Available online at www.CivileJournal.org
Civil Engineering Journal
(E-ISSN: 2476-3055; ISSN: 2676-6957)
Vol. 8, No. 12, December, 2022
3854
Some Approaches to the Prediction of Permeability Parameters
in a Finite Element Program for Early Warning
Krairoj Mahannopkul
1
, Chollada Kanjanakul
2*
1
Department of Teacher Training in Civil Engineering, King Mongkut’s University of Technology North Bangkok, Thailand.
2
Department of Civil Engineering, The Research Unit of Technology and Innovation on Civil Engineering (RICE), Rajamangala University of
Technology Srivijaya, Nakhon Si Thammarat, Thailand.
Received 01 October 2022; Revised 14 November 2022; Accepted 23 November 2022; Published 01 December 2022
Abstract
Recently, landslides often occurred in natural soil slopes in the tropical region, which correlate with the rainy season.
Rainfall infiltration leads to groundwater level fluctuations. The increased positive pore-water pressures due to rainfall
influence have affected the properties and behavior of the unsaturated soil slope. In this research, the Finite Element Method
of SEEP/W and SLOPE/W analyzes the factor safety of the slope affected by pore water pressure change due to rainfall.
The Soil Water Characteristic Curve (SWCC) and Hydraulic Conductivity function were obtained from sieve analysis and
Atterberg's limit. In addition, unsaturated soil properties from the UNSODA code are estimated based on grain-size
distribution using the SWRC program. The study area is in Khanom District, southern Thailand. The results show that the
soil slope at the site became unstable on November 18, 2021, with F.S. = 1.0, which agrees well with the date of the disaster.
In conclusion, the slope stability analysis without the parameters from the unsaturated soil hydraulic database (UNSODA)
leads to the F.S. value being higher than the actual value, and the alarm estimation would be inaccurate.
Keywords: Soil Water Characteristic Curve (SWCC); Unsaturated Soil Hydraulic Database (UNSODA); Early Warning; Slope Stability
Analysis; Rainfall Infiltration.
1. Introduction
Landslides in mountainous areas during heavy rainfall can result in the loss of lives and properties. Various
geotechnical studies have applied an early warning system for landslides to mitigate the risk [1]. For example, Yang et
al. [2] established the rainfall threshold for landslide activity in Dazhou, China, and used the parameter combined with
the intraday rainfall to represent the rainfall condition. Kardani et al. [3] used the finite element method (FEM) to
simulate slope stability and generate synthetic data for the training of the optimized machine learning methods (OML)
that were employed to predict slope stability on the testing dataset. Zheng et al. [4] proposed the triangular fuzzy number
and the analytic hierarchy process method with GIS that effectively predicted the distribution of geo-hazard risk in the
study region along the Chengdu-Kunming railway, southwestern China, generated from case studies within the past ten
years.
In a particular region, rainfall is the main factor for slope failures, and rainfall thresholds are the significant parameter
in forecasting the landslide probability [5, 6]. Many researchers have shown great potential in predicting landslide and
debris flow events by machine learning with time-series data processing methods based on continuous rainfall records.
Zhao et al. [7] used this approach to predict debris flow events based on continuous rainfall records from five rain gauges
in the catchment. Jiang et al. [8] presented the probabilistic rainfall thresholds for debris flow occurrences under specific
rainfall conditions by applying a Bayesian approach based on the Wenchuan earthquake.
* Corresponding author: chollada.k@rmutsv.ac.th, chollada-ka@hotmail.com
http://dx.doi.org/10.28991/CEJ-2022-08-12-014
© 2022 by the authors. Licensee C.E.J, Tehran, Iran. This article is an open access article distributed under the terms and
conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).