 190 to 4000ng·kg^‐1.     

Regression models for octanol‐water partition coefficients,and for bioconcentration in fish

Octanol‐water partition coefficients (P) of a number of organochlorine insecticides (OCs) are presented. The merits of log‐log regressions between experimental ? values and calculated estimates of P, solute activity coefficients in water, solute molecular surface area data, and reversed‐phase liquid‐chromatographic net retention data, are critically evaluated for several classes of pollutants: polycyclic aromatic hydrocarbons (PAHs), chlorinated benzenes, chlorinated biphenyls and OCs. Special attention is paid to the predictive accuracy of such semi‐empirical regressions in connection with possible effects of solute molecular shape and polarity.

Finally, bioconcentration and ‐accumulation of hydrophobic pollutants in fish are briefly discussed.