1980—2020年中国九大流域蒸散发及其组分时空评估

蒸散发是地表陆气水分交换的纽带,准确量化蒸散发的时空演变格局对于水资源规划与管理至关重要。本文基于GLEAM模型的蒸散发及其组分数据集,借助7个通量观测站数据、120个流域的流域水量平衡及PML_V2蒸散发产品,在中国九大流域系统评估了GLEAM-ET产品,分析了植被恢复背景下,蒸散发（ET）及其组分（植被蒸腾Ec,截留蒸发Ei,土壤蒸发Es）在1980—2020年的时空演变格局。本文主要得到以下结论：① GLEAM-ET产品在中国九大流域具有较好的适用性,其性能与气候类型有关,干旱区效果优于湿润区。此外,GLEAM与PML_V2模型在九大流域相关性较好（R>0.7）,分布格局与变化趋势整体保持一致。② 全国尺度上,ET均值为416.88 mm,增长速率为1.21 mm/a。Ec与ET均呈自东南向西北递减的分布格局,而Es与其相反。Ec、ET在九大流域均呈显著增加趋势（p<0.001）。Ei、Es在季风区流域分别呈显著增加和显著减小趋势;在内陆区流域呈不显著减小（p>0.05）和显著增加趋势。③在植被恢复背景下,ET组分比例发生了变化。Ec占比变化存在南北差异,南方流域Ec占比均减小,北方流域均增加。Ei占比在各流域均增加,Es占比均减小。黄河流域ET组分对植被恢复的响应最为明显,Ec占比增加了5.21%,Es占比减小了5.56%。     

江西省都昌县山地与平原植被界线提取

由于生长环境、地带性分布规律和垂直差异,山地与平原地区的不同植被类型在遥感影像上存在“同物异谱”及“异物同谱”现象,易导致土地类型的误分。为避免此类错分,在进行土地覆被类型解译之前,应首先确定平原、山地植被的边界。本文在遥感影像聚类分析、GIS空间分析及数理统计分析技术的支持下,以江西省都昌县北部地区为研究区域,基于高分一号（GF-1）遥感卫星影像及地形要素,完成了山地植被与平原植被的界线提取。实验结果显示,本研究得到的植被界线精度高达99.47%和96.28%。与单纯基于遥感影像分类得到的植被边界成果相比,精度提高了近25%和23%。研究证明,高分辨率遥感影像与地形因子结合后,计算得到的山地平原划分界线精度有明显提高,可以满足高分辨率遥感影像土地覆被类型解译研究的需要。     

Morphodynamics of an Artificial Cobble Beach in Tianquan Bay,Xiamen, China

By tracking and monitoring the profile configuration, topography, and hydrodynamic factors of an artificial cobble beach in Tianquan Bay, Xiamen, China over three consecutive years after its completion, we analyzed the evolution of its profile configuration and plane morphology, and its storm response characteristics. The evolution of the profile configuration of the artificial cobble beach in Tianquan Bay can be divided into four stages. The beach was unstable during the initial stage after the beach nourishment the profile configuration changed obviously, and an upper concave composite cobble beach formed gradually, which was characterized by a steep upper part and a gentle lower part. In the second stage, the cobble beach approached dynamic equilibrium with minor changes in the profile configuration. At the third stage the beach was in a high-energy state under the influence of Typhoon Meranti, and the response of the artificial cobble beach differed significantly from that of the low-tide terrace sandy beach. Within a short time, there was net onshore transport of cobbles in the cross-shore direction. The beach face was eroded, the beach berm was accumulated, and the slope of the beach was steepened considerably. In the alongshore direction, there was notable transport of cobbles on the beach from east to west along the shore, and the total volume of the beach decreased by 4.5×103 m~3, which accounted for 50% of the total amount of beach volume lost within three years. The fourth stage was the restoration stage after the typhoon, characterized by a little gentler profile slope and the increase in width and the decrease in height of beach berm. Because of the action of waves and the wave-driven longshore current caused by the specific terrain and landform conditions along the coast(e.g., coastal headlands, near-shore artificial structures, and reefs), the coastline of the artificial cobble beach gradually evolved from being essentially parallel to the artificial coast upon completion to a slightly curved parabolic shape, and three distinct erosion hotspots were formed on the west side of the cape and the artificial drainpipe, and the reefs. Generally, the adoption of cobbles for beach nourishment on this macro-tidal coast beach with severe erosion has yielded excellent stability and adaptability.     

基于深度学习的钨钼找矿靶区预测方法研究

随着矿产勘查工作由浅部矿向深部隐伏矿、由易识别矿向难识别矿发展,找矿难度日益增大,地质专家越来越重视新理论、新方法、新技术的应用。深度学习作为人工智能的前沿领域/技术,对于实现矿产资源预测“智能化预测评价”具有得天独厚的优势。本文以陕西省镇安县西部钨钼矿集区单元素化探异常原始数据为基础,提出了基于深度学习的钨钼矿产评价方法。该方法以归一化地球化学数据作为模型训练数据,通过深度学习中深度自编码网络方法实现异常值提取进而识别重点成矿有利地段,实现矿产资源找矿远景区定性预测。研究结果表明,在对957条单元素化探异常原始数据分类且做好模型标签后,整个过程在计算机的“黑盒子”中自动完成学习和预测,相较于传统预测研究方法,本文方法具有自动化程度高和客观性强的特征。此外,本文利用已知矿点构建训练数据集,采用随机森林方法对预测区进行矿产资源找矿靶区预测圈定,为进一步缩小找矿靶区范围提供科学依据。     

钻孔灌注桩桩身混凝土强度横向差异简析

根据实际现场施工和钻孔灌注桩检测资料对桩身边缘混凝土强度与桩中心混凝土强度有所不同进行初步探讨。     

Gamma-ray bursts from nearby neutron stars

We interpret the puzzling-ray bursts as emitted by cooling sparks from the surface of spasmodically accreting, old neutron stars. Their spiky, anisotropic radiation is oriented w.r.t. the galactic disk via interstellar accretion, whose orbital angular momentum tends to counteralign with the galactic spin; in this way, larger source numbers in directions of the galactic disk are compensated by smaller beaming probabilities, resulting in a near-isotropic arrival distribution, as observed by BATSE. The source distances range between 10 pc and 500 pc. Their radiated energies are of order 10³⁵ erg, corresponding to accreted clumps (blades) of typical mass 10¹⁵ g per burst. Magnetic surface field strengths range between 10¹⁰ and 10¹² G, somewhat weaker than those of newborn neutron stars.     

Small Fe bearing ring molecules of possible astrophysical interest: molecular properties and rotational spectra

A number of small, cyclic molecules containing several carbon atoms in their ring structure has been identified in different astrophysical environments. It is the aim of this work to study important molecular properties of such heterocyclic species bearing an iron atom, which is one of the most abundant cosmic elements. Quantum theoretical calculations based on a density functional approach have been employed to investigate physical properties of six small cyclic carbon and hydrocarbon molecules containing iron as a hetero atom, viz. FeC₂H _n and FeC₃H _n (n=0,2,4). The full geometry optimisation at the chosen level of electronic structure theory (B3LYP/6-31G(d)) including vibrational anharmonicities and non-rigidity, has furnished values for the rotational constants of these species to an expected accuracy of about one per cent. We present structural, electronic, vibrational, and rotational molecular properties including line frequencies, line strengths, and transition probabilities. These results may be helpful for identifying these molecules in future laboratory experiments in view of tentative astronomical observations.     

S－闭空间上的函数

研究了当Ｘ为Ｓ－闭空间时，Ｃ（Ｘ）的子集Ｆ为紧的充要条件，从而扩充了著名的Ａｓｃｏｌｉ定理。     

On measuring low-degree p-mode frequency splitting with full-disc integrated data

The standard method of measuring rotational splitting from solar full-disc oscillation data, based on maximum-likelihood fitting of multi-Lorentzian profiles to oscillation power spectra, systematically overestimates the splitting. One of the reasons is that the maximum likelihood estimators (MLE) become unbiased only asymptotically as the number of data tends to infinity; for a finite data set they are often biased, inducing a systematic error. In this paper we assess by Monte Carlo simulations the amount of systematic error in the splitting measurement, using artificially generated power spectra. The simulations are carried out for multiplets of degree     2 and 3 with various signal-to-noise ratios, linewidths and observing times. We address the possible use of non-MLE estimators that could provide a smaller or negligible systematic error. The implication for asteroseismology is also discussed.