多波地震深度学习的油气储层分布预测案例

有机并有效利用纵波与转换横波在油气储层敏感度上存在的差异,有助于突出地震油气储层特征,有助于提高地震油气储层分布边界刻画的精度.基于此,本文设计了一种卷积神经网络与支持向量机方法相结合的多波地震油气储层分布预测的深度学习法（Deep Learning Method）.首先,利用莱特准则剔除所生成的多波地震属性中可能存在的异常值降低网络变体数量.然后,通过能突出多波地震油气储层特征的聚类算法和无监督学习算法构建隐藏层,用于增加网络共享,提取油气特征.最后,将增加网络罚值后的井点样本作为支持向量机预测的输入样本,以降采样后的C3卷积层属性作为学习集,进行从已知到未知的地震油气储层的预测.本方案应用于HG地区晚三叠统HGR组的碳酸盐岩油气储层预测,所预测的地震油气储层边界更加清晰,预测结果与实际情况基本吻合.应用结果表明：本论文方案不仅具有可行性,且具有有效性.

     

考虑动态克林伯格系数的煤储层渗透率预测模型

随着储层压力的降低,克林伯格效应对渗透率的影响越来越大.现有的煤储层渗透率预测模型大都忽略了克林伯格系数的变化,其预测结果与实际生产存在一定的差异,尤其是在低储层压力阶段.本文以体积不变假设为基础,基于火柴棍模型给出在储层压力降低过程中动态克林伯格系数的计算公式,并建立考虑动态克林伯格系数的渗透率预测模型;深入分析在煤储层压力降低过程中,煤储层渗透率和克林伯格系数的变化规律.研究结果表明：随着储层压力的降低,克林伯格系数呈先增大后减小的变化趋势;在相同储层压力下,克林伯格系数随渗透率增加呈指数减小趋势,随温度增加呈线性增大趋势.本文建立的渗透率模型参数简单易获取,预测结果与实际煤储层渗透率变化规律符合性较好,尤其是在低储层压力阶段,能准确预测煤储层渗透率变化.

     

Weathering reflected by the chemical composition of alluvial soils from the Zeya and Selemdzha river valleys

The paper examines the chemical composition of alluvial soils from the Zeya and Selemdzha River valleys, one of the branch rivers in the Amur River basin. It has been established that the compositions of these soils are characterized by a relative shortage of practically all analyzed macro- and microelements as compared with UCC (upper continental crust) and PAAS (post- Arhean Australian average shale) compositions. The chemical composition of alluvial soils approximates that of the soils in NE China and Korea due to similar climatic and geographic conditions. On the basis of relationships between major and trace elements and peculiarities of trace elements concentration it is shown that the alluvial soils in the middle reaches of the Zeya River were formed by the sources largely transformed by chemical weathering processes. By comparison, the soils in the Selemdzha River and lower reaches of the Zeya River (below the Selemdzha River mouth) are dominated by material to a large degree reworked by the physical processes rather than chemical ones. This inference conforms with specific geological settings of the water-collector.     

Spatial distribution of cyanotoxins and ratios of microcystin to biomass indicators in the reservoirs of the Volga,Kama and Don Rivers,the European part of Russia

We studied the distribution of cyanotoxins and potential producers, as well as the variability of microcystin to biomass parameters (chlorophyll-a; MC/Chl-a; and biovolume; MC/BV) in 12 drinking water reservoirs of the world’s largest reservoir system, the Volga-Kama-Don cascade (European part of Russia) during the summers of 2016 and 2018. MC concentrations varied from below 0.1 μg L⁻¹ in June up to 16.4 μg L⁻¹ in August and exceeded 1 μg L^-1 in 25 % of the samples. This MC variability was associated to changes in the abundance of widespread bloom formers such as Microcystis spp., Dolichospermum spp. and Planktothrix agardhii. Ratios of MC/Chl-a and MC/BV_cyano ranged up to 0.88 μg μg⁻¹ and 4.5 μg mm⁻³, respectively. Together with microcystin profiles MC/BV_cyano ratios characterized cyanobacterial populations along the reservoir cascade and they indicated a potential toxin hazard better than MC/Chl-a. The neurotoxin anatoxin-a was observed only in the most southern and hypereutrophic Tsimlyansk Reservoir (maximum 0.01 μg L⁻¹). Toxin gene analysis revealed that MC mostly originated from Microcystis and Dolichospermum. During their co-existence up to 14 MC congeners co-occurred. Cuspidothrix issatschenkoi cf. Raphidiopsis mediterranea was identified as possible neurotoxin producers.     

The contribution of primary production to the overall organic matter pool in Volga reservoirs

The results of studies of the concentrations and the ratios between the forms of organic matter in Volga reservoirs are presented. The reservoirs under consideration have different morphometric characteristics. flowage, and trophic status. Total organic matter content varies from 7.1 to 11.8 g/m³ or from 39.1 to 70.8 g/m² with the proportion of labile fraction of 4.4–15.2%. The primary production of plankton, the daily value of which accounts for 14–40% of labile organic matter, has a notable effect on its concentration and input. Organic matter pool in Volga reservoirs forms under the effect of the morphometric features of reservoirs, water exchange, and the drainage area. The total characteristics of organic matter in the Upper Volga reservoirs have not experienced significant changes over the 20-year period. However, data for 2005 show that the contribution of primary production to the total concentration of organic matter and its labile fraction have decreased due to a decline in the productivity of planktonic communities recorded in recent years.     

Physical characteristics of high-rank coal reservoirs in different coal-body structures and the mechanism of coalbed methane production

The physical characteristics of coal reservoirs are important for evaluating the potential for gas desorption, diffusion, and seepage during coalbed methane (CBM) production, and influence the performance of CBM wells. Based on data from mercury injection experiments, low-temperature liquid nitrogen adsorption, isothermal adsorption, initial velocity tests of methane diffusion, and gas natural desorption data from a CBM field, herein the physical characteristics of reservoirs of high-rank coals with different coal-body structures are described, including porosity, adsorption/desorption, diffusion, and seepage. Geometric models are constructed for these reservoirs. The modes of diffusion are discussed and a comprehensive diffusion-seepage model is constructed. The following conclusions were obtained. First, the pore distribution of tectonically deformed coal is different from that of normal coal. Compared to normal coal, all types of pore, including micropores (<10 nm), transitional pores (10–100 nm), mesopores (100–1000 nm), and macropores (>1000 nm), are more abundant in tectonically deformed coal, especially mesopores and macropores. The increase in pore abundance is greater with increasing tectonic deformation of coal; in addition, the pore connectivity is altered. These are the key factors causing differences in other reservoir physical characteristics, such as adsorption/desorption and diffusion in coals with different coal-body structures. Second, normal and cataclastic coals mainly contain micropores. The lack of macropores and its bad connectivity limit gas desorption and diffusion during the early stage of CBM production. However, the good connectivity of micropores is favorable for gas desorption and diffusion in later gas production stage. Thus, because of the slow decline in the rate of gas desorption, long-term gas production can easily be obtained from these reservoirs. Third, under natural conditions the adsorption/desorption properties of granulated and mylonitized coal are good, and the diffusion ability is also enhanced. However, for in situ reservoir conditions, the high dependence of reservoir permeability on stress results in a weak seepage of gas; thus, desorption and diffusion is limited. Fourth, during gas production, the pore range in which transitional diffusion takes place always increases, but that for Fick diffusion decreases. This is a reason for the reduction in diffusion capacity, in which micropores and transitional pores are the primary factors limiting gas diffusion. Finally, the proposed comprehensive model of CBM production under in situ reservoir conditions elucidates the key factors limiting gas production, which is helpful for selection of reservoir stimulation methods.     

松辽盆地丰乐地区营城组火山岩储层预测

徐家围子断陷的主要产气层段为营城组火山岩段地层,所以研究该地层对寻找天然气藏具有非常重要的意义.本文利用丰乐地区的高分辨率地震资料、岩石物性资料、测井资料对下白垩统营城组火山岩进行构造精细解释,火山岩地震相及岩相分析从而进行储层预测,利用均方根振幅、瞬时频率能量及瞬时吸收系数能量等地震属性分析结合地震反演技术精细刻画火山岩储层.结果认为：均方根振幅、瞬时频率能量和瞬时吸收系数能量能较好预测火山岩相和储层物性;有利储层在反演剖面上表现为中低阻抗特征,中等密度;较好储层主要分布在工区的西部与中部.     

深度Transformer迁移学习的页岩气储层核心参数预测案例

地层纵横向非均质性强, 工区间数据分布存在差异.这导致基于已有工区数据构建的机器学习储层参数预测模型, 推广到新工区会存在较大预测误差.常规地质方法是在岩心与测井响应特征分析基础上建模, 利用测井资料计算储层参数, 流程复杂.该方法需要岩心校准模型, 同样难以快速推广到新的工区.考虑地层纵横向非均质性, 本文设计了一种深度Transformer迁移学习网络, 通过已有工区的测井与岩心资料构建预测模型, 实现未取心新工区储层参数快速准确预测.首先利用无监督学习算法-孤立森林剔除测井数据中存在的异常噪声数据.然后设计Transformer特征提取网络, 提高网络特征提取能力, 以此深入挖掘测井数据与储层参数的内在联系.最后设计深度迁移学习网络, 构建网络损失函数, 利用随机梯度下降算法优化网络参数, 实现储层参数准确预测.本方案应用于四川南部地区五峰组—龙马溪组页岩储层参数孔隙度、总有机碳含量和总含气量预测.实验结果与工区校正后计算结果、主流机器学习模型预测结果对比, 本方案结果与岩心数据具有更高的一致性.应用结果表明: 本文方案具有实用性、有效性和可推广性.

     

水库浮游细菌生产和呼吸过程代谢内外源有机碳研究

全球气候变化和人类活动的加剧影响水生生态系统中溶解性有机碳(DOC)的种类和浓度.浮游细菌生产和呼吸过程对碳源变化的响应,可能影响水体乃至区域的碳循环特征.本研究选取了南京市11座中小型水库,利用碳稳定同位素技术和双端元混合模型定量分析浮游细菌在生产和呼吸代谢过程中对内外碳源的利用.依据DOC中内外碳源的比例,将11座...     

Assessment of the environmental conditions of the Zeya River and its tributaries in the construction zone of the Nizhne-Zeiskaya HPP

Water chemistry, chlorophyll “a” concentration in phytoplankton and periphyton, and the structure of bacterioplankton communities of the Zeya River and its tributaries are assessed in the construction zone of the Nizhne-Zeiskaya HPP. The results of multidisciplinary studies demonstrate the effect of natural and anthropogenic factors on the formation of water quality in the examined part of the Zeya River.     

Deformations of Point and Midstream Bars on Mountain Rivers

Based on data of long-term permanent observations of the processes of erosion and accumulation in the Bzyb River basin, relationships between the deformations of the river channel, the hydrological regime of this mountain river, and the rate and character of the debris formation in different areas of its basin are analyzed.     

Prediction of sedimentation in reservoirs by combining catchment based model and stream based model with limited data

Estimation of sedimentation in reservoirs helps in the management and design of the reservoir's useful capacity. This research was done on the Awash River basin at the Koka Dam Reservoir in Ethiopia. The method applied was the loose integration of the Soil and Water Assessment Tool(SWAT) model and Hydrologic Engineering Center-River Analysis System(HEC-RAS) model for the estimation of the sediment load reaching the reservoir. The SWAT model was used for the estimation of erosion at the catchment level,and the HEC-RAS model was applied to estimate the sediment transport in the river channel. The implemented method allows sedimentation in the floodplains and bed shear stress to be considered in the sediment modeling, which cannot be considered in the SWAT model. In addition, the river cross sectional properties and the hydrodynamic processes in the rivers were considered in the modeling process. The data used in this study are a combination of i) observed data collected by government agencies, ii) data available online in data repositories, and iii) data extracted from remote sensing in the Shuttle Radar Topographic Mission(SRTM) Digital Elevation Model(DEM). The calibration and validation of the SWAT model was done by using Sequential Uncertainty Fitting(SUIF-2) calibration and validation tools. The HEC-RAS model was calibrated by adjusting the roughness factor. The output from the integrated approaches gives better estimates of flow and sediment near the inlet to the reservoir, with coefficients of determination of 0.85 and 0.67, respectively, and Nash Sutcliffe coefficients of model fit efficiency of 0.90 and 0.62, respectively, for daily simulations.     

Seasonal Short‐Term Prediction of Dissolved Oxygen in Rivers via Nature‐Inspired Algorithms

This study challenges the use of three nature‐inspired algorithms as learning frameworks of the adaptive‐neuro‐fuzzy inference system (ANFIS) machine learning model for short‐term modeling of dissolved oxygen (DO) concentrations. Particle swarm optimization (PSO), butterfly optimization algorithm (BOA), and biogeography‐based optimization (BBO) are employed for developing predictive ANFIS models using seasonal 15 min data collected from the Rock Creek River in Washington, DC. Four independent variables are used as model inputs including water temperature (T), river discharge (Q), specific conductance (SC), and pH. The Mallow's Cp and R² parameters are used for choosing the best input parameters for the models. The models are assessed by several statistics such as the coefficient of determination (R²), root‐mean‐square error (RMSE), Nash–Sutcliffe efficiency, mean absolute error, and the percent bias. The results indicate that the performance of all‐nature‐inspired algorithms is close to each other. However, based on the calculated RMSE, they enhance the accuracy of standard ANFIS in the spring, summer, fall, and winter around 13.79%, 15.94%, 6.25%, and 12.74%, respectively. Overall, the ANFIS‐PSO and ANFIS‐BOA provide slightly better results than the other ANFIS models.     

Oceanographic frontal structure and biological production at an ice edge

Marginal ice edge zones (MIZ) are unique frontal systems with air-ice-sea interfaces. Phytoplankton blooms, which occur along the edge of the melting ice pack in spring, are strongly related to the air-ice-sea interactive processes. In spring 1982, during a cruise to the Bering Sea ice pack, hydrographic sections, including standard biological oceanographic parameters, were collected across the MIZ showing such enhanced phytoplankton bloom populations in the ice edge. During this period the ice edge retreated at speeds of 6 to 38 cm s^?1. Associated with the retreating ice edge were a faster moving upper layer oceanic front that kept pace with the retreating ice edge, and a nearly stationary deeper front. In the presence of light, the phytoplankton blooms are shown to be associated with, and primarily controlled by enhanced density stratification and frontal structure due to ice melt during the spring ice retreat. The ice melt water forms stratification that helps to maintain the phytoplankton within the photic zone. The ice edge blooms can be differentiated from open water blooms by the stratification mechanism; in MIZ blooms stratification is due to low salinity melt water as opposed to temperature derived stratification in most open water blooms. In addition, in the series of cross sections collected, a unique biophysical interaction was observed when the MIZ front moving north with the spring retreat, came in contact with a fixed shelf front forming a ‘dish’ shaped hydrographic structure within which a major phytoplankton bloom was observed. We suggest that upwelling from the tidally driven shelf front supplied nutrients to the surface waters extending the life of the bloom. Wind-driven ice edge upwelling was also observed but was difficult to distinguish from the shelf front circulation.In this same set of ice edge cross sections, a cold water mass was observed at the surface in the MIZ. This water mass was subsequently overridden by warmer water forming a cold tongue structure above the pycnocline and seaward of the shelf front. We suggest that this cold tongue was transient in nature, and illustrative of one mechanism by which the T-S characteristics of high latitude shelf waters are formed and altered.     

Small Rivers and the Ecological State of an Area

Data from surveying small rivers in Moscow and Tver oblasts are used to assess the anthropogenic load on their watersheds. Expert estimates of the effect of such load onto the state of surface water are made. Integral characteristics and similarity criteria of the hydroecological conditions of small rivers are substantiated.     

Analysis of characteristic rhythms of heliogeomagnetic activity and their roles in synchronization of rhythms of biological objects

Rhythms of heliogeomagnetic activity and their dynamics in the solar activity (SA) cycles are analyzed. Heliogeomagnetic rhythms are the mixture of a periodic signal and noise component, which differs from a periodic signal by a substantial suddenness. Therefore, the filtering methods optimized for smoothness were used to analyze these rhythms and to separate rhythm periodic components. Short-period rhythms of the Kp index of geomagnetic activity and Wolf numbers (W) have been revealed. The rhythm dynamics in SA cycles has been indicated. Clearly defined near-seven-day, semi-seven-day, and other components, which are harmonics of the solar rotation period, are present in the rhythms of the Kp index in contrast to W. A comparison of the Kp and Dst rhythms indicates that the seasonal and annual variations in these indices are substantially similar. It has been indicated that the rhythms of heliogeomagnetic activity could be the external synchronizers of biological rhythms, having integrated into the endogenous temporal structure of these rhythms in the process of evolution, and that the dependence of the population macroparameters on W is much weaker than on Kp.     

Formation of Frazil Ice and Ice Jams on the Rivers of the Yenisei River Basin

The effect of different factors on the duration of spring ice jams and the flood water level in the Siberian rivers is considered. Observational data on the frequency of ice jams of different probabilities within some reaches of the Yenisei River are presented.     

On the Effect of the Froude Number and Hydromorphometric Parameters on Sediment Transport in Rivers

Water Resources - A number of requirements for sediment transport formulas are considered for using in mathematical and physical models of alluvial channels. Some formulas (Bagnold, Petrov,...