CAJ | 학술논문

第一次全国地理国情普查工作的一项重要内容就是调查地形地貌。针对在青藏高原等艰险地区开展相应工作仍存在诸多困难，该文基于美国“奋进号”航天飞机获取的 SRTM-DEM 数据，运用 GIS 技术提取了青藏高原的8个地貌因子，通过相关性分析最终选取地表粗糙度、高程、地势起伏度、坡度变率及坡向变率5个相关性相对较小的因子，开展宏观地貌类型划分及地貌分布特征分析工作，并基于一种简易模型提取并讨论了对青藏高原生态环境有重要影响的冰缘（冻土）地貌。研究表明：应用 SRTM-DEM 数据可快速有效地提取青藏高原等艰险地区的地貌因子，但某些因子之间具有极高的相关性，应用中需慎重选择。此外，通过 SRTM-DEM 数据提取地貌因子以进行宏观地貌类型划分及地貌特征分析的方法简单易行，能够为在青藏高原等艰险地区开展地理国情普查工作提供可靠的基础资料。
제일차전국지리국정보사공작적일항중요내용취시조사지형지모。침대재청장고원등간험지구개전상응공작잉존재제다곤난，해문기우미국“강진호”항천비궤획취적 SRTM-DEM 수거，운용 GIS 기술제취료청장고원적8개지모인자，통과상관성분석최종선취지표조조도、고정、지세기복도、파도변솔급파향변솔5개상관성상대교소적인자，개전굉관지모류형화분급지모분포특정분석공작，병기우일충간역모형제취병토론료대청장고원생태배경유중요영향적빙연（동토）지모。연구표명：응용 SRTM-DEM 수거가쾌속유효지제취청장고원등간험지구적지모인자，단모사인자지간구유겁고적상관성，응용중수신중선택。차외，통과 SRTM-DEM 수거제취지모인자이진행굉관지모류형화분급지모특정분석적방법간단역행，능구위재청장고원등간험지구개전지리국정보사공작제공가고적기출자료。
An important part of the first national geographic condition survey is to survey landform,but there are many difficulties to carry out the work in dangerous areas such as the Tibetan plateau.Based on SRTM3-DEM data which were acquired by U.S.Endeavour space shuttle in 2000,we used GIS technology to extract eight landform factors of the Tibetan plateau,and by calculating the correlation matrix,five factors which are less correlated were selected finally.On the basis of these factors,the landform of the Tibetan plateau was divided into different types,and then the distribution of them was analyzed.Furthermore,using a simple model we extracted the periglacial (permafrost)landforms which have made a major impact on the ecological environment of the Tibetan plateau.The result shows that using SRTM3-DEM data can extract the landscape factors quickly and efficiently in dangerous areas such as the Tibetan plateau,but we should face a problem that some factors are highly correlated.This paper also indicates that macro landform can be divided and analyzed effectively by SRTM3-DEM,and the results will provide reliable information for national geographic condition survey in dangerous areas.In addition, although the map of periglacial landforms that we got by our method has some errors,but the algorithm of the model can still give some references for further study.