位置:成果数据库 > 期刊 > 期刊详情页
Automatic recognition of loess landforms using Random Forest method
  • 时间:0
  • 分类:P931[天文地球—自然地理学]
  • 作者机构:[1]南京师范大学虚拟地理环境教育部重点实验室,南京210023, [2]江苏省地理信息资源开发与利用协蚓创新中心,南京210023, [3]江苏省地理环境演化国家重点实验室培育建设点,南京210023
  • 相关基金:国家自然科学基金项目(41671389,41601411)
中文摘要:

地貌的发育在特定的条件下往往呈现空间分布上由"新"至"老"的过渡,据此,对地貌类型与特征在空间上的序列采样,即可为研究某种地貌的个体发育提供基本依据。该方法即为地貌学研究中的空代时。本文首先介绍了空代时方法产生的背景与基本概念。分析了近年来空代时方法应用于河流地貌、构造地貌、河口海岸地貌等不同地貌类型演化过程的研究进展。在此基础上,明确了空代时方法在地貌学研究中的适用条件、影响因素及分类体系,并提出了地貌学空代时的研究范式。本文认为今后的研究工作一方面应充分利用海量的地理空间数据,运用空代时方法研究多种空间尺度下的地貌演化的问题;另一方面,应结合现有的物理机制和统计规律,构建集形、数、理一体化的地貌演化模型。

英文摘要:

Geomorphic evolution often presents a spatial pattern of a "young to old" distribution under certain natural environment condition, whereby sampling the geomorphic types and characteristics in spatial sequence can provide some evidence for the evolution of the individual geomorphologic object. This so- called space- for- time substitution has been a methodology in geomorphology research. This paper firstly introduced the basic concepts and background of space- for- time substitution, then a full review has been conducted of recent research progress in geomorphic evolution based on space-for-time substitution, such as fluvial landform, structural landform, estuarine landform and coastal landform. Finally, the explicit terms like suitable conditions, influencing factors and classifications have been summarized so that the research paradigm of space-for- time substitution was proposed. We argued that in the future, the researchers should focus on a full use of massive geographic data for geomorphic evolution research at multi- spatial scales, as well as an effective combination with physical mechanisms and statistical laws for a comprehensive geomorphic evolution modelling.

同期刊论文项目
同项目期刊论文