This repository contains the Jupyter Notebook and associated code for the paper titled "Synthetic Stream Gauges: An LSTM-Based Approach to Enhance River Streamflow Predictions in Unmonitored Segments (Under Review)", which describes the development and application of a synthetic stream gauge model using the Leaky Bucket concept combined with LSTM networks for analyzing synthetic data generated by this model.
The notebook illustrates the process of creating the Leaky Bucket model and the basin networks to produce synthetic data, as well as employing an LSTM model to separate the combined downstream flow of simulated main rivers into the individual upstream flows of its tributaries.
- Setup: Initial setup and library imports.
- Deep Learning Model: Definition and training of the neural network model.
- Final Model Evaluation: Evaluation of the model performance.
The code utilizes several libraries, including:
- Numpy
- Pandas
- Matplotlib
- Scikit-Learn
- PyTorch
- TQDM
- IPython
To run this notebook, please ensure you have installed all required libraries. A recommended approach is to create a virtual environment and install the dependencies via pip:
pip install numpy pandas matplotlib scikit-learn torch tqdm ipython
This project is licensed under the MIT License - see the LICENSE file in this repository for more details.
This work is part of ongoing research and has been submitted for review. We thank all contributors and collaborators for their insights and feedback.