Application of Deep Learning for the Analysis of the Spatiotemporal Prediction of Monthly Total Precipitation in the Boyacá Department, Colombia DOI Creative Commons

Johann Santiago Niño Medina,

Marco Javier Suárez Barón,

José Antonio Reyes Suarez

et al.

Hydrology, Journal Year: 2024, Volume and Issue: 11(8), P. 127 - 127

Published: Aug. 21, 2024

Global climate change primarily affects the spatiotemporal variation in physical quantities, such as relative humidity, atmospheric pressure, ambient temperature, and, notably, precipitation levels. Accurate predictions remain elusive, necessitating tools for detailed analysis to better understand impacts on environment, agriculture, and society. This study compared three learning models, autoregressive integrated moving average (ARIMA), random forest regression (RF-R), long short-term memory neural network (LSTM-NN), using monthly data (in millimeters) from 757 locations Boyacá, Colombia. The inputs these models were based satellite images obtained Climate Hazards Group InfraRed Precipitation with Station (CHIRPS) data. LSTM-NN model outperformed others, precisely replicating observations both training testing datasets, significantly reducing root mean square error (RMSE), deviations of approximately 19 mm per location. Evaluation metrics (RMSE, MAE, R2, MSE) underscored LSTM model’s robustness accuracy capturing patterns. Consequently, was chosen predict over a 16-month period starting August 2023, offering reliable tool future meteorological forecasting planning region.

Language: Английский

Annual Peak Runoff Forecasting Using Two-Stage Input Variable Selection-Aided k-Nearest-Neighbors Ensemble DOI
Wei Sun, Decheng Zeng, Shu Chen

et al.

Water Resources Management, Journal Year: 2025, Volume and Issue: unknown

Published: March 5, 2025

Language: Английский

Citations

0
Monthly Streamflow Forecasting for the Irtysh River Based on a Deep Learning Model Combined with Runoff Decomposition DOI Open Access

Ki Yong,

Mingliang Li, Peng Xiao

et al.

Water, Journal Year: 2025, Volume and Issue: 17(9), P. 1375 - 1375

Published: May 2, 2025

The mid- and long-term hydrological forecast is important for water resource management disaster prevention. Moreover, forecasts in the region with poorly observed field meteorological data are a great challenge traditional models due to complexity of processes. To address this challenge, machine learning model, particularly deep model (DL), provides new tool improving accuracy runoff prediction. In study, we took Irtysh River, one longest rivers Central Asia well-known trans-boundary river basin poor observations, as an example develop based on LSTM combined decomposition by Maximal Overlap Discrete Wavelet Transform (MODWT) process target variables predicting monthly streamflow. We also proposed XGBoost-SHAP (Extreme Gradient Boost-SHapley Additive Explanations) method identification predictors from large-scale indices streamflow forecast. results suggest that MODWT shows robustness between training test period. better performance than benchmark without illustrated increased NSE. well identified nonlinear relationship streamflow, determined can be physically explained. Compared mutual information method, improves study illustrate ability forecast, methods developed provide effective approach improve prediction scarcely catchments.

Language: Английский

Citations

0
A singular spectrum analysis-enhanced BiTCN-selfattention model for runoff prediction DOI
Wenchuan Wang,

Feng-rui Ye,

Yiyang Wang

et al.

Earth Science Informatics, Journal Year: 2024, Volume and Issue: 18(1)

Published: Dec. 12, 2024

Language: Английский

Citations

1
Application of Deep Learning for the Analysis of the Spatiotemporal Prediction of Monthly Total Precipitation in the Boyacá Department, Colombia DOI Creative Commons

Johann Santiago Niño Medina,

Marco Javier Suárez Barón,

José Antonio Reyes Suarez

et al.

Hydrology, Journal Year: 2024, Volume and Issue: 11(8), P. 127 - 127

Published: Aug. 21, 2024

Global climate change primarily affects the spatiotemporal variation in physical quantities, such as relative humidity, atmospheric pressure, ambient temperature, and, notably, precipitation levels. Accurate predictions remain elusive, necessitating tools for detailed analysis to better understand impacts on environment, agriculture, and society. This study compared three learning models, autoregressive integrated moving average (ARIMA), random forest regression (RF-R), long short-term memory neural network (LSTM-NN), using monthly data (in millimeters) from 757 locations Boyacá, Colombia. The inputs these models were based satellite images obtained Climate Hazards Group InfraRed Precipitation with Station (CHIRPS) data. LSTM-NN model outperformed others, precisely replicating observations both training testing datasets, significantly reducing root mean square error (RMSE), deviations of approximately 19 mm per location. Evaluation metrics (RMSE, MAE, R2, MSE) underscored LSTM model’s robustness accuracy capturing patterns. Consequently, was chosen predict over a 16-month period starting August 2023, offering reliable tool future meteorological forecasting planning region.

Language: Английский

Citations

0