Dynamical Evolution of Gamma-Ray Burst Remnants with Evolving Radiative Efficiency

In previous works, a generic dynamical model has been suggested by Huang et al., which is shown to be correct for both adiabatic and radiative blast-waves, in both ultra-relativistic and non-relativistic phases. In deriving their equations, Huang et al. have assumed that the radiative efficiency of the fireball is constant. They then applied their model directly to realistic cases where the radiative efficiency evolves with time. In this paper, we abandon the above assumption and re-derive a more accurate dynamical equation for gamma-ray burst remnants. Numerical results show that the model presented by Huang et al. is accurate enough in general cases.     

HYDROCHEMICAL STATE OF INTERMEDIATE CULTURE POND FOR CHINESE PRAWN JUVENILES

This study deals with the hydrochemical state and management measures for the intermediate culturepond for Chinese prawn (Penaeus orientalis) juveniles. The test result showed that the intermediate cultureof Chinese prawn juvenile can be advanced to around March 20 in the Shanghai area if they are culturedin ponds in simple greenhouses of plastic knitting cloth.The density of juveniles introduced was about35 million per hectare, survival was up to 55%.     

地勘单位实行两个转变中的设备管理

在实行“两个转变”中，地勘单位的设备管理要适应市场经济需要，紧密结合地勘单位在开发新产品、新工艺以及生产装备更新的需求，实现管理科学化、现代化、集约化，步入良性循环。     

基于深度学习网络的地震地质灾害识别研究

遥感影像识别方法是破坏性地震震后地质灾害快速、准确获取的重要方法之一,传统的遥感影像识别方法主要以人工目视识别方法和半自动识别方法为主,需投入大量的人力和时间。针对破坏性地震震后地质灾害解译时间长、投入人力多等问题,以2017年8月8日四川九寨沟7.0级地震震后高分辨率无人机遥感影像为研究样本,提出基于深度学习网络的地震地质灾害识别方法。首先结合震后遥感影像解译资料和现场调查资料,提取九寨沟地震地质灾害无人机遥感影像特征,并构建研究区地震地质灾害解译指标和分类数据集;然后采用DeepLabv3+网络结构及softmax损失函数,建立基于深度学习网络的地震地质灾害遥感影像图像语义分割模型方法;最后采用半监督学习方法进行结果验证。研究结果表明,基于深度学习网络的地震地质灾害识别方法可有效识别九寨沟地震地质灾害分布信息,整体分类识别准确率为94.22%,F₁分数值为0.77,结果具有较好的一致性和准确性,可提升地震现场灾情获取和重点地震隐患识别等工作效率及服务能力。     

云南巧家MS5.0地震灾害特点与减灾对策分析

2020年5月18日21时47分云南省巧家县小河镇发生M_S5.0地震,本文结合震区背景信息和现场调查结果,详细阐述了此次地震人员伤亡、房屋震害、地质灾害情况及致灾原因。分析认为,震区易发生小震致灾现象,与当地民居抗震性能较差、次生地质灾害严重、人口密集且居住地多位于山坡及山底等因素有关。为避免小震致灾,应提高震区及周边地区建筑物抗震能力,进行合理规划,迁移高风险区居民点。     

Modeling net primary productivity of the terrestrial ecosystem in China from 1961 to 2005简

Net primary productivity （NPP） is the structure and function of the ecosystem. NPP can most important index that represents the be simulated by dynamic global vegetation models （DGVM）, which are designed to represent vegetation dynamics relative to environ- mental change. This study simulated the NPP of China＇s ecosystems based on the DGVM Integrated Biosphere Simulator （IBIS） with data on climate, soil, and topography. The appli- cability of IBIS in the NPP simulation of China＇s terrestrial ecosystems was verified first. Comparison with other relevant studies indicates that the range and mean value of simula- tions are generally within the limits of observations; the overall pattern and total annual NPP are close to the simulations conducted with other models. The simulations are also close to the NPP estimations based on remote sensing. Validation proved that IBIS can be utilized in the large-scale simulation of NPP in China＇s natural ecosystem. We then simulated NPP with climate change data from 1961 to 2005, when warming was particularly striking. The following are the results of the simulation. （1） Total NPP varied from 3.61 GtC/yr to 4.24 GtC/yr in the past 45 years and exhibited minimal significant linear increase or decrease. （2） Regional differences in the increase or decrease in NPP were large but exhibited an insignificant overall linear trend. NPP declined in most parts of eastern and central China, especially in the Loess Plateau. （3） Similar to the fluctuation law of annual NPP, seasonal NPP also displayed an insignificant increase or decrease; the trend line was within the general level. （4） The re- gional differences in seasonal NPP changes were large. NPP declined in spring, summer, and autumn in the Loess Plateau but increased in most parts of the Tibetan Plateau.     

海风锋在渤海西岸局地暴雨过程中的作用

利用常规观测资料、地面加密自动站资料、多普勒雷达观测资料及中尺度TJ-WRF模式输出资料,对2009年7月6日天津宁河地区出现的暴雨天气过程进行了分析,重点分析了渤海湾海风锋对沿岸局地暴雨的触发机理。结果表明：此次局地暴雨过程是在有利天气背景条件下发生的,暴雨发生地存在局地层结不稳定和较好的水汽条件。海风锋本身有一辐合抬升区,区域内有弱对流存在;当海风锋移到局地存在层结不稳定且水汽充足的区域,其抬升区的辐合上升运动迅速加强,从而触发该地区雷暴的新生发展;海风锋与迎面移来的雷暴相遇,会对雷暴的加速发展起到加强作用。利用中尺度WRF模式输出资料进行分析看到,两条辐合线相交处易激发出强雷暴,雷暴出现在近地层大气暖干区的北端、湿空气的交汇处;海风锋对雷暴的新生发展有明显的触发抬升作用。     

Fluid–rock interaction in the Qitianling granite and associated tin deposits, South China: Evidence from boron and oxygen isotopes

The behavior of boron and its isotopes in fluid-dominant processes and hydrothermal alteration of granites is examined using fresh and altered granite samples from the Qitianling granite and associated hydrothermal tin deposits in South China. Boron concentrations are highest in the fresh granite (37 ppm) and depleted as a result of two stages of fluid mobility and fluid–rock interaction within the granite. Constraints provided both by δ¹¹B and δ¹⁸O data suggest that the first stage was related to exsolution of aqueous fluids from the granite magma at a temperature of > 450 °C. This was followed by further boron depletion in the granite by hydrothermal circulation of meteoric water at lower temperatures (~ 350 °C) and low water/rock ratios. The sensitivity of coupled boron and oxygen isotope systematics to these processes suggests that they can provide valuable constraints on of fluid mobility in granite and associated mineralization.     

基于BP人工神经网络的鹰潭市PM2.5和PM10浓度预测模型

利用2015—2019年鹰潭市5个大气成分监测站数据和气象站地面观测数据,运用主成分分析法,提取气象要素、气体污染物对PM2.5和PM10浓度影响的主要成分,调整BP人工神经网络的隐藏层个数和隐藏层节点数,构建基于BP人工神经网络的鹰潭市PM2.5和PM10浓度预测模型。结果表明：1） 气象要素中,共提取3个影响PM2.5、PM10浓度的主成分,分别为相对湿度、降水,气温、气压和风速,其中湿度、气温、风速与PM2.5、PM10浓度显著相关。2） 气体污染物中,共提取2个主成分,分别为SO2、NO2和O3,其中NO2、SO2与PM2.5、PM10浓度显著相关。3） 所建立的PM2.5、PM10浓度逐小时预测模型在20 h内预测性能良好,预测准确率分别为88%、86%,逐日预测模型在5 d内的预测性能良好,预测准确率分别为94%、92%,准确率较高,具有良好的预报性能。