Abstract
With the growing use of machine learning (ML) techniques in hydrological applications, there is a need to analyze the robustness, performance, and reliability of predictions made with these ML models. In this paper we analyze the accuracy and variability of groundwater level predictions obtained from a Multilayer Perceptron (MLP) model with optimized hyperparameters for different amounts and types of available training data. The MLP model is trained on point observations of features like groundwater levels, temperature, precipitation, and river flow in various combinations, for different periods and temporal resolutions. We analyze the sensitivity of the MLP predictions at three different test locations in California, United States and derive recommendations for training features to obtain accurate predictions. We show that the use of all available features and data for training the MLP does not necessarily ensure the best predictive performance at all locations. More specifically, river flow and precipitation data are important training features for some, but not all locations. However, we find that predictions made with MLPs that are trained solely on temperature and historical groundwater level measurements as features, without additional hydrological information, are unreliable at all locations.
Highlights
Groundwater is an important source of freshwater, accounting for almost 38% of the global irrigation demand (Siebert et al, 2010)
We make predictions for a time frame that was not used during optimization or training of the Multilayer Perceptron (MLP) (2 years unless otherwise specified), to assess the ability of the models to extrapolate beyond their training time frame
Deep learning (DL) models offer a promising alternative for capturing the complex interactions between features such as groundwater levels, river flow, temperature, and precipitation
Summary
Groundwater is an important source of freshwater, accounting for almost 38% of the global irrigation demand (Siebert et al, 2010). With growing economies and increasing food demand, the stress on freshwater aquifers has increased in places like North America and Asia (Aeschbach-Hertig and Gleeson, 2012). This situation is further aggravated by increased climate variability. In California, USA, groundwater provides nearly 40% of the water used by the state’s cities and farms. Many of the state’s groundwater basins have experienced long-term overdraft due to withdrawal rates exceeding recharge rates. The negative impacts of long-term overdraft include higher energy requirements for pumping water from deeper wells, land subsidence, reduced river flow, and impaired water quality (especially in coastal aquifers due to saltwater intrusion).
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.