Geospatial analysis of a coastal sand dune field evolution: Jockey's Ridge, North Carolina

Preservation and effective management of highly dynamic coastal features located in areas under development pressures requires in-depth understanding of their evolution. Modern geospatial technologies such as lidar, real time kinematic GPS, and three-dimensional GIS provide tools for efficient acquisition of high resolution data, geospatial analysis, feature extraction, and quantification of change. These techniques were applied to the Jockey's Ridge, North Carolina, the largest active dune field on the east coast of the United States, with the goal to quantify its deflation and rapid horizontal migration. Digitized contours, photogrammetric, lidar and GPS point data were used to compute a multitemporal elevation model of the dune field capturing its evolution for the period of 1974– 2004. In addition, peak elevation data were available for 1915 and 1953. Analysis revealed possible rapid growth of the dune complex between 1915–1953, followed by a slower rate of deflation that continues today. The main dune peak grew from 20.1 m in 1915 to 41.8 m in 1953 and has since eroded to 21.9 m in 2004. Two of the smaller peaks within the dune complex have recently gained elevation, approaching the current height of the main dune. Steady annual rate of main peak elevation loss since 1953 suggests that increase in the number of visitors after the park was established in 1974 had little effect on the rate of dune deflation. Horizontal dune migration of 3–6 m/yr in southerly direction has carried the sand out of the park boundaries and threatened several houses. As a result, the south dune section was removed and the sand was placed at the northern end of the park to serve as a potential source. Sand fencing has been an effective management strategy for both slowing the dune migration and forcing growth in dune elevation. Understanding the causes of the current movements can point to potential solutions and suggest new perspectives on management of the dune as a tourist attraction and as a recreation site, while preserving its unique geomorphic character and dynamic behavior.     

The dynamic characteristics and migration of a pyramid dune

The results of wind tunnel experiments and field observations show that when the intersection angle between airflow direction and dune crest (ridge) line is > 30°, a reverse vortex is formed. Because of the convergence of sand streams from the windward and lee slopes at the crest, sand accumulates in the crestal region, causing vertical growth. Nevertheless, studies also show that the common asymmetry of the two slopes of a dune may significantly influence the evolution of arms of a pyramid dune. The migration rates of pyramid dunes are mediated by the interplay of their arms moving transversely and the vertical growth in response to the variations in wind regimes. Comparing the effects of airflow transverse to a given arm with longitudinal airflow, it is indicated that the transverse airflow is more significant in controlling the arms of pyramid dunes. The whole body of the studied pyramid dune, particularly the upper quarter section, migrated SE direction during the monitoring period. The patterns of wind erosion and deposition change alternately with seasonal variations in wind directions. The W, NE and SE sides undergo constant erosion, deposition and both erosion and deposition, respectively. The results of long-term monitoring of a pyramid dune show that southerly winds, resulting from a local circulation, markedly affect the transverse migration of the whole pyramid dune.     

Variability in bedform morphology and kinematics with transport stage

Bedform geometry is widely recognized to be a function of transport stage. Bedform aspect ratio (height/length) increases with transport stage, reaches a maximum, then decreases as bedforms washout to a plane bed. Bedform migration rates are also linked to bedform geometry, in so far as smaller bedforms in coarser sediment tend to migrate faster than larger bedforms in finer sediment. However, how bedform morphology (height, length and shape) and kinematics (translation and deformation) change with transport stage and suspension have not been examined. A series of experiments is presented where initial flow depth and grain size were held constant and the transport stage was varied to produce bedload dominated, mixed‐load dominated and suspended‐load dominated conditions. The results show that the commonly observed pattern in bedform aspect ratio occurs because bedform height increases then decreases with transport stage, against a continuously increasing bedform length. Bedform size variability increased with transport stage, leading to less uniform bedform fields at higher transport stage. Total translation‐related and deformation‐related sediment fluxes all increased with transport stage. However, the relative contribution to the total flux changed. At the bedload dominated stage, translation‐related and deformation‐related flux contributed equally to the total flux. As the transport stage increased, the fraction of the total load contributed by translation increased and the fraction contributed by deformation declined because the bedforms got bigger and moved faster. At the suspended‐load dominated transport stage, the deformation flux increased and the translation flux decreased as a fraction of the total load, approaching one and zero, respectively, as bedforms washed out to a plane bed.     

Response of low‐angle dunes to variable flow

Current understanding of bedform dynamics is largely based on field and laboratory observations of bedforms in steady flow environments. There are relatively few investigations of bedforms in flows dominated by unsteadiness associated with rapidly changing flows or tides. As a consequence, the ability to predict bedform response to variable flow is rudimentary. Using high‐resolution multibeam bathymetric data, this study explores the dynamics of a dune field developed by tidally modulated, fluvially dominated flow in the Fraser River Estuary, British Columbia, Canada. The dunes were dominantly low lee angle features characteristic of large, deep river channels. Data were collected over a field ca 1·0 km long and 0·5 km wide through a complete diurnal tidal cycle during the rising limb of the hydrograph immediately prior to peak freshet, yielding the most comprehensive characterization of low‐angle dunes ever reported. The data show that bedform height and lee angle slope respond to variable flow by declining as the tide ebbs, then increasing as the tide rises and the flow velocities decrease. Bedform lengths do not appear to respond to the changes in velocity caused by the tides. Changes in the bedform height and lee angle have a counterclockwise hysteresis with mean flow velocity, indicating that changes in the bedform geometry lag changes in the flow. The data reveal that lee angle slope responds directly to suspended sediment concentration, supporting previous speculation that low‐angle dune morphology is maintained by erosion of the dune stoss and crest at high flow, and deposition of that material in the dune trough.     

What is preserved in the aeolian rock record? A Jurassic Entrada Sandstone case study at the Utah–Arizona border

Aeolian dune fields evolve from protodunes and small dunes into a pattern of progressively fewer, larger and more widely spaced dunes within limits defined by boundary conditions. However, the allogenic boundary conditions that promote aeolian dune‐field development, accumulation of strata and preservation of accumulated strata are not the same. Autogenic processes, such as dune interactions, scour‐depth variation along migrating dunes and substrate cannibalization by growing dunes, result in removal of the stratigraphic record. Moreover, dune‐field events may be collapsed into major erosional bounding surfaces. The question is what stages of evolving dune fields are represented in the rock record? This case study of ca 60 m of Jurassic Entrada Sandstone on the Utah/Arizona border (USA) defines stratigraphic intervals by gross architecture of bounding surfaces and sets of cross‐strata. The interpreted intervals in stratigraphic order consist of: (i) a lower sabkha bed that transitions upward into erosional remnants of small sets representing an initial wet aeolian system; (ii) large, compound cross‐strata representing a mature dune field; (iii) isolated scour‐fill representing negatively climbing dunes that produced ca 25 m of palaeo‐topographic relief; (iv) downlapping sets that fill the landscape‐scale relief; (v) four intervals of stacked climbing sets that each represent short periods of time; and (vi) an upper sabkha bed that again transitions into small sets representing a wet system. Accumulations appear to be associated with sediment pulses, a rising water table, and filling of scoured troughs and landscape‐scale depressions. Preservation of the accumulations is selective and associated with a rising water table, burial and subsidence. The preserved record appears remarkably incomplete. Speculation about missing strata gravitates towards cannibalization of the record of early dune‐field construction, and strata removed during the formation of bounding surfaces. This local Entrada record is thought to represent a point in the spectrum of preservation styles in the rock record.     

控制点影像库在遥感影像几何纠正中的应用

影像几何纠正是遥感图像处理的重要技术。为了克服现有方法在实际应用中的不足,提出利用控制点影像库进行影像自动匹配,然后进行几何变化的方式。这种方法简单有效,快速稳定,匹配得到的同名点数量足够且分布均匀,并通过引入最小二乘匹配使配准达到子像素精度,提高几何纠正的速度与可靠性。     

基于DEM数据的巴丹吉林沙漠沙丘分布规律及其形态参数

沙丘形态特征是风沙地貌研究的重要内容。数字高程模型（DEM）因卓越的三维地形表达能力被广泛应用于沙丘形态研究中。长期以来,受交通条件和研究方法的制约以及高大沙山分布位置的限定,对巴丹吉林沙漠沙丘形态的研究主要集中在沙漠东南部,缺乏对整个沙漠沙丘分布规律及其形态参数特征的系统认知。而对整个沙漠沙丘形态特征的研究,是区域风沙地貌形成与演化研究的重要组成部分。以巴丹吉林沙漠内所有独立沙丘为研究对象,利用研究区DEM数据和一种新的算法,研究了沙漠内两种主要类型沙丘（横向沙丘、星状沙丘）的分布规律及其形态参数特征。结果表明：沙漠内独立沙丘大约有6 033座,高度9~433 m,高度超过350 m的巨型沙山共有53座,其中星状沙山7座,最大高度383 m;横向沙山46座,最大高度433 m。横向沙丘高度与等效沙丘厚度呈正相关的线性函数关系,星状沙丘高度与等效沙丘厚度呈正相关的对数函数关系,表明在现代气候和环境条件下,横向沙丘高度仍在增加,而星状沙丘高度增长已经趋缓。两类沙丘的主轴向大都垂直于研究区主风向NW-SE。横向沙丘起沙风风向单一,沙丘轴向集中在NE-SW和NNE-SSW方向;星状沙丘起沙风风向略为复杂,沙丘存在多个次级轴向。     

雅鲁藏布江河谷加查-米林段沙丘成因

加查-米林段位于雅鲁藏布江河谷中游下段,分布沙丘89片。通过实地考察、遥感卫星影像解译、DEM分析、粒度分析等方法对加查-米林段沙丘成因进行分析。结果表明：依据沙丘沙物质来源可将加查-米林段沙丘划分为河漫滩型沙丘、阶地型沙丘和谷坡型沙丘。河漫滩型沙丘形成与雅鲁藏布江水位关系密切。丰水期洪水淹没河漫滩,泥沙沉积;枯水期水位降低,泥沙出露,经风的吹蚀、搬运、堆积形成河漫滩型沙丘。阶地型沙丘和谷坡型沙丘形成与表层覆盖的黄土层（粉质砂土）破坏有关。雅鲁藏布江不断下切,在阶地及谷坡不同高度残留古河流冲积沙,遭受风蚀后风蚀堆积物形成古沙丘。因环境演变,古沙丘表面沉积黄土,形成致密坚硬的黄土层,不易被风吹蚀,成为古沙丘的保护层。黄土层受风蚀、水蚀及人类活动破坏,导致下伏古沙丘裸露,形成现代流动沙丘。     

Semantic segmentation of high spatial resolution images with deep neural networks

Availability of reliable delineation of urban lands is fundamental to applications such as infrastructure management and urban planning. An accurate semantic segmentation approach can assign each pixel of remotely sensed imagery a reliable ground object class. In this paper, we propose an end-to-end deep learning architecture to perform the pixel-level understanding of high spatial resolution remote sensing images. Both local and global contextual information are considered. The local contexts are learned by the deep residual net, and the multi-scale global contexts are extracted by a pyramid pooling module. These contextual features are concatenated to predict labels for each pixel. In addition, multiple additional losses are proposed to enhance our deep learning network to optimize multi-level features from different resolution images simultaneously. Two public datasets, including Vaihingen and Potsdam datasets, are used to assess the performance of the proposed deep neural network. Comparison with the results from the published state-of-the-art algorithms demonstrates the effectiveness of our approach.     

一种基于复合金字塔模型的三维模型数据组织方法

针对异构三维模型数据的统一组织和集成管理问题,本文以目前主流金字塔模型、三维场景瓦片的数据组织方法和地图服务规范等相关技术为基础,提出了一种基于复合金字塔模型的三维模型数据组织方法。该方法首先构建覆盖测区的地理格网模型,以此作为异构三维模型数据集成的空间框架。其次,建立三维模型数据与地理格网单元的映射关系,从而构建复合金字塔模型,实现异构三维模型数据的统一管理。最后,在数据结构上对现有数据格式进行扩展,设计了一种面向网络传输的数据格式储存三维模型数据,完成异构三维模型数据的一体化存储。试验结果表明,该方法能够实现异构三维模型数据统一组织与集成管理。