地形起伏度最佳分析区域预测模型

地形起伏度指分析区域内最高点和最低点之差,反映宏观区域内地形的起伏特征,是描述地貌形态的定量指标。确定最佳分析区域是地形起伏度提取算法的核心步骤,以及决定地形起伏度提取结果有效性的关键。本文以全国范围内随机选取的78个实验区域、三种不同尺度的DEM数据作为实验对象,分别进行系列分析区域尺度的地形起伏度计算,建立了基于微观地形特征因子的地形起伏度最佳分析区域预测模型。实验表明：相同区域、不同尺度的DEM数据提取的地形起伏度存在差异,DEM尺度相差较小时,地形起伏度的差异也较小;地形起伏度和实验区域的最大高程、区域高差、平均坡度和平均坡度变率等地形特征因子存在强相关关系;当置信水平为0.05时,预测模型拟合参数的准确率达到95%以上,证明预测模型可以有效地确定最佳分析区域的取值范围。     

The political ecology of hegemony in depression-era British Columbia, Canada: Masculinities, work and the production of the forestscape

This article attempts to empirically demonstrate how the struggle for bourgeois hegemony in depression-era British Columbia, Canada, was fought for through the production of new natures. Bringing together Antonio Gramsci’s conceptualization of hegemony with marxist understandings of political ecology, I examine how the legitimacy of particular groups’ dominance over subordinate groups and the survival of specific social relations was built and contested through the (re)making of the material-symbolic landscape. However, I also take seriously Stuart Hall’s argument that we must take note of the multi-dimensional character of hegemony by paying attention to the entanglement of class, gender and ecological relations during the 1930s. In order to demonstrate these arguments I examine the economic, social, moral and ecological crisis that rippled across the socionatural fabric of B.C. during the depression years. I detail how the federal and provincial states responded to the interlaced crises of class, gender and ecological relations through launching a series of public works programs and training programs. These projects were intended to modernize the forestry industry and remake unemployed men in body and soul. In doing so, I demonstrate how ideologies regarding nature come to be both enrolled in the struggle for hegemony and materialized in the making of the forestscape. By weaving theoretical insights through the socionatural history of British Columbia, I demonstrate how a gramscian sensibility pushes us to take seriously the relationality of socionatural processes and the embededdness of concepts in material histories.     

Tectonic vs. gravitational morphostructures in the central Eastern Alps (Italy): Constraints on the recent evolution of the mountain range

Deep-seated gravitational slope deformations (DSGSDs) influence landscape development in tectonically active mountain ranges. Nevertheless, the relationships among tectonics, DSGSDs, and topography are poorly known. In this paper, the distribution of DSGSDs and their relationships with tectonic structures and active processes, surface processes, and topography were investigated at different scales. Over 100 DSGSDs were mapped in a 5000 km² sector of the central Eastern Alps between the Valtellina, Engadine and Venosta valleys. Detailed lineament mapping was carried out by photo-interpretation in a smaller area (about 750 km²) including the upper Valtellina and Val Venosta. Fault populations were also analysed in the field and their mechanisms unravelled, allowing to identify different structural stages, the youngest being consistent with the regional pattern of the ongoing crustal deformation. Finally, four DSGSD examples have been investigated in detail by geological and 2D geomechanical modelling.DSGSDs affect more than 10% of the study area, and mainly cluster in areas where anisotropic fractured rock mass and high local relief occur. Their onset and development is subjected to a strong passive control by mesoscopic and major tectonic features, including regional nappe boundaries as well as NW–SE, N–S and NE–SW trending recent brittle structures. The kinematic consistency between these structures and the pattern of seismicity suggests that active tectonics may force DSGSDs, although field evidence and numerical models indicate slope debuttressing related to deglaciation as a primary triggering mechanism.     

Socio-economic implications based on interviews with fishermen following the Indian Ocean tsunami

Following the devastating tsunami of 26 December 2004 in the Indian Ocean, there was a need to give a voice to the affected population. Hence a survey was conducted in the tsunami-affected regions of India. The tsunami mainly affected the states of Tamil Nadu, Kerala and Andhra Pradesh and the Union Territory of Pondicherry, all in south India, as well as the Andaman and Nicobar Islands of India in the Bay of Bengal. For various logistical reasons, no survey was conducted in the Andaman and Nicobar Islands. The survey was conducted during 21 January to 19 February 2005 and from 1 March to 8 March 2005. A total of eight people, arranged into four teams, simultaneously conducted the survey based upon a prepared questionnaire comprising a total of 16 questions. The total number of villages surveyed was 161, and the overall results of the survey are reported here. Among many observations, capacity building during the construction process, relocation and housing issues and tsunami education and awareness were prominent.     

特征筛选方法在高分辨率遥感影像建筑物提取中的应用

城市建筑物信息的自动提取是城市遥感的关键技术之一,由于阴影、下垫面等多因素干扰,建筑物信息提取精度往往不稳定。本文以Pleiades卫星影像为数据源,通过改进Relief F特征筛选方法,探讨建筑物信息提取精度提高的可行性。首先构建高分辨率遥感影像建筑物基础特征空间,然后利用改进型Relief F算法分析特征对象的权重并筛选出最优特征,最后用监督分类、无特征筛选分类和基于改进型Relief F特征筛选等3种方法分别提取研究区建筑物信息,并结合实地调查数据进行精度验证。结果表明,基于改进型Relief F特征筛选的分类方法提取精度能够达到91.34%,较其他两种方法提取精度分别提高了34.31%和5.62%,且运算速度快,自动识别效率高。     

龙门山中北段流域地貌特征及其构造意义

龙门山中北段位于青藏高原东缘,该区作为高原向东扩展的前缘部位,其地形与河流水系的演化记录了高原隆升与挤压扩展及其气候环境效应的各种信息。龙门山中北段构造活动有明显差异,从中段逆冲为主转化为北段的走滑为主,本文采用定量化地貌参数从构造地貌的角度揭示了区域构造活动的差异。龙门山中北段地貌因子（坡度、地形起伏度和条带状剖面）的阶梯状分布特点,显示了高原扩展的逆冲推覆特征,在中央断裂处构造抬升作用最强,同时显示出了南北向构造活动减弱的趋势,由中段的逆冲转换为北段逆冲兼走滑的形式。北川-映秀断裂两侧流域的HI值也显示了断裂上盘高、下盘低、沿走向减弱的趋势。综合分析认为,本区构造活动是地貌演化的主控因素,龙门山中北段地形存在差异,北川-映秀断裂两侧的小流域地貌指数分析显示,构造抬升活动自南向北减弱,中段以逆冲为主,北段为逆冲兼走滑。     

新疆地震应急救援物资管理信息系统的设计与实现

根据地震应急救援工作的要求,介绍了新疆地震应急救援物资管理信息系统的设计背景,分析了系统的设计思想与目标,并对各功能模块进行了详细的描述,阐述了该系统的实现方案。     

特征优选与卷积神经网络在农作物精细分类中的应用研究

农作物的精细分类一直是农业遥感领域的热点,对农作物估产和种植结构监管有重要意义。深度学习的出现为农作物分类准确性的提升提供了新的思路。本文提出一种特征优选与卷积神经网络（Convolutional Neural Networks, CNN）相结合的多光谱遥感农作物分类方法,用以解决精细分类问题。实验以哨兵2号遥感影像为数据源,基于多光谱遥感影像的波段反射率与包括归一化植被指数在内的10种植被指数,利用Relief F算法进行特征增强与优选,获取最优特征集,从而设计出基于特征优选的CNN分类方法,并对河南省原阳县主要农作物水稻、玉米、花生进行分类识别与制图,分类精度达到96.39%。同时,选用支持向量机、CNN方法分别对研究区农作物进行分类识别。对比分析3种方法的分类结果,发现本文提出的基于最优特征集的CNN农作物分类方法表现最优,CNN方法次之,支持向量机方法表现最差。实验结果表明：① 利用Relief F算法能够对特征贡献度进行排序,完成特征筛选,得到包含24个特征的最优特征子集,训练精度达到99.89%;② 基于最优特征集的CNN方法能够在最大程度上提取高精度差异性特征,实现对农作物的精细分类,且相比CNN和支持向量机的农作物分类方法,本文方法表现更佳。     

南黄海海底地势分区

本文对南黄海海底地势分区进行了研究,结果表明,南黄海是半封闭的内陆浅海,为新生代沉积盆地,深受外力作用的塑造。海底由西部缓倾斜地势区、西南部辐射状地势区、东部朝鲜半岛西岸水下冲刷槽沟地势区和中部浅海平原及黄海海槽地势区组成,形成由东西两侧向中部缓倾斜的基本面貌。     

Relative size of fluvial and glaciated valleys in central Idaho

Quantitative comparisons of the morphometry of glaciated and fluvial valleys in central Idaho were used to investigate the differences in valley relief and width in otherwise similar geologic and geomorphic settings. The local relief, width, and cross-sectional area of valleys were measured using GIS software to extract information from USGS digital elevation models. Hillslope gradients were also measured using GIS software. Power-law relationships for local valley relief, width, and cross-sectional area as a function of drainage area were developed. Local valley relief in glaciated valleys relates to drainage area with a power-law exponent similar to fluvial valleys, but glaciated valleys are deeper for a given drainage area. Local valley width in glaciated valleys is greater than in fluvial valleys, but the exponent of the power-law relationship to drainage area is similar in both valley types. Local valley cross-sectional area in glaciated valleys increases with drainage area with a power-law exponent similar to fluvial valleys, however, glacial valleys have roughly 80% greater cross-sectional area. Steep valley walls in glaciated basins increase the potential for bedrock landsliding relative to fluvial basins. Both the Olympic Mountains of Washington and valleys in central Idaho show relationships in which glaciated valleys are up to 30% deeper than fluvial valleys despite differences in lithology, tectonic setting, and climate.