Forecasting Multiple Groundwater Time Series with Local and Global Deep Learning Networks

Time series data from environmental monitoring stations are often analysed with machine learning methods on an individual basis, however recent advances in the machine learning field point to the advantages of incorporating multiple related time series from the same monitoring network within a 'global' model. This approach provides the opportunity for larger training data sets, allows information to be shared across the network, leading to greater generalisability, and can overcome issues encountered in the individual time series, such as small datasets or missing data. We present a case study involving the analysis of 165 time series from groundwater monitoring wells in the Namoi region of Australia. Analyses of the multiple time series using a variety of different aggregations are compared and contrasted (with single time series, subsets, and all of the time series together), using variations of the multilayer perceptron (MLP), self-organizing map (SOM), long short-term memory (LSTM), and a recently developed LSTM extension (DeepAR) that incorporates autoregressive terms and handles multiple time series. The benefits, in terms of prediction performance, of these various approaches are investigated, and challenges such as differing measurement frequencies and variations in temporal patterns between the time series are discussed. We conclude with some discussion regarding recommendations and opportunities associated with using networks of environmental data to help inform future resource-related decision making

Artificial neural network · Autoregressive integrated moving average · Autoregressive model · Data mining · Field (mathematics · Machine learning · Missing data · Multilayer perceptron · Resource (disambiguation · Series (stratigraphy · Statistics · Time series · Computer Science · Hydrological Forecasting Using AI · Hydrology and Watershed Management Studies · Mathematics · Time Series Analysis and Forecasting · Artificial Intelligence

• Forecasting heat-related impacts with multivariate multi-step time series models using advanced deep learning

  Open Access•Jung-Ching Kan, Marlon Vieira Passos et al.•Sustainable Cities and Society•2026

  Add to list

• Time Series Analysis by State Space Methods

  James Durbin, Siem Jan Koopman•Time Series Analysis by State…•2012

  Add to list

• Deep learning and process understanding for data-driven Earth system science

  Open Access•Markus Reichstein, Gustau Camps-Valls et al.•Nature•2019

  Add to list

• The self-organizing map

  Open Access•Teuvo Kohonen•Proceedings of the IEEE•1990

  Add to list

• Finding Structure in Time

  Open Access•Jeffrey L Elman•Cognitive Science•1990

  Add to list