Rapid Lateglacial tree population dynamics and ecosystem changes in the eastern Baltic region

A growing body of evidence implies that the concept of 'treeless tundra' in eastern and northern Europe fails to explain the rapidity of Lateglacial and postglacial tree population dynamics of the region, yet the knowledge of the geographic locations and shifting of tree populations is fragmentary. Pollen, stomata and plant macrofossil stratigraphies from Lake Kurjanovas in the poorly studied eastern Baltic region provide improved knowledge of ranges of north‐eastern European trees during the Lateglacial and subsequent plant population responses to the abrupt climatic changes of the Lateglacial/Holocene transition. The results prove the Lateglacial presence of tree populations (Betula, Pinus and Picea) in the eastern Baltic region. Particularly relevant is the stomatal and plant macrofossil evidence showing the local presence of reproductive Picea populations during the Younger Dryas stadial at 12 900–11 700 cal. a BP, occurring along with Dryas octopetala and arctic herbs, indicating semi‐open vegetation. The spread of Pinus–Betula forest at ca. 14 400 cal. a BP, the rise of Picea at ca. 12 800 cal. a BP and the re‐establishment of Pinus–Betula forest at ca. 11 700 cal. a BP within a span of centuries further suggest strikingly rapid, climate‐driven ecosystem changes rather than gradual plant succession on a newly deglaciated land. Copyright © 2009 John Wiley & Sons, Ltd.     

Nutrient (P, N) dynamics in the southwestern Kattegat, Scandinavia: sedimentation and resuspension effects

 The yearly nutrient supply from land and atmosphere to the study area in SW Kattegat is 10 900 tons of N and 365 tons of P. This is only few percent of the supply from adjacent marine areas, as the yearly transport through the study area is 218 000 tons of N and 18 250 tons of P. Yearly net deposition makes up 1340 tons of N (on average 2.5 g m^–2 yr^–1) and 477 ton of P (on average 0.9 g m^–2 yr^–1). Shallow-water parts of the study area have no net deposition because of frequent (>35% of the year) resuspension. Resuspension frequency in deep water is <1% of the year. Resuspension rates, as averages for the study area, are 10–17 times higher than net deposition rates. Because of resuspension, shallow-water sediments are coarse lag deposits with small amounts of organic matter (1.1%) and nutrients (0.04% N and 0.02% P). Deep-water sediments, in contrast, are fine grained with high levels of organic matter (11.7%) and nutrients (0.43% N and 0.15% P). Laboratory studies showed that resuspension changes the diffusive sediment water fluxes of nutrients, oxygen consumption, and penetration into the sediment. Fluxes of dissolved reactive phosphate from sediment to water after resuspension were negative in organic-rich sediments (13.2% organic matter) with low porosity (56) and close to zero in coarse sediments with a low organic matter content (2.3%) and high porosity (73). Fluxes of inorganic N after resuspension were reduced to 70% and 0–20% in relation to the rates before resuspension, respectively. Received: 10 July 1995 · Accepted: 19 January 1996     

Reliability assessment of serviceability performance of braced retaining walls using a neural network approach

In urban environments, one major concern with deep excavations in soft clay is the potentially large ground deformations in and around the excavation. Excessive movements can damage adjacent buildings and utilities. There are many uncertainties associated with the calculation of the ultimate or serviceability performance of a braced excavation system. These include the variabilities of the loadings, geotechnical soil properties, and engineering and geometrical properties of the wall. A risk‐based approach to serviceability performance failure is necessary to incorporate systematically the uncertainties associated with the various design parameters. This paper demonstrates the use of an integrated neural network–reliability method to assess the risk of serviceability failure through the calculation of the reliability index. By first performing a series of parametric studies using the finite element method and then approximating the non‐linear limit state surface (the boundary separating the safe and ‘failure’ domains) through a neural network model, the reliability index can be determined with the aid of a spreadsheet. Two illustrative examples are presented to show how the serviceability performance for braced excavation problems can be assessed using the reliability index. Copyright © 2005 John Wiley & Sons, Ltd.     

Origin of short period comets

Reasons for interest in the origin of short-period comets and the difficulties of computing their long-term dynamcal evolution are reviewed. The relative advantages of a source region in an extended inner core of the Oort cloud or a compact comet belt just beyond the planetary system are finely balanced, and it is premature to consider the problem solved. A complication is that some comets belonging to the Jupiter family may be part of a time-dependent system, possibly the remains of a giant comet such as Chiron which could have been part of the system 104 yr ago. The origin of short-period comets plays a pivotal role in many areas of solar system science: planet formation, the source of water (possibly life) on the terrestrial planets, the cratering record on the terrestrial planets and satellites of the outer planets, and the environmental impact posed by massive bodies and their decay products in the Earth's near-space environment.     

Dynamics of Mars-orbiting dust: Effects of light pressure and planetary oblateness

This paper deals with dynamics of impact ejecta from Phobos and Deimos initially on near-circular equatorial orbits around Mars. For particles emitted in a wide size regime of 1 micron and greater, and taking into account the typical particle lifetimes to be less than 100 years, the motion is governed by two perturbing forces: solar radiation pressure and influence of Mars' oblateness. The equations of motion of particles in Lagrangian non-singular elements are deduced and solved, both analytically and numerically, for different-sized ejecta. We state that the coupled effect of both forces above is essential so that on no account can the oblateness of Mars are be neglected. The dynamics of grains prove to be quite different for the ejecta of Phobos and Deimos. For Deimos, the qualitative results are relatively simple and imply oscillations of eccentricity and long-term variations of orbital inclination, with amplitudes and periods both depending on grain size. For Phobos, the dynamics are shown to be much more complicated, and we discuss it in detail. We have found an intriguous peculiar behavior of debris near 300 µm in size. Another finding is that almost all the Phobos ejecta with radii less than 30 µm (against the values of 5 to 20 µm adopted earlier by many authors) should be rapidly lost by collisions with martian surface. The results of the paper may be the base for constructing an improved model of dust belts that presumably exist around Mars.     

Phase transitions in the ISM a source of dissipative behaviour

By considering a simple fluid model, we investigate the role of phase transitions in the ISM on the galaxy- scale gas dynamics. Cooling and heating timescales in the ISM are typically shorter than typical galactic rotation timescales, so the individual phases in the ISM can be assumed to be in temperature equilibrium with the radiation field. Using this approximation we can construct an equation of state which depends upon the average density and mass fractions in the individual phases. Previous studies suggest that there is an equilibrium phase fraction as a function of pressure. We incorporate evolution towards this equilibrium state as a relaxation term with a time to obtain equilibrium . We derive a condition in terms of a critical Mach number when one dimensional shocks should be continuous. For small values of the relaxation time we show that the relaxation term acts like a viscosity. We show with one dimensional simulations that increasing causes shocks to become smoother. This work suggests that phase changes can strongly effect the gas dynamics of the ISM across spiral arms and bars.     

基于A- CGAN的深反射地震数据随机噪声压制方法研究

基于深度学习的地震数据噪声压制方法是当前地震数据去噪处理的重要方向。深度学习方法突破了传统滤波处理的局限,在对常规地震数据的噪声压制中表现出效率高、信噪分离效果好的特点。但针对深部弱有效反射数据,当前的深度学习方法特征提取能力有限,难以取得较好的去噪效果。笔者等结合深反射地震数据特点,针对当前深度学习噪声压制方法在特征提取及对数据集依赖上的局限,提出了基于注意力循环生成对抗网络(Attention Cycle- Consistent Generative Adversarial Networks,A- CGAN)的深反射地震数据随机噪声压制方法。借助循环一致生成对抗网络(Cycle- Consistent Generative Adversarial Networks,Cycle- GAN)的域映射思想,降低对数据集的要求。为了构建适用于深反射地震数据的去噪网络,从3个方面对Cycle- GAN进行改进：在Cycle- GAN的生成器（去噪器）中加入残差结构和注意力机制,用于加深网络深度和提高其特征提取能力;在Cycle- GAN的鉴别器中使用块判决,提高鉴别精度和准确度;在损失函数部分加入感知一致性损失函数,提升网络模型恢复纹理细节信息的能力。通过合成地震数据和实际深反射地震数据测试,验证了优化算法的有效性,体现了良好的应用价值。     

预测有效波高的深度学习模型研究

研究基于RNN、LSTM、GRU深度学习模型,针对NOAA浮标数据集中的44013、44014、44017浮标的数据,通过斯皮尔曼相关性分析提高模型预测效果。实验结果表明,在进行相关性分析后,S-RNN、S-LSTM、 S-GRU的预测效果均比原始RNN、LSTM、GRU模型预测效果好。此外,提出一种基于LSTM的LSTM-Attention 波高预测模型,并进行相关实验,量化LSTM-Attention模型的预测效果,实验结果表明LSTM-Attention模型有更好的预测效果。为评估模型的泛化能力,研究还提出了一种采用邻近浮标数据进行学习,预测浮标缺失数据的方 法。实验结果表明,该方法的预测精度可以达到97.93%。本研究为海浪预测提供了新的方法和思路,也为未来深 度学习模型在海浪预测中的应用提供了参考。     

高速远程崩滑动力学的研究现状及发展趋势

高速远程崩滑动力学机理的研究,一直是国内外工程地质与山地灾害领域崩塌、滑坡灾害研究的前沿热点问题。对这一研究领域,最近几十年来所取得的研究成果、研究现状及发展趋势进行了述评。厘定了高速远程崩滑的定义;详细地阐述了国内外众多典型重大高速远程崩滑灾害事件的几何学、运动学特征;概述了迄今为止所提出的高速远程崩滑形成的各种动力学机理;分析与讨论了高速远程崩滑动力学机理研究的发展趋势与动向;提出了今后进一步研究的总体思路、目标与应重点解决的关键问题。