Estimation and evaluation of drainage basins with landslides using decision tree technique and ASTER data

DOI Web Site 4 Citations 3 References

Bibliographic Information

Other Title
  • Ｄｅｃｉｓｉｏｎ　ｔｒｅｅによる地すべり発生流域の推定とその検証―ＡＳＴＥＲデータを用いて―
  • Decision tree ニ ヨル ジスベリ ハッセイ リュウイキ ノ スイテイ ト ソノ ケンショウ ASTER データ オ モチイテ

Search this article

Abstract

In this study, drainage basins with landslides in the Akaishi Mountains are estimated using the decision tree technique which is a kind of data mining technique. Decision tree technique is characterized by the estimation processes which are explicitly and quantitatively represented by Tree-structures. The purpose of this study is to clarify the relationship between landslide and topography/lithology using Tree-structure. To estimate drainage basins with landslides, topographic characteristics (elevation, slope, profile curvature, plan curvature, dissection and undissection) and geological data were used as explanatory variables. Topographic characteristics were calculated from the DEMs (Digital Elevation Models) of EOS-Terra/ASTER. Objective variable is landslide area data.<br>Correctness of the estimation is 71.2%. The estimation processes quantitatively indicate that the lithology, plan curvature and the slope angle play important roles for estimating drainage basins with landslides. In addition, drainage basins where landslides occur frequently tend to have concave terrain and lower slope angle. While in the Setogawa group, distribution of landslides is not uniform, and landslides are concentrated in the drainage basins where slope angles are relatively small.<br>It is concluded that the decision tree technique can indicate the order of important explanatory variables clearly. This technique can also indicate the quantitative relationship between landslides and topography/lithology.

Journal

Citations (4)*help

See more

References(3)*help

See more

Keywords

Details 詳細情報について

Report a problem

Back to top