Using machine learning to determine acceptable levels of groundwater consumption in Iran Sami Ghordoyee Milan a, ⁎, Zahra Kayhomayoon b , Naser Arya Azar c , Ronny Berndtsson d , Mohammad Reza Ramezani e , Hamid Kardan Moghaddam f a Department of Water Engineering, College of Aburaihan, University of Tehran, Tehran, Iran b Department of Geology, Payame Noor University, Tehran, Iran c Department of Water Engineering, Faculty of Agriculture, University of Tabriz, Tabriz, Iran d Division of Water Resources Engineering & Centre for Advanced Middle Eastern Studies, Lund University, Lund, Sweden e School of Engineering and Built Environment, Griffith University, Kessels Road, Nathan, Qld 4111, Australia f Water Research Institute, Ministry of Energy Water Research Institute, Tehran, Iran abstract article info Article history: Received 24 April 2022 Received in revised form 22 November 2022 Accepted 23 November 2022 Available online 28 November 2022 Editor: Dr. Jingzheng Ren Groundwater footprint index (GFI) is an essential indicator to assess the sustainability of groundwater aquifers. Prediction of future GFI can significantly help managers and decision-makers of groundwater supply to better plan for future resilient consumption of surface and groundwater. In this context, artificial intelligence and ma- chine learning models can aid to predict GFI in view of lacking or uncertain data. We used this technique to pre- dict GFI for 178 Iranian aquifers. To our knowledge, this is the first time that GFI was predicted using machine learning models. Four models, i.e., adaptive neuro-fuzzy inference system, least-squares support vector regres- sion, random forest, and gene expression programming, were used to predict GFI. Systematic combinations of eight variables, including precipitation, recharge, return water, infiltration from the river to the aquifer, ground- water exploitation, aquifer area, evaporation, and river drainage from the aquifer were used in the form of nine input scenarios for GFI prediction. The results showed that inclusion of all input variables gave the best results for predicting the GFI. Predicted GFIs were generally between 0.5 and 8 with an average of 1.9. A value above 1 in- dicates that groundwater consumption is not resilient that can adversely affect available groundwater resources in the future. Over-use of groundwater can lead to land subsidence. Especially, aquifers located in Qom, Qazvin, Varamin, and Hamedan provinces of Iran may be affected due to large over-use. Among the four models, least- squares support vector regression resulted in the highest prediction performance. Due to the poor performance of adaptive neuro-fuzzy inference system, the novel Harris hawks optimization algorithm was used to improve the performance of adaptive neuro-fuzzy inference system. The Harris hawks optimization - adaptive neuro- fuzzy inference system hybrid model improved the GFI prediction performance. Machine learning methods improve prediction of GFI for aquifers and thus, can be used to better manage groundwater in areas with less reliable data. © 2022 Institution of Chemical Engineers. Published by Elsevier Ltd. All rights reserved. Keywords: Groundwater footprint Groundwater stress Land subsidence Machine learning 1. Introduction Due to decreasing surface water in arid and semiarid regions, groundwater resources are becoming increasingly important for eco- nomic development (Ashraf et al., 2021). At present, >36 % of drinking water and 42 % of agricultural needs in the world are supplied by groundwater (Kourakos et al., 2019). However, increasing unmanaged exploitation of groundwater resources causes problems such as aquifer depletion (Moghaddam et al., 2021a, 2021b), ground surface subsi- dence (Yazdian et al., 2021), declining water quality (Chen et al., 2019), saline water intrusion (Nasiri et al., 2021), agricultural land loss (Chanapathi et al., 2019), and depopulation in rural areas (Zhou et al., 2017; Hoekstra et al., 2018). Besides this, unregulated exploitation of groundwater resources affects and damages the hydraulic connection among various hydrological systems, and reduces the storage capacity of aquifers (Theesfeld, 2010). An essential aspect of recognizing aquifers in terms of exploitation and development potential is a correct understanding of the balance Sustainable Production and Consumption 35 (2023) 388–400 ⁎ Corresponding author. E-mail addresses: s.milan@ut.ac.ir (S. Ghordoyee Milan), Zkayhomayoon@pnu.ac.ir (Z. Kayhomayoon), ronny.berndtsson@tvrl.lth.se (R. Berndtsson), mohammad.ramezani@griffithuni.edu.au (M.R. Ramezani), h.kardan@wri.ac.ir (H. Kardan Moghaddam). https://doi.org/10.1016/j.spc.2022.11.018 2352-5509/© 2022 Institution of Chemical Engineers. Published by Elsevier Ltd. All rights reserved. Contents lists available at ScienceDirect Sustainable Production and Consumption journal homepage: www.elsevier.com/locate/spc