元胞自动机和克隆选择算法的蓝藻水华预测模型

针对蓝藻水华频发影响水体质量和生态环境的问题,该文以太湖为例,探讨了元胞自动机和克隆选择算法在蓝藻水华预测方面的可行性。使用遥感经验算法计算水体叶绿素a浓度;将叶绿素a浓度分布数据作为预测模型的输入数据,将基于平衡方程的预报模型与元胞自动机进行耦合,对蓝藻水华时空变化过程进行了动态模拟;在预测方面实现了蓝藻水华空间分布情况的可视化,并采用克隆选择算法优化了预测模型中的重要参数。研究结果表明:利用克隆选择算法优化的元胞自动机模型能够在蓝藻水华预测中直观地显示蓝藻水华的时空变化特征;能够在短时间内重复性地预演蓝藻水华的变化过程,为决策者提供有用的参考。     

Household level influences on fragmentation in an African park landscape

The process of landscape fragmentation outside park borders occurs through the actions of people living near the boundaries. In the Kibale National Park landscape in western Uganda, human-landscape relationships are typified by small-scale subsistence agriculture, in which households rely on resources provided in forests and wetlands, whose use is in turn shaped by perceptions of resource availability. To understand and manage for fragmentation of resource pools, modeling and identifying the proximate drivers, and thus enacted resource extraction and utilization – is of fundamental importance. We combine landscape analysis at the household scale, using remotely sensed data, with household surveys, to understand the potential human drivers of local scale landscape change. We found strong evidence for a local household zone (LHZ) effect on fragmentation patterns with geographical and socioecological heterogeneities in LHZ impact. Differences were influenced by wealth, and in some cases, tribal identity. The perception of crop raiders – primarily baboons and small monkeys, but also elephants and other animals – may have largely shaped human-environment interactions, and were associated with fragmentation. Ninety-two percent of the best fit models included the attitude that the park should stay, but associated it with increased fragmentation, suggesting that the uncharacteristic non-hostile attitude about Kibale does not directly translate into conservation-friendly local human-environment interactions. This study provides insight into park–neighbor interactions and the influence of the LHZ on protected-area landscapes, and it points to important points in the system for collaborative opportunities to engage communities and conservation managers.     

A discrete spatial choice model of burglary target selection at the house-level

This article studies how burglars select a house to burglarize. We draw on the rational choice perspective to investigate how burglars select a target by relying on house-related attributes to optimize a combination of perceived rewards, efforts and risk. It extends current applications of the discrete spatial choice framework to burglary target selection by adopting the house as the spatial unit of analysis and studies burglars' target selection process in a larger and more diverse study area than that of earlier studies. Using data on 650 residential burglaries and on approximately 500,000 residential properties in the Belgian province East Flanders, we consider a discrete spatial choice model of burglary target selection to establish which house-related attributes influence burglars' target selection process. Our findings demonstrate that terraced houses, houses without a garage, houses that have not been outfitted with a central heating and/or air-conditioning system and houses nearby burglars' residences are more likely to be selected. Overall, our analysis suggests that burglars rely on effort-related attributes to distinguish between targets while higher perceived rewards actually decrease the odds of a house being burglarized. Risk-related attributes are unimportant for burglars' target choice.     

阶段性干旱胁迫对小麦光合生理特性及产量结构的影响

在人工遮雨的条件下,采用盆栽的种植方式探究"皖麦68"营养生长期(返青期—开花期)及生殖生长期(开花期—成熟期)轻度干旱胁迫(土壤相对含水量为55%±5%)及复水(土壤相对含水量为70%±5%)对其光合生理特性及产量结构的影响。结果表明:返青期至成熟期充分供水(CK)的小麦旗叶光合参数和产量最高。开花至成熟期复水(DN)的小麦叶片在复水后光合能力迅速恢复,表现出了超补偿效应:光合速率(16.43μmol/(m~2·s))甚至超过了CK(15.01μmol/(m~2·s));采用非直角双曲线模型拟合小麦旗叶的光响应曲线,其中DN的曲角θ最大;DN产量较CK略有降低但千粒重为34.51 g,高于CK(34.44 g)。开花至成熟期轻度干旱(ND)及全生育期轻度干旱(DD)的小麦光合特征参数与产量均显著降低。DD产量最低、品质最差,但其收获指数I_H高于CK、仅次于DN。在小麦返青期—开花期进行水分管理适量减少灌溉,开花期—成熟期复水能够提升籽粒的干物质积累量,获得较高的产量及品质。     

MIXTURE ANGLYSIS USING FACTOR ANGLYSIS I: CALIBRATION AND QUANTITATION

One of the major application areas of factor analysis, multivariate calibration and quantitation, is coveredin this review. The algorithms, methodologies and applications covered include principal componentregression, target transformation factor analysis, singular value decomposition and rank annihilationfactor analysis. Many important areas of research having relevance to multivariate calibration andquantitation problems are also covered in this review, including background correction, measurementerror, rank determination, cross-validation, figures of merit, detection of invalid samples, experimentaldesign, sample selection, statistical inference and wavelength selection.     

基于模糊综合评判的机械零件材料选择方法

针对在选择机械零件材料时采用传统方法所具有的主观性缺陷,本文结合设计实例,运用模糊数学理论,建立了零件材料选择的模糊综合评判的数学模型,能从多种可用材料中较客观、合理地确定出最适用的一种材料,从而提高了设计质量.     

SHORT COMMUNICATION MODEL SELECTION BY THE PROGRAM MINIGLASS IN COMPLEX FORMATION STUDIES

The model selection from EMF data of pH titration curves has been studied for binary and ternarysystems(Ni~(2+)-glycine and Ni~(2+)-methionine-taurine respectively)by using the MINIGLASSprogram.Comparison between experimental and calculated titration curves with this interactive programindicates the presence(and the pH range)of the different species in equilibrium.     

Multi-electrode resistivity survey for groundwater exploration in the Harare greenstone belt, Zimbabwe

A multi-electrode resistivity survey, carried out over metasedimentary strata and metavolcanics in the Harare greenstone belt in northeastern Zimbabwe as part of a groundwater resources investigation, illustrates the ability of this technique to produce high-resolution images of the subsurface, which are useful for groundwater resources assessment. The resistivity results provide a clear view of the thickness of the weathered regolith and of the distribution of the various lithological units. Using a combination of apparent formation resistivity and inferred depth of weathering, it is possible to characterize the various lithologies on the geophysical profiles. These assigned lithologies show excellent correlation with the mapped geology, and the main lithologies, metabasalt, meta-arenite, granodiorite and banded iron formation can be clearly identified. The banded iron formation is characterized by low resistivity values, while a combination of the depth of weathering and resistivity values are used to distinguish between the meta-arenite on one hand and the metabasalt and granodiorite on the other. The multi-electrode method is successful in identifying potentially favourable zones for obtaining groundwater, such as areas with a maximum depth of weathered regolith, zones of fracturing and faulting, and high porosity and permeability zones associated with lithological contacts.

---

Resumen Un estudio de resistividad multi-electrodo que se llevó a cabo en estratos meta sedimentarios y metavolcánicos en el cinturón de rocas verdes de Harare, en la parte noreste de Zimbabwe, como parte de una investigación de recursos de agua subterránea, ilustra la capacidad de esta técnica para producir imágenes de alta resolución de la subsuperficie, las cuales son útiles para la evaluación de los recursos da agua subterránea. Los resultados de resistividad suministran una visión clara el espesor del regolito meteorizado y de la distribución de varias unidades litológicas. Mediante el uso de una combinación de resistividad aparente de formación y de una profundidad inferida de meteorización, es posible caracterizar las diferentes litologías en los perfiles geofísicos. Estas litologías asignadas muestran una excelente correlación con la cartografía geológica y se pueden identificar las litologías predominantes como metabasalto, meta-arenita, granodiorita y formación de hierro bandeado. Esta última se caracteriza por valores bajos de resistividad, mientras que se utiliza una combinación de profundidad de meteorización y valores de resistividad, para distinguir entre meta-arenita por un lado y metabasalto y granodiorita por el otro. El método multi-electrodo es exitoso para identificar zonas potencialmente favorables para captar agua subterránea, las cuales pueden ser áreas con una profundidad máxima de regolito meteorizado, o bien zonas de fracturamiento y fallamiento, o también zonas de porosidad y permeabilidad altas asociadas con contactos litológicos.

---

Résumé La prospection électrique par la méthode de multiélectrodes, effectuée pour linvestigation des eaux souterraines cantonnés dans des roches métamorphiques dorigine sédimentaire et volcanique de la ceinture de Harare-Zimbabwe a mis en évidence les possibilités de cette technique de réaliser des images à grande résolution du sous-sol qui sont très utiles pour lévaluation de la ressource en eaux souterraines. Daprès la distribution de la résistance il résulte une image très claire de lépaisseur des regolithes altérés ainsi que de la distribution des différentes unités lithologiques. Daprès les profiles géophysiques on peut caractériser les différents lithologies en utilisant les résistivités apparentes de la formation respective et les informations sur la profondeur de la zone altérée. Les lithologies ainsi déterminées montrent une excellente corrélation avec les cartes géologiques et on peut très bien identifier les lithologies principales comme les metabasaltes et les meta-arenites, les granodiorites et les formations de fer rubanées. Cettes dernières sont caractérisées par des faibles valeurs de la résistivité, tandis que en utilisant les résistivité et la profondeur de la zone altérée on peut distinguer les meta-arenites de granodiorites et des metabasaltes. La méthode des multiélectrodes est donc capable didentifier des formations favorables à laccumulation en eau comme les zones de faille et des fractures, ainsi que les zones à grande porosité et perméabilité, associées aux contactes lithologiques.

     

基于CIRCLE特征变换的点群选取改进算法

本文在介绍了已有的“基于Circle特征变换的点群选取算法”(测绘科学2005年第3期)基础上,提出了一种改进算法。详细分析了已有算法在线化简过程中对聚类的过度依赖性和线化简过程中的逻辑不一致性等两个主要不足,对特征空间中的线化简等关键步骤进行了改进。最后就改进算法的结果与原先算法的结果进行了对比分析与评价。     

线要素多尺度表达的一种新方法

本文依据多比例尺条件下矢量数据的绘制原理,提出了多尺度线化简与线选取方法。多尺度线化简算法依比例尺将空间坐标转换为像素坐标,根据像素坐标删除曲线中重合点与共线点,综合后的曲线与原始曲线在各比例尺下绘制结果相同;多尺度线选取算法利用仿射变换结果快速计算线要素之间的连通关系,然后渐进删除长度小于阈值的悬线与孤立弧段,保持了线要素的连通性。对不同类型数据的实验验证了本文算法的有效性和可行性。