Articles | Volume 20, issue 8
https://doi.org/10.5194/hess-20-3379-2016
https://doi.org/10.5194/hess-20-3379-2016
Research article
 | 
23 Aug 2016
Research article |  | 23 Aug 2016

Case-based knowledge formalization and reasoning method for digital terrain analysis – application to extracting drainage networks

Cheng-Zhi Qin, Xue-Wei Wu, Jing-Chao Jiang, and A-Xing Zhu

Abstract. Application of digital terrain analysis (DTA), which is typically a modeling process involving workflow building, relies heavily on DTA domain knowledge of the match between the algorithm (and its parameter settings) and the application context (including the target task, the terrain in the study area, the DEM resolution, etc.), which is referred to as application-context knowledge. However, existing DTA-assisted tools often cannot use application-context knowledge because this type of DTA knowledge has not been formalized to be available for inference in these tools. This situation makes the DTA workflow-building process difficult for users, especially non-expert users. This paper proposes a case-based formalization for DTA application-context knowledge and a corresponding case-based reasoning method. A case in this context consists of a series of indices that formalize the DTA application-context knowledge and the corresponding similarity calculation methods for case-based reasoning. A preliminary experiment to determine the catchment area threshold for extracting drainage networks has been conducted to evaluate the performance of the proposed method. In the experiment, 124 cases of drainage network extraction (50 for evaluation and 74 for reasoning) were prepared from peer-reviewed journal articles. Preliminary evaluation shows that the proposed case-based method is a suitable way to use DTA application-context knowledge to achieve a marked reduction in the modeling burden for users.

Download
Short summary
Application of digital terrain analysis (DTA), which is typically a modeling process involving workflow building, relies heavily on DTA domain knowledge. However, the DTA knowledge has not been formalized well to be available for inference in automatic tools. We propose a case-based methodology to solve this problem. This methodology can also be applied to other domains of geographical modeling with a similar situation.