Open AccessLand surface temperature (LST) and soil moisture index (SMI) to identify slope stability
Author(s) -
Sutanto Trijuni Putro,
Naima Azaiez Arif,
T Sarastika
Publication year - 2022
Publication title -
iop conference series. earth and environmental science
Language(s) - English
Resource type - Journals
eISSN - 1755-1307
pISSN - 1755-1315
DOI - 10.1088/1755-1315/986/1/012022
Subject(s) - environmental science , linear regression , stability (learning theory) , water content , vegetation (pathology) , satellite , landslide , remote sensing , regression analysis , soil science , mathematics , geology , computer science , statistics , geomorphology , physics , medicine , geotechnical engineering , pathology , astronomy , machine learning
Scientists widely use satellite images for scientific purposes, including investigation on earth science and environmental issues. Developing of many environmental models is due to replicating the natural process. Landslide is a known natural process controlled by slope stability which incorporates many parameters such as soil water content, morphology, and meteorological factor. Both LST and SMI were derived from satellite images, while SMI was the derivation of LST, meanwhile the use of both parameters in determining slope stability was rarely done. This research explores the use of LST and SMI in slope stability modeling. The LST analysis was calculated based on SEBAL (Surface Energy Balance Algorithms) using Landsat 8 imagery. The LST was then used to construct the SMI. Slope stability (FS) was calculated using the Selby model. All those variables were then cross-plotted in a regression to find the R2 value. The result shows a weak connection between FS-LST and FS-SMI with the R2 value of 9,09% and 8,16%. A stronger connection is only demonstrated in FS-Slope regression with a value of 70,98%. The weak R2 indicates that the model is not fit to calculate the FS of the Selby model. The LST and SMI were derived from satellite images and did not directly correspond to the soil characteristic as SMI was derived from LST and vegetation indices. Further empirical data collection needs to be used to build a better model on FS.