基于AMT的龙首山成矿带西岔地段马路沟断裂带深部发育特征

为减少铀矿钻探工程风险,基于音频大地电磁法(AMT)资料,对龙首山成矿带中的马路沟断裂及次级断裂的深部发育特征进行了探究。通过对马路沟断裂与次级断裂正演理论模型响应特征分析,论证了方法的可行性,在此基础上,结合地质、岩石电性参数和钻探资料,对反演电阻率断面进行了推断解释,大致查明了区内断裂的深部发育特征。研究结果表明:马路沟断裂(F101)走向NW,倾角约80°,其中X501—X502剖面之间倾向NE,X502—X506剖面倾向SW;次级断裂F102、F103走向NW,均倾向NE,倾角75°~80°;上述断裂具有切割深度大、倾角较陡特征。该结果为后期钻探工程的布置提供了依据,提升了铀矿地质勘查的效果与效益。     

中国煤层气资源控制程度及可靠性分析

准确认识深部条件下气体和水分的赋存状态、相对含量及分布特征,对煤层气高效勘探开发具有重要指导意义。基于理论模型、分子模拟和气水演变分析,明确了煤层中气、水的赋存状态,揭示了气、水动态运聚界限和动态演化过程。考虑煤−水界面作用、水的可动性及赋存状态,煤中水可分为可动水(重力水和毛细水)、束缚水(吸附水、沸石水、结晶水和层间水)、结构水,其中毛细水、重力水和吸附水由孔隙主导,沸石水、结构水、结晶水和层间水由矿物主导。分子模拟结果显示,水分子在0.7 nm孔隙中可以饱和充填,吸附和解吸路径一致,在更大孔隙中出现弱吸附层和自由态。水分子吸附过程表现为单分子含氧基团吸附、单层强吸附、多层弱吸附、水团簇形成和充填孔隙等阶段。甲烷分子在1.5 nm孔隙可存在3层稳定充填吸附,在较大孔隙中(>1.5 nm)即以单层吸附和游离态共存,游离态在介孔及更大孔隙中普遍存在。结合上述吸附−游离气存在界限,改进了游离气和吸附气理论计算公式,为含气量计算提供新思路。深部热成因煤层气是煤大规模生排烃之后的残余气,在排烃过程中发生气驱水和水分蒸发扩散,残余水分为束缚水和结构水,后期无法改变。假定静水压力20 MPa,在0、5、10、15和20 MPa储层压力下,外来水分可入侵最大孔径为7、9、13、27 nm和不侵入。受差异保存条件控制,煤成气除了形成超压和欠压等差异含气系统外,还可能在煤系形成多类型含气模式。上述研究明确了煤层气、水微观赋存机制及形成演化模式,对深部煤层气富集特征及高效开发设计具有指导意义。

     

Overview of US Minerals Management Service Activities in Deepwater Research

The US Minerals Management Service (MMS) is responsible for safe and environmentally sound management of offshore oil and gas resources. In recent years, there has been a tremendous increase in both activities and plans for deepwater development of these resources in the Gulf of Mexico (GOM). As the industry meets the enormous technical challenges involved, there is a concomitant effort to assure environmental protection of deep-ocean and coastal resources. This paper will outline the research planning and program currently underway to assess impacts, identify potential problems, and discover more about unknown resources of the deep ocean.     

论海洋深度空间控制权——从海洋深度空间能源开发谈起

论文在提出海洋和“新海洋”广度、深度空间概念的基础上,进一步诠释了作为新的海洋权益的控制权,及其在未来政治、经济、外交、军事等领域的重要作用。由于海洋蕴藏着巨大的可再生能源,论文从能源开发利用的视角分析了开发利用海洋深度空间的重要战略意义,并对解决我国海洋争端,开发海洋新能源及综合践行海洋深度空间权战略提出了应对策略。     

Effects of land use and cultivation time on soil organic and inorganic carbon storage in deep soils

The vertical distribution and exchange mechanisms of soil organic and inorganic carbon(SOC, SIC) play an important role in assessing carbon(C) cycling and budgets. However, the impact of land use through time for deep soil C(below 100 cm) is not well known. To investigate deep C storage under different land uses and evaluate how it changes with time, we collected soil samples to a depth of 500 cm in a soil profile in the Gutun watershed on the Chinese Loess Plateau(CLP); and determined SOC, SIC, and bulk density. The magnitude of SOC stocks in the 0–500 cm depth range fell into the following ranking: shrubland(17.2 kg m~(-2)) grassland(16.3 kg m~(-2)) forestland(15.2 kg m~(-2)) cropland(14.1 kg m~(-2)) gully land(6.4 kg m~(-2)). The ranking for SIC stocks were: grassland(104.1 kg m~(-2)) forestland(96.2 kg m~(-2)) shrubland(90.6 kg m~(-2)) cropland(82.4 kg m~(-2)) gully land(50.3 kg m~(-2)). Respective SOC and SIC stocks were at least 1.6-and 2.1-fold higher within the 100–500 cm depth range, as compared to the 0–100 cm depth range. Overall SOC and SIC stocks decreased significantly from the 5 th to the 15 th year of cultivation in croplands, and generally increased up to the 70 th year. Both SOC and SIC stocks showed a turning point at 15 years cultivation, which should be considered when evaluating soil C sequestration. Estimates of C stocks greatly depends on soil sampling depth, and understanding the influences of land use and time will improve soil productivity and conservation in regions with deep soils.     

A hybrid integrated deep learning model for the prediction of citywide spatio-temporal flow volumes

ABSTRACT

The spatio-temporal residual network (ST-ResNet) leverages the power of deep learning (DL) for predicting the volume of citywide spatio-temporal flows. However, this model, neglects the dynamic dependency of the input flows in the temporal dimension, which affects what spatio-temporal features may be captured in the result. This study introduces a long short-term memory (LSTM) neural network into the ST-ResNet to form a hybrid integrated-DL model to predict the volumes of citywide spatio-temporal flows (called HIDLST). The new model can dynamically learn the temporal dependency among flows via the feedback connection in the LSTM to improve accurate captures of spatio-temporal features in the flows. We test the HIDLST model by predicting the volumes of citywide taxi flows in Beijing, China. We tune the hyperparameters of the HIDLST model to optimize the prediction accuracy. A comparative study shows that the proposed model consistently outperforms ST-ResNet and several other typical DL-based models on prediction accuracy. Furthermore, we discuss the distribution of prediction errors and the contributions of the different spatio-temporal patterns.