Time Distributed Deep Learning models for Purely Exogenous Forecasting. Application to Water Table Depth Prediction using Weather Image Time Series

2024-09-20

Environmental Modelling & Software

AI Summary

In regions with sparse, delayed, and discontinuous hydrological observations, reliable groundwater level (GWL) forecasting remains challenging. Method: This paper proposes a novel paradigm for predicting GWL depth using only readily available, high-quality meteorological image time series. To address future shift issues in exogenous-variable-driven sequence modeling, we introduce TDC-UnPWaveNet—a purely exogenous, non-autoregressive adaptation of WaveNet—featuring a novel Channel Distributed layer. Complementarily, TDC-LSTM is designed to correct systematic biases. The two models jointly optimize dynamic correlation and predictive accuracy. Results: Evaluated in the Grana-Maira watershed, TDC-LSTM significantly reduces mean bias, while TDC-UnPWaveNet substantially improves temporal correlation and Kling–Gupta Efficiency (KGE) by up to 23.6%. This framework provides a generalizable deep learning solution for sustainable groundwater management in data-scarce regions.

Technology Category

• Machine Learning: Time-Series/Data Streams
• Natural Language Processing: Language Grounding & Multi-modal NLP
• Data Mining & Knowledge Management: Data Stream Mining

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Groundwater resources are one of the most relevant elements in the water cycle, therefore developing models to accurately predict them is a pivotal task in the sustainable resources management framework. Deep Learning (DL) models have been revealed very effective in hydrology, especially by feeding spatially distributed data (e.g. raster data). In many regions, hydrological measurements are difficult to obtain regularly or periodically in time, and in some cases, last available data are not up to date. Reversely, weather data, which significantly impacts water resources, are usually more available and with higher quality. More specifically, we have proposed two different DL models to predict the water table depth in the Grana-Maira catchment (Piemonte, IT) using only exogenous weather image time series. To deal with the image time series, both models are made of a first Time Distributed Convolutional Neural Network (TDC) which encodes the image available at each time step into a vectorial representation. The first model, TDC-LSTM uses then a Sequential Module based on an LSTM layer to learn temporal relations and output the predictions. The second model, TDC-UnPWaveNet uses instead a new version of the WaveNet architecture, adapted here to output a sequence shorter and completely shifted in the future with respect to the input one. To this aim, and to deal with the different sequence lengths in the UnPWaveNet, we have designed a new Channel Distributed layer, that acts like a Time Distributed one but on the channel dimension, i.e. applying the same set of operations to each channel of the input. TDC-LSTM and TDC-UnPWaveNet have shown both remarkable results. However, the two models have focused on different learnable information: TDC-LSTM has focused more on lowering the bias, while the TDC-UnPWaveNet has focused more on the temporal dynamics maximising correlation and KGE.

Problem

• Predict water table depth using weather image time series
• Develop DL models for purely exogenous forecasting in hydrology
• Address data scarcity in hydrological measurements with weather data

Innovation

• Time Distributed CNN encodes weather image series
• LSTM layer learns temporal relations for predictions
• UnPWaveNet adapts WaveNet for future sequence output