On the optically biaxial character and heterogeneity of anthracites

The results of the study of optical properties of 13 anthracites from different parts of the world are presented in this paper. Measurements of reflectance values were made on non-oriented vitrinite grains for a minimum of 300 points per sample. The reconstruction of Reflectance Indicating Surfaces (RIS) were made by Kilby's method [Kilby, W.E., 1988. Recognition of vitrinite with non-uniaxial negative reflectance characteristics. Int. J. Coal Geol. 9, 267–285; Kilby, W.E., 1991. Vitrinite reflectance measurement — some technique enhancements and relationships. Int. J. Coal Geol. 19, 201–218]. It was found that the use of Kilby's method for strongly anisotropic materials like anthracites did not give unambiguous results. Some improvement in Kilby's method, consisting of the division of the cumulative cross-plot into several elemental components, is suggested. Each elemental cross-plot corresponds to a textural class of anthracite, which is characterized by the values of RIS main axes R_MAX^(k), R_INT^(k) and R_MIN^(k) (k=1,2,…n; n — number of classes). The global texture of anthracite is characterized as a RIS with main axes calculated as the weighted means of , and for each class of this anthracite.The division of cumulative Kilby's cross-plot on elemental components makes possible the calculation of new coefficients H_t and H₁₀ characterizing the heterogeneity of the structure and texture of anthracites. The results of our study show that all anthracites have biaxial negative textures, but their heterogeneity varies in a wide range of H_t and H₁₀ coefficients depending upon the individual coal basin.     

网状河比较沉积学研究

根据比较沉积学原理,论述了现代网状河沉积的沉积过程、沉积韵律特征、平面形态及其环境背景,进而对古代相关沉积类型进行了比较研究,对网状河的形成机制进行探讨和总结,以指导地下同类沉积砂体的油气勘探和开发。     

基于高斯混合模型的建筑物QuickBird多光谱影像数据分类研究

针对任何一种遥感影像数据的信息提取都有其无法逾越的理论极限,正确认识这种极限有利于明确相关算法研究的方向,降低工程应用成本.制约影像信息提取精度的"同物异谱"现象以及与之相关的影像对象"光谱异质性"问题正是科学认识这种极限的关键和切入点.城市下垫面中的建筑物屋顶材料不同,光谱反射率也不同,"同物异谱"现象严重.基于高斯混合模型的期望最大(Expectation Maximization,EM)估计算法,能为分析建筑物类内以及同一建筑物对象内部光谱异质性程度提供科学依据, 进而提高分类精度.本文以QuickBird多光谱影像为实证研究数据, 利用高斯混合模型及其EM估计算法拟合出不同材料屋顶的密度分布, 实现建筑物影像对象分类, 得到优于传统监督、非监督分类的结果.     

Recognition of shallow karst water resources and cave potentials using thermal infrared image and terrain characteristics in semi-arid regions of Iran

Shallow karst water resources and caves may influence land surface temperatures due to cold transfer property of rocks and evaporation from buried karst. The objective of this research was to develop a method for recognition of karst areas based on evaluating the surface characteristics that manifest itself by low land surface temperature in the satellite images. Investigation of thermal ETM⁺ image of the study region in Iran showed that parts of carbonate rocks that bear karst water are relatively cooler compared to areas with similar terrain conditions. Relational modeling provided useful information on spatial distribution of areas that have the potential to hold karst water resources and/or caves. Further inspection of ASTER images, along with geotechnical, geophysical and geological field surveys verified the approach. Significant correlation was found between electrical resistivity and thermal band values. The method may be used as a primary exploratory tool for shallow karst water explorations in similar areas.     

Spatial and seasonal variations of fish assemblages in mangrove creek systems in Zanzibar (Tanzania)

Spatial and seasonal variations of fish assemblage composition were studied in three non-estuarine mangrove creeks of Zanzibar (Tanzania). Fish were collected monthly for one year at three sites (lower, intermediate and upper reaches) in each creek using a seine net (each haul covering 170 m²). Density, species number and diversity of fish were all higher at sites with dense cover of macrophytes (seagrass and macroalgae) than over unvegetated sandy sites. In general, fish assemblages mainly comprised juveniles of a few abundant taxa, e.g. Mugil cephalus, Mugilidae spp. and Leiognathus equulus at sites with mud substratum and Gerres oyena, Lethrinus harak and Sillago sihama at sites dominated by macrophytes. Multivariate analyses revealed significant separations in fish assemblage composition within the two creeks where the bottom substratum differed among sites. Overall, season seemed to have little effect on density, species number, diversity index (H′) and assemblage structure of fish. Water condition variables were also relatively stable across the season, although a short-term fluctuation primarily induced by decreased salinity, occurred during the heavy rains in April and May. Fish assemblage structure was not significantly affected by any of the abiotic factors tested. However, significant regressions were found between the other fish variables and environmental variables, but since these associations were mostly species-specific and generally inconsistent, we suggest that the overall distribution patterns of fish were mainly an effect of particular substrate preferences of fish species rather than contemporary water conditions.     

农户保护性耕作技术采纳行为及其影响因素研究综述

保护性耕作对于防止水土流失、改善土壤质量、保障国家粮食安全等具有重要意义。农户作为保护性耕作技术采纳的直接决策者,研究其采纳行为及其影响因素有助于促进保护性耕作技术的推广应用。基于此,论文系统梳理了国内外农户保护性耕作技术采纳行为及其影响因素的相关研究,以期为更有效的技术推广提供参考。结果发现：不同类型农户的保护性耕作技术采纳行为存在较大差异;农户保护性耕作技术采纳的研究区域较为集中;影响农户保护性耕作技术采纳行为的因素主要包括基本因素(个体特征和家庭特征)和差别因素(社会网络、风险偏好与风险感知、政府政策、心理认知及区位特征)2类。基于上述分析,论文提出了未来重点的研究方向：技术推广要考虑农户群体内的异质性;应注重各种研究理论间的融合;未来可进一步加强多学科交叉的系统研究。     

基于流动单元划分的储层测井二次解释

宝浪油田三工河组储层是在辫状河三角洲环境下沉积的河道砂体,非均质性严重,常规测井解释以砂层为解释单元,忽视砂层内部物性及渗流特征的差异,测井解释精度低.通过对砂层细分流动单元,采用聚类分析的方法将Ⅱ油组储层划分为5类独立的流动单元,在不同的流动单元内建立不同的测井解释模型.结果表明:模型计算出的孔隙度和渗透率精度更高,分别达到90%和80%以上,较好地反映了储层内部的非均质性特征,为精细油藏描述提供可靠的孔隙度、渗透率参数.     

地理流的空间模式：概念与分类

地理流被定义为地理对象在不同空间位置间的流动,而地理流通常涵盖了物质、信息、资金、能量等的空间转移。位置之间各种流的出现与强化,改变了以距离衰减效应为基本规律的传统认知。因此,针对地理流的研究将有助于从新的角度理解地理格局及其形成机理。本文认为可采用2种模型对地理流进行表达：① 将地理流抽象为包含起点和终点坐标的四元组（又称地理流正交模型）;② 将其抽象为包含起点坐标、流长度与流方向的模型（又称地理流极坐标模型）。在地理流表达模型的基础上,流空间可以定义为由两个二维空间通过笛卡尔积形成的四维空间。本文给出了流空间中4种距离（最大距离、加和距离、平均距离和加权距离）、不同距离下地理流的r邻域体积以及地理流密度等测度的定义。根据地理流极坐标模型中不同要素随机性的组合方式,将地理流模式分为随机、丛集、聚散、社区、并行与等长6种常见的单一模式,并梳理了不同模式的识别方法。单一模式之间的组合可以形成混合模式,而多类流数据之间可以形成多元流模式。在未来的研究中,地理流的基础统计理论、地理流模式的挖掘方法以及在实际问题中的应用将是其重要的发展方向。