The luminescence dating chronology of a deep core from Bosten Lake (NW China) in arid central Asia reveals lake evolution over the last 220 ka

The lacustrine deposits of lakes in arid central Asia (ACA) potentially record palaeoclimatic changes on orbital and suborbital time scales, but such changes are still poorly understood due to the lack of reliable chronologies. Bosten Lake, the largest freshwater inland lake in China, is located in the southern Tianshan Mountains in central ACA. A 51.6‐m‐deep lacustrine succession was retrieved from the lake and 30 samples from the succession were used for luminescence dating to establish a chronology based on multi‐grain quartz OSL and K‐feldspar post‐IR IRSL (pIRIR₂₉₀) dating. Quartz OSL ages were only used for samples from the upper part of the core. The K‐feldspar luminescence characteristics (dose recovery test, anomalous fading test, first IR stimulation temperature plateau test) are satisfactory and from the relationship amongst the quartz OSL, IR₅₀ and pIRIR₂₉₀ doses we infer that the feldspar signals are likely to be well bleached at deposition. Bacon age‐depth modelling was used to derive a chronology spanning the last c. 220 ka. The chronology, lithology and grain‐size proxy record indicate that Bosten Lake formed at least c. 220 ka ago and that lake levels fluctuated frequently thereafter. A stable deep lake occurred at c. 220, 210–180, c. 165, 70–60, 40–30 and 20–5 ka, while shallow levels occurred at c. 215, 180–165, 100–70, 60–40 and 30–20 ka. Bosten Lake levels decreased by at least ~29 m and possibly the lake even dried up between c. 160 and c. 100 ka. We suggest that the water‐level fluctuations in the lakes of ACA may not respond directly to climatic changes and may be affected by a number of complex factors.     

基于软硬协作决策的半监督珊瑚礁底质分类方法

珊瑚礁底质分类对海洋资源开发和海洋生态环境保护起到至关重要的作用。目前,深度学习语义分割方法在遥感图像分类领域应用广泛,但在底质分类方面的研究较少。由于基于全监督深度学习的方法中逐像素标注标签的成本较高,不适用于大规模、高频次的底质分类工作,基于半监督的深度学习方法能够有效利用已标注标签为无标签数据产生伪标签,从而有效降低人工成本,然而现有半监督方法的性能易受伪标签噪声的干扰。针对以上问题,本文提出了一种基于软硬协作决策的半监督底质分类方法。首先,利用多模型联合决策生成高质量的伪标签;然后,提出了一种能够顾及伪标签像素置信度的损失函数来指导模型进行训练;最后,采用软硬协作的决策方式得到精确的底质分类结果。在美属维尔京群岛圣克罗伊岛北部的巴克岛礁和夏威夷群岛的中途岛东南约400 km处的珍珠与爱马仕环礁的浅层底栖生物栖息地地图数据集上评估了本文方法的精度,实验结果表明,本文提出的方法与全监督学习方法精度相当,比主流的语义分割方法精度平均高3.08%,能够有效服务于珊瑚礁底质调查工作。     

香花石包头矿黄河矿及蓟县矿的拉曼光谱特征

首次测定了中国矿物学家发现的矿物———香花石、包头矿、黄河矿及蓟县矿的拉曼光谱图。经对仪器准确性和精密度的测定表明,测定拉曼峰位的最大偏差为±１ｃｍ－１。     

湖南夏季雨量场的EOF的稳定性及其长期预报

利用1958～1995年湖南夏季雨量场,分析了经验正交函数(EOF),结果表明,它能反映降水的空间分布特征.将该场划分为1958～1983年和1970～1995年两段,对原始场和两个新场的EOF进行了比较和讨论,发现在一定条件下,EOF是稳定的.将该场序列延伸至1997年,并划分为1958～1977年和1978～1997年两个不重合的时段;结果EOF仍较稳定,这可能与场内平均相关状况的差异较小有关.在此基础上,采用主成份筛选的建模方案,制作了雨量场的长期预报.     

微波诱导等离子体原子发射光谱法测定铍铬钼钒锆锗的研究

研究了用微波诱导等离子体原子发射光谱法(MIP-AES)测定Be、Cr、Mo、V、Zr和Ge。确定了最佳实验条件,其检出限分别为:0.038,0.030,0.015,0.030,0.024和0.11μg/ml。对实际样品进行了测定,结果令人满意。     

Statistics of underwater ambient noise at high sea states arisen from typhoon out zones in the Philippine Sea and South China Sea

Oceanic noise is the background interference in sonar performance prediction and evaluation at high sea states. Statistics of underwater ambient noise during Typhoons Soulik and Nida were analyzed on the basis of experimental measurements conducted in a deep area of the Philippine Sea and the South China Sea. Generated linear regression, frequency correlation matrix (FCM), Burr distribution and Gumbel distribution were described for the analysis of correlation with environmental parameters including wind speed (WS), significant wave height (SWH), and the inter-frequency relationship and probability density function of noise levels (NLs). When the typhoons were quite close to the receivers, the increment of NLs exceeded 10 dB. Whilst ambient noise was completely dominated by wind agitation, NLs were proportional to the cubic and quintic functions of WS and SWH, respectively. The fitted results between NLs and oceanic parameters were different for “before typhoon” and “after typhoon”. The fitted slopes of linear regression showed a linear relationship with the logarithm of frequency. The average observed typhoon-generated NLs were 5 dB lower than the Wenz curve at the same wind force due to the insufficiently developed sea state or the delay between NLs and WS. The cross-correlation coefficient of FCM, which can be utilized in the identification of noise sources in different bands, exceeded 0.8 at frequencies higher than 250 Hz. Furthermore, standard deviation increased with frequency. The kurtosis was equal to 3 at >400 Hz approximately. The characteristics of NLs showed good agreement with the results of FCM.     

地震数据流服务器的优化与改进

为提高测震数据流服务器的性能及安全性,提出了采用LVS服务器集群技术构建测震数据流服务器集群的方案,通过选用恰当的IP负载均衡技术及连接调度技术,较好的克服了单台服务器性能瓶颈问题,同时增强了地震数据流服务器系统的容错性及可靠性。     

A combined image matching method for Chinese optical satellite imagery

Image matching is one of the key technologies for digital Earth. This paper presents a combined image matching method for Chinese satellite images. This method includes the following four steps: (1) a modified Wallis-type filter is proposed to determine parameters adaptively while avoiding over-enhancement; (2) a mismatch detection procedure based on a global-local strategy is introduced to remove outliers generated by the Scale-invariant feature transform algorithm, and geometric orientation with bundle block adjustment is employed to compensate for the systematic errors of the position and attitude observations; (3) we design a novel similarity measure (distance, angle and the Normalized Cross-Correlation similarities, DANCC) which considers geometric similarity and textural similarity; and (4) we introduce a hierarchical matching strategy to refine the matching result level by level. Four typical image pairs acquired from Mapping Satellite-1, ZY-1 02C, ZY-3 and GeoEye-1, respectively, are used for experimental analysis. A comparison with the two current main matching algorithms for satellite imagery confirms that the proposed method is capable of producing reliable and accurate matching results on different terrains from not only Chinese satellite images, but also foreign satellite images.     

2012年12月20-23日巴州寒潮降雪天气的特征分析

利用常规观测资料、自动站、T639客观分析资料,结合天气学原理和天气动力学诊断分析方法,对2012年12月20-23日巴音郭楞蒙古自治州（以下简称巴州）的寒潮天气过程的成因和降雪天气特征进行分析,并给出巴州降雪天气概念模型。结果表明：这次寒潮降雪过程是北欧阻塞高压向东南衰退,导致脊前横槽转成竖槽东南移,高空冷空气南压至40°N附近,造成巴州强降温、强降雪等寒潮天气,降雪阶段关键影响系统是700 hPa上的冷暖切变和风场辐合,水汽主要源自中低层西北方的经向和低层东南方的纬向输送;后续的冷凝降雪和阴雪阶段天气局地特征明显,水汽源自850 hPa至近地层局地的垂直输送。     

青藏高原冬春积雪影响亚洲夏季风的研究进展

青藏高原积雪作为亚洲夏季降水异常的一个预测因子已有很长历史。统计分析表明,青藏高原冬春积雪与印度夏季降水反相关,与我国长江中下游夏季降水多为正相关,不同作者对此现象给出了一些可能的解释,如地表反照率效应和积雪一水文效应。但积雪的这种跨季节滞后影响的具体物理过程仍有争议。另一方面,青藏高原冬春积雪资料存在不确定性,这对相关物理机制的理解造成了很大困难。通过回顾和梳理前人在该领域的研究成果,提出了高原积雪影响亚洲季风的未来研究方向。