Spatial-temporal graph neural networks for groundwater data
Scientific Reports volume 14, Article number: 24564 (2024 ) Cite this article
This paper introduces a novel application of spatial-temporal graph neural networks (ST-GNNs) to predict groundwater levels. Groundwater level prediction is inherently complex, influenced by various hydrological, meteorological, and anthropogenic factors. Traditional prediction models often struggle with the nonlinearity and non-stationary characteristics of groundwater data. Our study leverages the capabilities of ST-GNNs to address these challenges in the Overbetuwe area, Netherlands. We utilize a comprehensive dataset encompassing 395 groundwater level time series and auxiliary data such as precipitation, evaporation, river stages, and pumping well data. The graph-based framework of our ST-GNN model facilitates the integration of spatial interconnectivity and temporal dynamics, capturing the complex interactions within the groundwater system. Our modified Multivariate Time Graph Neural Network model shows significant improvements over traditional methods, particularly in handling missing data and forecasting future groundwater levels with minimal bias. The model’s performance is rigorously evaluated when trained and applied with both synthetic and measured data, demonstrating superior accuracy and robustness in comparison to traditional numerical models in long-term forecasting. The study’s findings highlight the potential of ST-GNNs in environmental modeling, offering a significant step forward in predictive modeling of groundwater levels.
The complexity of groundwater level (GWL) prediction and modeling arises from its nonlinearity and sensitivity to various hydrological, meteorological, and anthropogenic influences1,2,3. Conventional GWL models, such as physical-based and traditional statistical methods, are limited by their high parameterization needs. This requirement makes the models computationally difficult to calibrate, as each parameter must be accurately estimated to reflect real-world conditions. Additionally, the models are challenged by computational intensity and difficulty in capturing the temporal evolution and nonlinearity in GWL data4,5,6,7. In this context, deep learning models and, in particular, spatial-temporal graph neural networks (ST-GNNs) offer promising new avenues for accurate and efficient groundwater forecasting. Our study focuses on leveraging ST-GNNs to predict groundwater levels in the Overbetuwe area, Netherlands. The selection of this area is driven by the availability of data8.
