Spatial and temporal variations in pH and total alkalinity at the beginning of the rainy season in the Changjiang Estuary, China

The results of field observation carried out in May 2003 were used to examine pH and total alkalinity behaviors in the Changjiang Estuary. It was showed that PH and total alkalinity took on clear spatial variations in values with the minima in the low salinity region. Like salinity, transect distributions of PH and total alkalinity (TA) in a downriver direction had a sharp gradient each. These gradients appeared in such a sequence that the TA gradient was earlier than salinity and PH gradients, and the salinity gradient was earlier than the PH gradient. These distribution characteristics seemed to be strongly influenced by the mixing process of freshwater and seawater, for both PH and total alkalinity had significant linear relationships with salinity and temperature. For PH, phytoplankton activities also had a significant impact upon its spatial distribution. During a period of 48 h, PH and total alkalinity changed within wide ranges for every layer of the two anchor stations, namely, Stas 13 and 20, which were located at the mixed water mass and seawater mass, respectively. For both Stas 13 and 20, PH and TA fluctuation of every layer could be very wide during a 4 h period. As a whole, the data of the two anchor stations showed that neither variations in salinity and temperature nor phytoplankton activities were the main factors strongly influencing the total alkalinity temporal variability on a small time scale. The data of Sta. 20 implied that both salinity variation and phytoplankton activities had a significant influence on PH temporal variability, but the same conclusion could not be drawn from the data of Sta. 13.

Viewing DVM via general behaviors of zooplankton: A way bridging the success of individual and population

In this paper, we viewed the diel vertical migration (DVM) of copepod in the context of the animal's immediate behaviors of everyday concerns and constructed an instantaneous behavioral criterion effective for DVM and non-DVM behaviors. This criterion employed the function of 'venturous revenue' (VR), which is the product of the food intake and probability of the survival, to evaluate the gains and losses of the behaviors that the copepod could trade-off. The optimal behaviors are to find the optimal habitats to maximize VR. Two types of VRs are formulated and tested by the theoretical analysis and simulations. The sensed VR, monitoring the real-time changes of trade-offs and thereby determining the optimum habitat, is validated to be the effective objective function for the optimization of the behavior; whereas, the realized VR, quantifying the actual profit obtained by an optimal copepod in the sensed-VR-determined habitat, defines the life history of a specific age cohort. The achievement of a robust copepod overwintering stock through integrating the dynamics of the constituent age cohorts subjected to the instantaneous behavioral criterion for DVM clearly exemplified a possible way bridging the immediate pursuit of an individual and the end success of the population. (c) 2005 Published by Elsevier Ltd.

Molecular phylogeny of some genera of Pamphagidae (Acridoidea, Orthoptera) from China based on mitochondrial 16S rDNA sequences

Based on the mitochondrial 16S ribosomal DNA partial sequences (473 bp) of 9 species of Pamphagidae (Acridoidea, Orthoptera) from China and of 4 species of Pamphagidae and 2 species of Pyrgomorphidae and Acrididae (as outgroups) retrieved from GenBank, we constructed the molecular phylogeny using the Neighbor Joining (NJ) and Minimum Evolution ( ME) methods based on the nucleotide Kimura 2-parameter model. The results of our study shown that: 1) the ranges of the 16S rDNA nucleotide divergence between two species of a genus were 0.21%, among genera of a subfamily were 0.42-3.38%, and among subfamilies of Pamphagidae were 1.90-8.88%, respectively. The phylogenetic tree shows that: 1) all Pamphagidae taxa form a monophyletic clade, and are well separated from the outgroup; 2) the African taxa Porthetinae (Lobosceliana brevicornis) and Akicerinae (Batrachotetrix sp.) are distinctly separated from the Chinese taxa Prionotropisinae; 3) Haplotropis bruneriana and Glauia terrea of Pamphaginae are nested in the middle of the tree, but their phylogenetic status is uncertain in this study; 4) 8 genera of Asiotmethis, Beybienkia, Mongolotmethis, Sinotmethis, Rhinotmethis, Filchnerella, Eotmethis and Pseudotmethis from China are all grouped into the subfamily Prionotropisinae, but their phylogenetic relationships are not clearly resolved.

利用BP神经网络实现三维物体姿态的测定

本文利用BP(Back-Propagation)人工神经网络对三维物体的姿态测定进行了研究。姿态测定一直缺少通用而实际的方法,人工神经网络由于具有强大的自组织、自适应学习能力,迅速的并行信息处理能力,可望解决这个问题。但现有BP算法存在训练慢和易陷入局部最小两个问题.本文提出的级联形式网络结构,使BP网络的训练速度大为提高,陷入局部最小的可能性大为降低。利用这种级联结构对飞机模型姿态测定,取得了较好的实验结果。

基于粗糙集-BP神经网络的发动机故障诊断

由于发动机光谱分析监控数据中磨损微粒种类过多,如果将这些微粒信息直接作为神经网络的输入,则存在输入层神经元过多、网络结构复杂等诸多问题。本文将粗糙集引入到发动机故障诊断中来,利用粗糙集在属性约简方面的优势,删除冗余磨损微粒,提取出重要磨损微粒,并将其作为BP神经网络的输入,建立发动机故障诊断模型。该方法降低输入层的神经元个数,简化了网络结构,缩短网络训练时间,并且由于剔除了冗余磨损微粒,减少了由该部分微粒信息不准确而带来的误差,有效提高了故障诊断的精确度。最后通过算例分析验证了相关算法和诊断模型的准确性和有效性。

基于改进的模糊C均值的BP分类器设计

提出了一种基于改进模糊C均值的BP神经网络分类器的设计,通过改进的模糊C均值算法对大量的数据进行聚类划分,然后设计BP神经网络对划分后的数据进行训练和测试,最后由计算机进行综合判断.试验证明该分类器是有效的,可以对高速公路车辆的车型进行迅速判别.

现代动物食谱及吕梁山北端全新世风成沉积研究

The carbon isotope compositions of plants in Xilinguole, Haibei and Jilong and the tooth enamel of animals in Jilong have been analyzed. The study of carbon isotope discrimination of plants along altitude gradient among the three areas has found decreased discrimination.The eolian deposits of Heyeping can be used to reconstruct Holocene climate variability in the region. Some proxies indicates that the 8. 2 ka event can be found in north China, and it is warm and arid in the 4900～6800(or 6900) cal. a BP.

岱海岩芯沉积的粒度组成与湖区全新世降水变化历史

Sediment cores DH99a and DH99b recovered in the central part of Daihai Lake in north-central China were analysed at 2- to 4-crn intervals for grain-size distribution. Grain-size distributions of the lake sediments are inferred to be a proxy for past changes in East Asian monsoon precipitation, such that greater silt-size percentage and higher median grain size reflect increased monsoonal precipitation rates. The grain-size record of Daihai Lake sediments spanning the last ca 11,000 yr indicates that the monsoonal precipitation in the lake region can be divided into three stages: the Early, Middle and Late Holocene. During the Early Holocene before ca 7900 cal yr BP, the median grain size (Md) and the silt-fraction content were relatively low and constant, suggesting relatively low precipitation over the lake region. The Middle Holocene between ca 7900 and 3100 cal yr BP was marked by intensified and highly variable monsoonal precipitation, as indicated by high and variable Md values and silt contents of the lake sediments. During this period, average precipitation rate gradually increased from ca 7900 to 6900 cal yr BP, displayed intense oscillations between ca 6900 and 4400 cal yr BP, and exhibited a decreasing trend while fluctuating from ca 4400 to 3100 cal yr BP Although generally high during the Middle Holocene, both the Md and the silt content assumed distinctly low values at the short intervals of ca 6500-6400, 6000-5900, 5700-5600, 4400-4200 cal yr BP, implying that monsoonal precipitation might have been significantly reduced during these intervals. During the Late Holocene since ca 3100 cal yr BP, grain-size values suggest that precipitation decreased. However, during the Late Holocene, relatively higher Md values and silt contents occurring between ca 1700 to 1000 cal yr BP may denote an intensification of hydrological cycles in the lake area. Changes in the East Asian monsoonal precipitation were not only directly linked with the changing seasonality of solar insolation resulting from progressive changes in the Earth's orbital parameters, but also may have been closely related to variations in the temperature and size of the Western Pacific Warm Pool, in the intensity of the El Nino-Southern Oscillation, and in the path and strength of the North Equatorial Current in the western Pacific.

The ion-molecule reaction after multiphoton ionization in the binary cluster of ammonia and methanol

The binary cluster (CH3OH)(n)(NH3)(m) was studied by using a multiphoton ionization time-of-flight mass spectrometer (MPI-TOFMS). The measured two series of protonated cluster ions: (CH3OH)(n)H+ and (CH3OH)(n)NH4+ (1 less than or equal to n less than or equal to 14) were attributed to the ion-molecule reaction in the binary cluster ions. The mixed cluster of CH3OD with ammonia was also studied. The results implied that the proton transfer probability from the OD group was larger than that from CH3 group. The ab initio calculation of the binary cluster was carried out at the HF/STO-3G and MP2/6-31G** levels of theory, and indicated that the latter process of the proton transfer must overcome a barrier of similar to 29 kcal/mol. (C) 1999 Elsevier Science B.V. All rights reserved.

Especiação iônica da solução percolada de um solo tratado com lama de cal.

2005

Crescimento de Eucalyptus benthamii submetido à aplicação de lama de cal e cinza de madeira.

2015

Kennedy, Johnson and NATO: Britain, America and the Dynamics of Alliance, 1962-68

Priest, Andrew, Kennedy, Johnson and NATO: Britain, America and the Dynamics of Alliance, 1962-68 (New York: Routledge, 2006), wpp.xiv+222 RAE2008

The church in Corea by Mark Napier Trollope, Bp. of Korea

http://www.archive.org/details/churchincorea00troluoft

Fuzzy ARTMAP, Slow Learning and Probability Estimation

A nonparametric probability estimation procedure using the fuzzy ARTMAP neural network is here described. Because the procedure does not make a priori assumptions about underlying probability distributions, it yields accurate estimates on a wide variety of prediction tasks. Fuzzy ARTMAP is used to perform probability estimation in two different modes. In a 'slow-learning' mode, input-output associations change slowly, with the strength of each association computing a conditional probability estimate. In 'max-nodes' mode, a fixed number of categories are coded during an initial fast learning interval, and weights are then tuned by slow learning. Simulations illustrate system performance on tasks in which various numbers of clusters in the set of input vectors mapped to a given class.