西南低涡的一些基本事实及初步分析

本文统计了1970～1989年20年西南低涡生成的资料,得到了西南低涡生成的月际和年际分布的其本事实。在此基础上分析了生成低涡的500hPa环流背景和九龙涡生成的地理环境,试图揭示西南涡生成的机理,最后,文章利用非热成风适应原理,对低涡生成机理进行了理论分析。     

海洋模式对1995年5月南海AXBT海温资料的同化分析

利用美国普林斯顿大学海洋模式 (POM )对 1995年 5月南海AXBT海洋观测资料和MISST卫星观测的海面温度进行模式四维同化分析研究。POM模式同化处理使用最优插值分析与逼近方法 ,从模式积分到第 4年开始 ,使用COADS气候平均风应力场、净热通量场强迫海洋 ,从第 5年 5月开始加入 1995年 5月的MISST和AXBT资料同化积分 30d ,然后对同化结果进行分析。结果表明 ,资料同化可以改进模式对温度场的预报 ,也可改进模式对海流的模拟。分析结果还显示 ,1995年 5月南海浅层有暖涡和冷涡形成     

华北地区夏季水资源特征分析及其对气候变化的响应（Ⅰ）:近40年华北地区夏季水资源特征分析

近年来 ,淡水资源匮乏问题已经成为全人类普遍关注的严重问题。华北地区是我国的政治、经济和文化中心 ,该区水资源严重不足 ,缺水问题一直困扰着该区经济的发展和人民生活水平的提高。本文利用NCEP/NCAR的 40a(195 8— 1997年 )再分析资料分析发现 ,近 40a来 ,华北地区年平均水资源略有盈余 ,但水资源年内分布很不均匀 ,一年内仅 6 9月份水资源有盈余 ,其它各月均有不同程度亏损。对华北地区水资源最充足的夏季水资源各分量进行EOF分析 ,结果表明 ,华北地区夏季水资源各分量空间分布的最主要特征都表现为全区一致型 ;最主要的时间变化趋势都表现为自 6 0年代至 80年代中期持续下降 ,80年代中期以后略有回升 ,但回升幅度较小 ,降水仍然低于 40a来的平均水平。对各物理量进行M -K突变检验的结果表明 ,各量在 6 0年代中期前后都发生了一次突变过程 ,但发生时间略有差别     

淮河流域洪涝变化动力系统研究

非线性现象是现代科学最重要的研究课题之一，地理现象无不存在着非线性，目前，地理学对非线性现象的研究多从分形角度开展，对其本质－混沌的研究较少，在先前关于淮河流域洪涝变化序列及其非线性特征研究的基础上，运用混沌理论和微分议程理论，对淮河流域洪涝变化混沌动力系统进行了初步研究，重建了一个3维2次非线性模型，研究表明，淮河流域洪涝变化混沌动力系统具有较Lorenz等典型混沌系统更复杂的形式。     

Feeding ecology of flatfish juveniles (pleuronectiformes) in Mar Chiquita coastal lagoon (Buenos Aires, Argentina)

During a two-year period in Mar Chiquita coastal lagoon (Buenos Aires, Argentina), the diet of juvenileParalichthys orbignyanus andOncopterus darwinii sorted according to size, area, and season was analyzed. Sampling was carried out using a beach seine net in three zones of different salinity. The small size of the individuals collected and the high percentage of full stomachs, corresponding with high food availability in the lagoon, indicated that Mar Chiquita is used as a nursery ground. The diet ofP. orbignyanus consisted of crustaceans and polychaetes, whereasO. darwinii fed predominantly on small crustaceans (gammarid amphipods, harpacticoid copepods, mysids, and ostracods); polychaetes were less important. Diet of both flatfish species varied with length. Smaller individuals were microphagous, while bigger specimens included larger prey in their diet. Prey taxa number increased for both species, seawards to the mouth of the lagoon, probably related to the salinity gradient. Prey diversity forP. orbignyanus was highest in autumn and spring, and the lowest percentage of stomachs with food occurred in winter. Prey diversity was lowest forO. darwinii in spring and the highest percentage of empty stomachs was found in autumn.     

GIS分幅数字化数据的模糊匹配

在GIS数字化输入空间数据时，经常碰到分幡接边时的数据匹配问题，本文提出了一种建立于综合考虑距离和整体结构的模糊数据匹配算法，并给出了应用实例。     

Effect of the time factor on the temperature dependence of electrical resistivity of some rocks

Summary The electrical AC (10³ Hz) resistivity of some rocks was measured in a dry regime in air and in an argon atmosphere under various heating regimes. It is shown that long-term measurements provide us additional information about factors influencing resistivity and activation energy which cannot be obtained by means of standard measurements.     

Some characteristics of quasi-biennial geomagnetic oscillations of solar origin

Summary The monthly means of the geomagnetic index aa and of the relative sunspot numbers were used to analyse the quasi-biennial oscillations (QBO) in geomagnetic activity related to an analogous variation in solar activity. Statistical methods of time series analysis were applied: autocorrelation and power spectrum, cross-correlation, complex demodulation, and the digital filter technique. The aim of this investigation was not only the detection of QBO in geomagnetic activity, which has been done by various authors, but also the determination of the time variation of the period, phase and amplitude of these oscillations in the given time interval with reference to the results of Apostolov [2] and Apostolov and Letfus [4]. The following main results have been obtained: The characteristic period of QBO in geomagnetic activity is shorter and varies in a narrower interval than that in sunspot activity; the lifetime of the oscillations is about 22 years and roughly coincides with Hale's cycle; the amplitude of QBO in geomagnetic activity has its maximum on the decreasing branch of the sunspot cycle; the QBO in geomagnetic activity follows the analogous variation in sunspot activity with a delay of 2 to 6 months for various solar cycles, as compared with the 3 to 22 moths for the raw, unfiltered data. The causes of the differences in the characteristics of QBO in geomagnetic and sunspot activity are considered and it is shown that these differences do not contradict the solar control of this phenomenon.

---

¶rt;uaum u¶rt;a aa u mum ua nm unau ¶rt; aaua au¶rt;mu uu () aumo amumu, a mmmuu auauu amumu. nau ¶rt;u mamumuu m¶rt; aaua ¶rt;: nm mu, amau u u aau, na ¶rt;¶rt;u, ua umau. m u¶rt;au m ¶rt;u uaum amumu, m ¶rt;a au amau, m m n¶rt;u auau nu¶rt;a, a u anum¶rt; mu uu ¶rt; ¶rt;a umaa u, uma mam nma [2] u nma u ma [4]. u n ¶rt;u mam: aamumuu nu¶rt; aum amumu u um uma amumu nm; uu uu nuuum a 22¶rt;a u a u a; anum¶rt;a aum amumu ¶rt;muam aua uu a na¶rt;a mu ua nm; aum amumu ¶rt;m aau auau amumu nm n¶rt;au 2 n 6 a ¶rt; a u n au n¶rt;au 3 n 22 a ¶rt; umua naa ¶rt;a. ¶rt;am nuu auu aamumu aum u amum u naa, m mu auu nmum m m u.

     

Comparative performance of convolutional neural network,weighted and conventional support vector machine and random forest for classifying tree species using hyperspectral and photogrammetric data

ABSTRACT

The classification of tree species can significantly benefit from high spatial and spectral information acquired by unmanned aerial vehicles (UAVs) associated with advanced classification methods. This study investigated the following topics concerning the classification of 16 tree species in two subtropical forest fragments of Southern Brazil: i) the potential integration of UAV-borne hyperspectral images with 3D information derived from their photogrammetric point cloud (PPC); ii) the performance of two machine learning methods (support vector machine – SVM and random forest – RF) when employing different datasets at a pixel and individual tree crown (ITC) levels; iii) the potential of two methods for dealing with the imbalanced sample set problem: a new weighted SVM (wSVM) approach, which attributes different weights to each sample and class, and a deep learning classifier (convolutional neural network – CNN), associated with a previous step to balance the sample set; and finally, iv) the potential of this last classifier for tree species classification as compared to the above mentioned machine learning methods. Results showed that the inclusion of the PPC features to the hyperspectral data provided a great accuracy increase in tree species classification results when conventional machine learning methods were applied, between 13 and 17% depending on the classifier and the study area characteristics. When using the PPC features and the canopy height model (CHM), associated with the majority vote (MV) rule, the SVM, wSVM and RF classifiers reached accuracies similar to the CNN, which outperformed these classifiers for both areas when considering the pixel-based classifications (overall accuracy of 84.4% in Area 1, and 74.95% in Area 2). The CNN was between 22% and 26% more accurate than the SVM and RF when only the hyperspectral bands were employed. The wSVM provided a slight increase in accuracy not only for some lesser represented classes, but also some major classes in Area 2. While conventional machine learning methods are faster, they demonstrated to be less stable to changes in datasets, depending on prior segmentation and hand-engineered features to reach similar accuracies to those attained by the CNN. To date, CNNs have been barely explored for the classification of tree species, and CNN-based classifications in the literature have not dealt with hyperspectral data specifically focusing on tropical environments. This paper thus presents innovative strategies for classifying tree species in subtropical forest areas at a refined legend level, integrating UAV-borne 2D hyperspectral and 3D photogrammetric data and relying on both deep and conventional machine learning approaches.