Abstract:Rainfall and reservoir water level are commonly regarded as the two major influencing factors for reservoir landslides and are employed for landslide displacement prediction, yet their daily data are readily available with current monitoring technology, which makes a more refined analysis possible. However, until now, few efforts have been made to predict landslide displacements using daily data, which is likely to substantially improve accuracy and is crucial for landslide early warning. A novel feature enhancement approach for extracting critical characteristics from daily rainfall and reservoir water level data for use in landslide displacement prediction is proposed in this study. Six models, including gated recurrent units (GRUs), long short-term memory (LSTM), and support vector regression (SVR) with an unenhanced dataset and GRU-E, LSTM-E, and SVR-E with an enhanced dataset, were employed for displacement predictions at four GPS monitoring stations on the Baijiabao landslide, a typical step-like reservoir landslide. The results show that the accuracy values of all the enhanced models were significantly improved, and the GRU-E model achieved the most significant improvement, with the RMSE decreasing by 24.39% and R2 increasing by 0.2693, followed by the LSTM-E and SVR-E models. Further, the GRU-E model consistently outperformed the other models, achieving the highest R2 of 0.6265 and the lowest RMSE of 16.5208 mm, significantly superior than the others. This study indicates the feasibility of improving the accuracy of landslide monthly displacement predictions with finer monitoring data and provides valuable insights for future research.

What problem does this paper attempt to address?

The main problem that this paper attempts to solve is to improve the accuracy of landslide displacement prediction. Specifically, the author proposes a new feature - enhancement method, using more fine - grained time - scale data (such as daily rainfall and reservoir water - level data) to improve the landslide displacement prediction model. Traditional landslide displacement prediction models mainly use monthly cumulative rainfall and average reservoir water - level as input features, and these features may not be able to capture the significant changes in rainfall distribution and reservoir water - level fluctuations within a month. Therefore, by introducing more detailed daily data, the changes in influencing factors can be described more accurately, thereby improving the prediction accuracy of the model. ### Background of the Paper Landslide is a complex geological phenomenon, which may lead to casualties, infrastructure damage and natural environment destruction. The Three Gorges Reservoir Area (TGRA) is located along the Yangtze River for 600 kilometers. Due to seasonal rainfall and periodic fluctuations in reservoir water - level, there are more than 4,200 potential landslide sites. For example, in 2003, when the reservoir water - level reached 135 meters, the Qianjiangping landslide was re - activated, causing a 30 - meter - high water wave to capsize boats, resulting in 24 deaths and 346 houses being destroyed. In order to reduce the losses caused by landslides, a professional monitoring network has been established in the Three Gorges Reservoir Area, providing a large amount of on - site landslide monitoring data. These data make it possible to provide early warnings through landslide displacement prediction, providing a basis for risk management decisions. ### Research Objectives The main objective of this paper is to propose a new feature - enhancement method based on the multi - layer Recurrent Neural Network (RNN) model to solve the limitations of traditional input features (such as monthly monitoring data). Specifically, the research proposes a new method for constructing a landslide displacement prediction data set from daily rainfall and reservoir water - level data, which can describe the changes in influencing factors more accurately, thereby improving the prediction accuracy. ### Methods 1. **Landslide Monitoring Data Collection**: Collect long - term landslide surface displacement measurement data as well as rainfall and reservoir water - level data. 2. **Data Pre - processing and Data Set Construction**: - **Feature Enhancement**: Calculate the mean, median, maximum, minimum, variance and standard deviation of the daily reservoir water - level, as well as the classification and percentage of daily rainfall intensity. - **Data Set Construction**: Use the sliding - window technique to construct training and testing data sets. 3. **Multi - layer RNN Model**: Adopt a multi - layer RNN model, including LSTM and GRU units, for landslide displacement prediction. 4. **Hyper - parameter Optimization**: Select the best hyper - parameters through grid search and Particle Swarm Optimization algorithm. 5. **Performance Evaluation**: Use Mean Squared Error (MSE), Root Mean Squared Error (RMSE) and Coefficient of Determination (R²) for model evaluation. ### Results The experimental results show that the accuracy of all enhanced models has been significantly improved, among which the GRU - E model performs the best, with the RMSE reduced by 24.39% and the R² increased by 0.2693. The prediction performance of the GRU - E model on four GPS monitoring stations is better than that of other models, reaching the highest R² value of 0.6265 and the lowest RMSE value of 16.5208 mm. ### Conclusions This research proves that using more fine - grained monitoring data (such as daily data) can significantly improve the accuracy of monthly landslide displacement prediction, and provides valuable insights for future landslide prediction research.

A Novel Approach to Enhance Landslide Displacement Prediction with Finer Monitoring Data: A Case Study of the Baijiabao Landslide

A Novel Hybrid LMD-ETS-TCN Approach for Predicting Landslide Displacement Based on GPS Time Series Analysis

A Data‐driven Intelligent Model for Landslide Displacement Prediction

Landslide Displacement Prediction Based on Multivariate LSTM Model

A Novel Approach for Displacement Interval Forecasting of Landslides with Step-Like Displacement Pattern

Data-augmented Landslide Displacement Prediction Using Generative Adversarial Network

Study and verification on an improved comprehensive prediction model of landslide displacement

Landslide Displacement Prediction Based on Time Series Analysis and Double-BiLSTM Model

A Novel Model for Landslide Displacement Prediction Based on EDR Selection and Multi-Swarm Intelligence Optimization Algorithm

A Hybrid Recurrent Model Based on LSTM and Statistical Evaluation to Predict Soil Displacement for Landslide Monitoring

Data Mining and Deep Learning for Predicting the Displacement of "Step-like" Landslides

A novel data-driven hybrid intelligent prediction model for reservoir landslide displacement

Study on Wavelet Multi-Scale Analysis and Prediction of Landslide Groundwater

Prediction Interval Estimation of Landslide Displacement Using Bootstrap, Variational Mode Decomposition, and Long and Short-Term Time-Series Network

Using Complementary Ensemble Empirical Mode Decomposition and Gated Recurrent Unit to Predict Landslide Displacements in Dam Reservoir

Combining Numerical Simulation and Deep Learning for Landslide Displacement Prediction: An Attempt to Expand the Deep Learning Dataset

A novel displacement prediction method using gated recurrent unit model with time series analysis in the Erdaohe landslide

Displacement prediction of step-like landslides based on feature optimization and VMD-Bi-LSTM: a case study of the Bazimen and Baishuihe landslides in the Three Gorges, China

Prediction of landslide step-like displacement using factor preprocessing-based hybrid optimized SVR model in the Three Gorges Reservoir, China

Spatiotemporal prediction of landslide displacement using deep learning approaches based on monitored time-series displacement data: a case in the Huanglianshu landslide

Prediction of Landslide Displacement Based on the Combined VMD-Stacked LSTM-TAR Model