Toward Urban Water Security: Broadening the Use of Machine Learning Methods for Mitigating Urban Water Hazards

Abstract

Due to the complex interactions of human activity and the hydrological cycle, achieving urban water security requires comprehensive planning processes that address urban water hazards using a holistic approach. However, the effective implementation of such an approach requires the collection and curation of large amounts of disparate data, and reliable methods for modeling processes that may be co-evolutionary yet traditionally represented in non-integrable ways. In recent decades, many hydrological studies have utilized advanced machine learning and information technologies to approximate and predict physical processes, yet none have synthesized these methods into a comprehensive urban water security plan. In this paper, we review ways in which advanced machine learning techniques have been applied to specific aspects of the hydrological cycle and discuss their potential applications for addressing challenges in mitigating multiple water hazards over urban areas. We also describe a vision that integrates these machine learning applications into a comprehensive watershed-to-community planning workflow for smart-cities management of urban water resources.

Driving Forces and Socio-Economic Impacts of Low-Flow Events in Central Europe: A Literature Review Using DPSIR Criteria

“…Table 1 lists the 33 papers identified through the keyword search described in Section 2 and the selection process shown in Figure 2. Bryan et al, 2020 [33] 1 UK A Cap-Net UNDP, 2015 [34] 2 Global A Ribbe et al, 2013 [35] 2 Mekong River Basin A Tsakiris, 2016 [36] 1 Global A Wang et al, 2019 [37] 1 Heihe River (China) A WWF and GIWP, 2016 [13] 2 Global A Al Hussain, 2017 [38] 5 Lower Teesta River Basin (Bangladesh) B Assubayeva, 2022 [39] 5 Central Asia B Chilikova-Lubomirova et al, 2020 [40] 4 Bulgaria B Chung et al, 2009a [41] 1 South Korea B Chung et al, 2009b [42] 1 South Korea B Flörke et al, 2011 [43] 4 Europe B Holman et al, 2021 [44] 1 UK B Ilcheva et al, 2015 [45] 1 Southeast Europe B Kolcheva et al, 2016 [46] 1 Bulgaria B Kossida, 2015 [47] 5 Greece B Kovar et al, 2009 [48] 4 Czech Republic B Olsson et al, 2010 [49] 4 Europe B WHO, 2011 [50] 2 Europe B Vargas Amelin, 2016 [51] 5 Spain B Zucaro et al, 2017 [52] 1 Italy B Allen-Dumas et al, 2021 [53] 1 Global C Daoud, 2015 [54] 5 Egypt C Eddoughri et al, 2022 [55] 1 Morocco C Mishra et al, 2018 [56] 3 Vietnam C Nyangena, 2018 [57] 5 Kenya C Perović et al, 2021 [58] 1 Serbia C Pociask-Karteczka et al, 2018 [59] 3 Poland C Reckermann et al, 2022 [60] 1 Baltic Sea region C Soares et al, 2019 [61] 1 Portugal C Sperotto, 2013 [62] 5 most of the low-flow consequences descr the experts interviewed, see Appendix B low-flow consequences and conflicts in transcribed so that they could be eval interviews (see Section 3.6), the statemen…”

“…Lower Teesta River Basin (Bangladesh) B Assubayeva, 2022 [39] 5 Central Asia B Chilikova-Lubomirova et al, 2020 [40] 4 Bulgaria B Chung et al, 2009a [41] 1 South Korea B Chung et al, 2009b [42] 1 South Korea B Flörke et al, [43] 4 Europe B Holman et al, 2021 [44] 1 UK B Ilcheva et al, 2015 [45] 1 Southeast Europe B Kolcheva et al, 2016 [46] 1 Bulgaria B Kossida, 2015 [47] 5 Greece B Kovar et al, [48] 4 Czech Republic B Olsson et al, 2010 [49] 4 Europe B WHO, 2011 [50] 2 Europe B Vargas Amelin, 2016 [51] 5 Spain B Zucaro et al, 2017 [52] 1 Italy B Allen-Dumas et al, 2021 [53] 1 Global C Daoud, 2015 [54] 5 Egypt C Eddoughri et al, 2022 [55] 1 Morocco C Mishra et al, 2018 [56] 3 Vietnam C Nyangena, 2018 [57] 5 Kenya C Perović et al, 2021 [58] 1 Serbia C Pociask-Karteczka et al, 2018 [59] 3 Poland C Reckermann et al, 2022 [60] 1 Baltic Sea region C Soares et al, [61] 1 Portugal C Sperotto, 2013 [62] 5 North Adriatic coast C Swart et al, [63] 4 Europe C Wade et al, [64] 4 UK C 15 A = 6, B = 17, C = 12 the interview partners, the associated partners of the DryRivers project we consulted. In addition, existing cooperation networks were used.…”

An automated machine learning methodology for the improved prediction of reference evapotranspiration using minimal input parameters

“…The automated ML pipeline creation of AutoML methodologies, regardless of dataset size and, without human involvement, aids in formulating a uniform learnable problem and thus creating a generalized model by using a pooled dataset containing data from multiple locations. According to the review by Allen‐Dumas et al (2021), there is an urgent need to integrate various ML methods to address multiple water hazards such as floods, drought, water contamination and soil erosion in urban water systems. Being a dynamic ML modelling tool, AutoML could ignite a surge in the ML applications in hydrology such as monitoring, early warning, prediction of urban water hazards and so on.…”

“…multiple locations. According to the review byAllen-Dumas et al (2021), there is an urgent need to integrate various ML methods to address multiple water hazards such as floods, drought, water contamination and soil erosion in urban water systems. Being a dynamic ML modelling tool, AutoML could ignite a surge in the ML applications in hydrology such as monitoring, early warning, prediction of urban water hazards…”

Application of Gis in the Assessment of Flood Risk in the Region Zenica - Doboj Canton

“…Trend analyses show that major flood disasters as well as damages and losses generated by them have increased drastically in recent years (Berz, 2000;Barredo, 2007). Because of the accelerated pace of anthropogenic activity, hazard frequency and intensity is exacerbated, requiring immediate delivery of science-based solutions for mitigation, resilience, and adaptation that can be quickly deployed in any hazard-prone area (Allen-Dumas, Xu, Kurte, Rastogi, 2021). Mitigating these urban water hazards is challenging for watershed management and the urban planning community (Eriksson et al, 2015).…”