Short-term Hourly Streamflow Prediction with Graph Convolutional GRU Networks

TL;DR

This paper introduces a Graph Convolutional GRU model for short-term hourly streamflow prediction, demonstrating improved accuracy over baseline methods in flood-prone areas, aiding in disaster preparedness.

Contribution

The study presents a novel Graph Convolutional GRU network specifically designed for short-term streamflow forecasting using river network data.

Findings

Outperforms persistence baseline in accuracy

Surpasses convolutional bidirectional GRU in performance

Effective for flood prediction in climate change scenarios

Abstract

The frequency and impact of floods are expected to increase due to climate change. It is crucial to predict streamflow, consequently flooding, in order to prepare and mitigate its consequences in terms of property damage and fatalities. This paper presents a Graph Convolutional GRUs based model to predict the next 36 hours of streamflow for a sensor location using the upstream river network. As shown in experiment results, the model presented in this study provides better performance than the persistence baseline and a Convolutional Bidirectional GRU network for the selected study area in short-term streamflow prediction.