Automatic Determination of Stellar Atmospheric Parameters Using Neural Networks and Instance-Based Machine Learning

In this article we show how machine learning methods can beeffectively applied to the problem of automatically predictingstellar atmospheric parameters from spectral information, a veryimportant problem in stellar astronomy. We apply feedforwardneural networks, Kohonen's self-organizing maps andlocally-weighted regression to predict the stellar atmosphericparameters effective temperature, surface gravity and metallicityfrom spectral indices. Our experimental results show that thethree methods are capable of predicting the parameters with verygood accuracy. Locally weighted regression gives slightly betterresults than the other methods using the original dataset asinput, while self-organizing maps outperform the other methods when significant amounts of noise are added. We also implemented a heterogeneous ensemble of predictors, combining the results given by the three algorithms. This ensemble yields better results than any of the three algorithms alone, using both the original and the noisy data.     

Comparison of four meteorite penetration funnels in the Campo del Cielo crater field,Argentina

Abstract– More craters may be discovered in the future, but as it is currently known, the Campo del Cielo crater field is 18 km long by 4 km at its widest point. Such a distribution of craters suggests that the parent meteoroid entered and traversed the atmosphere at a very low angle relative to horizontal. The crater field contains at least 20 small craters produced by the larger fragments of the parent meteoroid. Four of these are explosion analog craters and the rest are penetration funnels. During four field seasons, we have constructed topographic and magnetic maps of four of the penetration funnels as found, and then dug trenches across them to learn their original structures and recover meteorites preserved within them. Structures of these penetration funnels indicate very low angles of impact, i.e., 9–16° relative to horizontal. This supports the idea that the parent meteoroid traversed the atmosphere at a low angle. Data given here for the four penetration funnels include projectile masses, lengths, widths, depths, and estimates of impact angles and azimuths. One of the penetration funnels described here (No. 6) can almost be classified as an explosion analog crater.     

Expression of putative zinc-finger protein lcn61 gene in lymphocystis disease virus China （LCDV-cn） genome

An open reading frame (lcn61) of lymphocystis disease virus China (LCDV-cn), probably responsible for encoding putative zinc-finger proteins was amplified and inserted into pET24a (+) vector. Then it expressed in E. coli BL21 (DE3), and His-tag fusion protein of high yield was obtained. It was found that the fusion protein existed in E. coli mainly as inclusion bodies. The bioinformatics analysis indicates that LCN61 is C2H2 type zinc-finger protein containing four C2H2 zinc-finger motifs. This work provides a theory for functional research of lcn61 gene. Supported by High Technology Research and Development Program of China (863 Program, No. 2006AA100309)     

3D Step-by-step inversion strategy for audio magnetotellurics data based on unstructured mesh

A three-dimensional (3D) step-by-step inversion strategy for audio magnetotellurics(AMT) is investigated in this study.The objective function is minimized by it...     

Current and thermohaline characteristics of the Arabian Sea during January 1998

Based on a ship survey during January 1998, the characteristics of the flow, the thermohaline properties and the volume transport of the Arabian Sea are discussed. A strong westward flow exists between 10.5?N and 11?N, part of which turns to the south as the Somali current near the coast at about 10?N and the rest turns north. At the passage between the African continent and the So- cotra Island, the northern branch separates into two flows: the left one enters the passage and the right one flows eastward along the southern slope of the island. Off the island the flow separates once more, most of it meandering northeast and a small fraction flow- ing southeast. Volume transport calculation suggests that the tidal transport is one or two orders of magnitude smaller than the total transport in this region and it becomes more important near the coast. The average velocity of the flow in the upper layer (0-150 m) is about 20 cm s-1, with a maximum of 53 cm s-1 appearing east of the Socotra Island, and the subsurface layer (200-800 m) has an aver- age velocity of 8.6 cm s-1; the velocity becomes smaller at greater depths. The depth of the seasonal thermocline is about 100 m, above which there is a layer with well mixed temperature and dissolved oxygen. High-salinity and oxygen-rich water appears near the surface of the northern Arabian Sea; a salinity maximum and oxygen minimum at 100 m depth along 8?N testifies the subduction of surface water from the northern Arabian Sea. Waters from the Red Sea and the Persian Gulf also influence the salinity of the area.     

VARIATIONAL ITERATION SOLVING METHOD FOR SEA-AIR OSCILLATOR MODEL OF INTERDECADAL CLIMATE FLUCTUATIONS

1IN T R O D U C T IO NInteractionsbetween the ocean and atmosphere con-tributeto climatefluctuationosver a broad spectrum oftime scales.Studiesof those interactionhsave farfo-cused on El Ni o-SouthernOscillation(ENSO)phe-nomenon thathas a period of3to4yea…     

A search for periodicities from a ULX in the LINER galaxy NGC 4736

We report our findings on a new quasi-periodic oscillation (QPO) and a long period from the ultraluminous X-ray source (ULX) X-2 in nearby galaxy NGC 4736 based on the Chandra and XMM-Newton archival data. To examine the timing properties, power density spectra of the source have been obtained using Fast Fourier Transform. Also the spectral parameters of the source have been calculated by obtaining and fitting the energy spectra. Power density spectrum of this source reveals a QPO peak at \(0.73_{-0.14}^{+0.16}~\mathrm{mHz}\) with an fractional rms variability of 16 % using the Chandra data (in the year 2000-lower state of the source). The XMM-Newton data analysis indicates a peak at \(0.53_{-0.35}^{+0.09}~\mathrm{mHz}\) with a fractional rms variation of 5 % (in the year 2006-higher state of the source). These recovered QPOs overlap within errors and may be the same oscillation. In addition, we detect a long periodicity or a QPO in the Chandra data of about (5.2±2.0)×10^?5 Hz (～5.4 h) over 3 σ confidence level. If this is a QPO, it is the lowest QPO detected from a ULX. The mass of the compact object in ULX X-2 is estimated using the Eddington luminosity and a disk blackbody model in the range (10–80) M _⊙.     

A proposal for the theoretical analysis of the interactive coupled effects between urbanization and the eco-environment in mega-urban agglomerations

Mega-urban agglomerations are strategic core areas for national economic development and the main regions of new urbanization. They also have important roles in shifting the global economic center of gravity to China. However, the development of mega-urban agglomerations has triggered the interactive coercion between resources and the eco-environment. The interactive coupled effects between urbanization and the eco-environment in mega-urban agglomerations represent frontier and high-priority research topics in the field of Earth system science over the next decade. In this paper, we carried out systematic theoretical analysis of the interactive coupling mechanisms and coercing effects between urbanization and the eco-environment in mega-urban agglomerations. In detail, we analyzed the nonlinear-coupled relationships and the coupling characteristics between natural and human elements in mega-urban agglomerations. We also investigated the interactive coercion intensities between internal and external elements, and the mechanisms and patterns of local couplings and telecouplings in mega-urban agglomeration systems, which are affected by key internal and external control elements. In addition, we proposed the interactive coupling theory on urbanization and the eco-environment in mega-urban agglomerations. Furthermore, we established a spatiotemporal dynamic coupling model with multi-element, multi-scale, multi-scenario, multi-module and multi-agent integrations, which can be used to develop an intelligent decision support system for sustainable development of mega-urban agglomerations. In general, our research may provide theoretical guidance and method support to solve problems related to mega-urban agglomerations and maintain their sustainable development.