Thermodynamic conditions of framework grain dissolution of clastic rocks and its application in Kela 2 gas field

Feldspar and clastic debris are the most important constituent framework grains of sedimentary clastic rocks and their chemical dissolution plays an essential role in the formation and evolution of the secondary pore in the reservoir rocks. On the basis of thermodynamic phase equilibrium, this study investigates the chemical equilibrium relationships between fluid and various plagioclase and K-feldspar in diagenesis of the sediments, particularly, the impact of temperature and fluid compositions (pH, activity of K+, Na+, Ca2+ and so on) on precipitation and dissolution equilibria of feldspars. Feldspar is extremely easily dissolved in the acid pore water with a low salinity when temperature decreases. The dissolution of anorthite end-member of plagioclase is related to the Ca content of the mineral and the fluid, higher Ca either in the mineral or in the fluid, easier dissolution of the feldspar. Moreover, the dissolution of albite end-member of plagioclase is related to Na of both the mineral and fluid,     

Taxonomic composition and seasonality of diatoms in three Dinaric karstic lakes in Croatia

The composition and temporal distribution of planktonic diatoms were studied in three small shallow karstic lakes – Modro Oko, Desne, and Kuti – throughout the seasons in 1994, 1995, 1997, and 1998. Physico-chemical parameters did not differ among the three lakes, except for NO₃ concentrations. Altogether, 137 diatom taxa within 41 genera were found in all three lakes combined. The highest number of taxa belonged to the genera Surirella (13), Navicula (12), andCymbella (9). Most taxa were tychoplanktonic, with true panktonic taxa contributing only 14% of the total. Despite their physical-chemical similarity, the lakes showed differences in the structure and dynamics of their diatom communities. Blooms occurred at different times in each of the lakes, and community similarity (Jaccard index) among all three was low (17.6–29.1%). The highest mean values of surface diatom abundance in Modro Oko (2199 cells L⁻¹), Desne (7900 cells L⁻¹), and Kuti (20690 cells L⁻¹) were not significantly different. Margalef's species richness index did not differ either among the lakes or seasonally. Only five diatom taxa in quantitative samples, and 22 taxa in net-samples, were common to all lakes. The most frequent diatoms were Cyclotella striata (56–93%), Fragilaria ulna (50–67%), and Asterionella formosa (27–83%). The most abundant taxa in Modro Oko, Desne, and Kuti were, respectively, Cymbella sp. 1 (402 cells L⁻¹), Cyclotella striata (3000 cells L⁻¹) and Asterionella formosa (20690 cells L⁻¹). Eighteen taxa were recorded for the first time in Croatia: Amphora commutata, Caloneis amphisbaena var. subsalina, C. silicula var. peisonis, Campylodiscus echeneis, Coscinodiscus lacustris, Denticula elegans, Gyrosigma acuminatum var. lacustre, G. distortum var. parkeri, G. fasciola, Navicula menisculus var. menisculus, N. pusilla, N. helensis, Stauroneis parvula var. prominula, Staurosira construens f. subsalina, Stephanodiscus sp. 1, Surirella gracilis, S. ovalis, and S. striatula. The genus Stephanodiscus was recorded for the first time in Croatia.     

天津沿海赤潮发生的基本特征研究

使用《天津市海洋环境状况公报》中的数据,分析天津沿海赤潮发生的特征。通过分析发现2005-2017年天津海域共发生赤潮30次,累计发生面积4 695 km2。天津沿海赤潮发生主要集中在夏季,又以8月最多,占赤潮发生总数的30%。通过赤潮优势种的分析发现近年来天津海域赤潮生物的优势种类有不断增加的趋势。同时本文从水文气象条件诱发赤潮的角度分析得出了天津沿海最易发生赤潮的天气形势,天津沿海25~27℃水温范围内赤潮发生较多,最适宜赤潮发生的风速为3、4级风,风向为东南风、西南风。     

复杂卫星图像中的小目标船舶识别

船舶作为海上的重要目标,实现对船舶自动识别有重要的意义。针对卫星图像中云雾、海岸背景等复杂海情对船舶识别带来的干扰,以及小目标船舶高漏检率问题,本文提出一种多尺度深度学习模型训练策略,在此基础上构建了一种船舶识别的深度学习网络,该网络可分为多尺度训练、特征提取、生成目标建议区域、船舶分类这4个部分。首先,采用多尺度的训练策略,将多尺度的船舶样本送入网络中进行训练,这样在训练样本中加入了大量小目标船舶的样本,使网络充分提取到小目标船舶的特征;其次,通过卷积神经网络对目标船舶进行特征自适应提取;然后,目标区域建议网络可依据卷积神经网络提取到的特征,在图像中找到感兴趣目标区域,即框定船舶的位置;最后,通过多个全连接层的组合,将高维特征映射到一个4元组中,再运用分类函数输出每一类船舶的概率值,概率值最大的则为该船舶的类别。同时为解决云雾遮挡和海岸背景的干扰,采用了一种负样本增强学习的方法,在样本数据集中加入了大量只含有云雾和海岸背景的图片,进行负样本扩充,增强网络模型对云雾及海岸背景的特征学习能力,以此解决复杂海情的影响。实验结果表明,所提方法有效解决了复杂海情条件下的船舶识别难,以及小目标船舶识别难的问题,实现了复杂海情条件下的船舶识别。同时,与现有成熟的深度学习目标识别算法相比,本文算法的精确度和召回率分别提升了6.98%和18.17%,所训练的模型具有良好的泛化能力和鲁棒性。     

基于ArcGIS平台与质检软件的地理国情普查快速质量控制方法

"信息化质检平台IGCES"及"地理国情普查成果质检软件"是某省地理国情普查采用的主要人机交互质量控制软件。该类软件使用复杂,某些检查项耗时长,难以满足快速化质检的要求,本文结合地理国情普查具体实施要求,提出"人工详查、基于ArcGIS 10.1平台与质检软件概查"的快速质检方法,为国家战略规划制定、空间规划管理、区域政策制定、灾害预警、科学研究等领域的地理国情监测提供有力的数据保障。     

地理世情监测标准建设初探

探讨地理世情监测标准规范建设的需求与意义,归纳梳理国内外相关标准规范建设的现状,分析目前我国地理世情监测标准规范建设存在的不足,提出相关建议,以推动地理世情监测工作的开展。     

基于MapGIS的地理国情普查研究与应用

随着第一次全国地理国情普查工作的全面开展,出现了众多地理国情信息数据处理软件,数据的采集和制作也遇到了各种各样的问题,我们利用MapGIS平台对软件的内部模块和功能进行了有效的利用,并得到了很好的应用效果。     

地理国情普查项目地表覆盖数据成果的质量控制

地理国情普查项目是全新的测绘地理信息项目,文章结合生产阶段及检查验收过程中地表覆盖数据生产容易出现的质量问题,通过认真分析影响质量的因素,从管理要素、生产过程、两级检查3个方面谈了解决该问题的部分质量控制措施,以便于更好地提高地理国情普查项目成果的质量。