Document Type

Data

Publication Date

4-17-2024

Abstract

Natural disasters pose significant threats to the environment and make land unsuitable for man. The lava from volcanic eruptions rapidly moves on the Earth surface and affects both natural and man-made features on the Earth. Remote sensing is relevant in extracting information about the areas affected by lava flows without physically visiting such areas. The objective of this study is to develop a workflow for determining the optimal ground sample distance of a digital elevation model to delineate the lava flow paths using object-based image analysis and geomorphologic analysis. For the experiment, the satellite images and digital elevation model of Afar region of Ethiopia was used. The satellite images of the affected areas were downloaded and segmented using large-scale mean shift image segmentation algorithm in the Orfeo Toolbox to find the lava objects. The main centerline of the lava object from the 2008 ASTER image and the flow paths from the digital elevation model with varying ground sample distance were used to find the optimal ground sample distance. In the result, the optimal resolution for delineating lava flow was found to be 36m with 145.38m RMSE of minimum distances. The result was further verified using the 2019 Landsat 8 images of the area affected by the 2017 volcanic eruption, which remained active at least to the 2019. The verification showed 36m ground sample distance resulted in 9th best result out of 91 candidates. The study showed the workflow using remote sensing and open source software to find the optimal ground sample distance of a digital elevation model for lava flow path delineation. The study is expected to contribute to the lava flow analysis efforts by scientists and researchers.

Creative Commons License

Creative Commons Attribution 4.0 International License
This work is licensed under a Creative Commons Attribution 4.0 International License.

Share

COinS