A Parsimonious Setup for Streamflow Forecasting using CNN-LSTM

Sudan Pokharel; Tirthankar Roy

A Parsimonious Setup for Streamflow Forecasting using CNN-LSTM

Machine Learning 2024-04-12 v1

Authors: Sudan Pokharel , Tirthankar Roy

Abstract

Significant strides have been made in advancing streamflow predictions, notably with the introduction of cutting-edge machine-learning models. Predominantly, Long Short-Term Memories (LSTMs) and Convolution Neural Networks (CNNs) have been widely employed in this domain. While LSTMs are applicable in both rainfall-runoff and time series settings, CNN-LSTMs have primarily been utilized in rainfall-runoff scenarios. In this study, we extend the application of CNN-LSTMs to time series settings, leveraging lagged streamflow data in conjunction with precipitation and temperature data to predict streamflow. Our results show a substantial improvement in predictive performance in 21 out of 32 HUC8 basins in Nebraska, showcasing noteworthy increases in the Kling-Gupta Efficiency (KGE) values. These results highlight the effectiveness of CNN-LSTMs in time series settings, particularly for spatiotemporal hydrological modeling, for more accurate and robust streamflow predictions.

Keywords

long short-term memory time series forecasting

Cite

@article{arxiv.2404.07924,
  title  = {A Parsimonious Setup for Streamflow Forecasting using CNN-LSTM},
  author = {Sudan Pokharel and Tirthankar Roy},
  journal= {arXiv preprint arXiv:2404.07924},
  year   = {2024}
}

A Parsimonious Setup for Streamflow Forecasting using CNN-LSTM

Abstract

Keywords

Cite

Related papers