Hydrological model parameter regionalization: Runoff estimation using machine learning techniques in the Tha Chin River Basin, Thailand.

Understanding hydrological processes necessitates the use of modeling techniques due to the intricate interactions among environmental factors. Estimating model parameters remains a significant challenge in runoff modeling for ungauged catchments. This research evaluates the Soil and Water Assessment Tool's capacity to simulate hydrological behaviors in the Tha Chin River Basin with an emphasis on runoff predictions from the regionalization of hydrological parameters of the gauged basin, Mae Khlong River Basin. Historical data of Mae Khlong River Basin from 1993 to 2017 were utilized for calibration, followed by validation using data from 2018 to 2022. •Calibration results showed the SWAT model's reasonable accuracy, with R² = 0.85, and the validation with R² of 0.64, indicating a satisfactory match between observed and simulated runoff.•Utilizing Machine Learning (ML) techniques for parameter regionalization revealed nuanced differences in model performance. The Random Forest (RF) model exhibited an R² of 0.60 and the Artificial Neural Networks (ANN) model slightly improved upon RF, showing an R² of 0.61 while the Support Vector Machine (SVM) model demonstrated the highest overall performance, with an R² of 0.63.•This study highlights the effectiveness of the SWAT and ML techniques in predicting runoff for ungauged catchments, emphasizing their potential to enhance hydrological modeling accuracy. Future research should focus on integrating these methodologies in various basins and improving data collection for better model performance.

Imputation of missing daily rainfall data; A comparison between artificial intelligence and statistical techniques.

Handling missing values is a critical component of the data processing in hydrological modeling. The key objective of this research is to assess statistical techniques (STs) and artificial intelligence-based techniques (AITs) for imputing missing daily rainfall values and recommend a methodology applicable to the mountainous terrain of northern Thailand. In this study, 30 years of daily rainfall data was collected from 20 rainfall stations in northern Thailand and randomly 25-35 % of data was deleted from four target stations based on Spearman correlation coefficient between the target and neighboring stations. Imputation models were developed on training and testing datasets and statistically evaluated by mean absolute error (MAE), root mean square error (RMSE), coefficient of determination (R2), and correlation coefficient (r). This study used STs, including arithmetic averaging (AA), multiple linear regression (MLR), normal-ratio (NR), nonlinear iterative partial least squares (NIPALS) algorithm, and linear interpolation was used.•STs results were compared with AITs, including long-short-term-memory recurrent neural network (LSTM-RNN), M5 model tree (M5-MT), multilayer perceptron neural networks (MLPNN), support vector regression with polynomial and radial basis function SVR-poly and SVR-RBF.•The findings revealed that MLR imputation model achieved an average MAE of 0.98, RMSE of 4.52, and R2 was about 79.6 % at all target stations. On the other hand, for the M5-MT model, the average MAE was 0.91, RMSE was about 4.52, and R2 was around 79.8 % compared to other STs and AITs. M5-MT was most prominent among AITs. Notably, the MLR technique stood out as a recommended approach due to its ability to deliver good estimation results while offering a transparent mechanism and not necessitating prior knowledge for model creation.