Groundwater is a vital source of freshwater, supporting the livelihood of over two billion people worldwide. The quantitative assessment of groundwater resources is critical for sustainable management of this strained resource, particularly as climate warming, population growth, and socioeconomic development further press the water resources. Rapid growth in the availability of a plethora of in-situ and remotely sensed data alongside advancements in data-driven methods and machine learning offer immense opportunities for an improved assessment of groundwater resources at the local to global levels. This systematic review documents the advancements in this field and evaluates the accuracy of various models, following the protocol developed by the Center for Evidence-Based Conservation. A total of 197 original peer-reviewed articles from 2010–2020 and from 28 countries that employ regression machine learning algorithms for groundwater monitoring or prediction are analyzed and their results are aggregated through a meta-analysis. Our analysis points to the capability of machine learning models to monitor/predict different characteristics of groundwater resources effectively and efficiently. Modeling the groundwater level is the most popular application of machine learning models, and the groundwater level in previous time steps is the most employed input data. The feed-forward artificial neural network is the most employed and accurate model, although the model performance does not exhibit a striking dependence on the model choice, but rather the information content of the input variables. Around 10–12 years of data are required to develop an acceptable machine learning model with a monthly temporal resolution. Finally, advances in machine and deep learning algorithms and computational advancements to merge them with physics-based models offer unprecedented opportunities to employ new information, e.g., InSAR data, for increased spatiotemporal resolution and accuracy of groundwater monitoring and prediction.

Accurate spatial estimation of precipitation is central to hydroclimatic studies. In this study, deterministic and probabilistic Inverse Distance Weighting (IDW) methods were utilized to provide spatial estimations of precipitation. To improve their accuracies, a modified version of IDW was developed by separation of IDW weights in XY-plane and Z direction. Moreover, four different weight functions were used to investigate how changing pattern of the weights relative to the distance and elevation affect the accuracy of the interpolation. Finally, four IDW variants were implemented for spatial modeling of precipitation in Central Plateau watershed located in Iran, using monthly data of 582 rain gauge stations during 2005 to 2015. Based on the statistics, modified IDW methods were about 8% more accurate than the standard IDW methods. Additionally, Correlated Triple Collocation (CTC) analysis showed that modified models have resulted in 12% smaller RMSEs in comparison with the standard models. Graphical Abstract

Flooding is one of the most dangerous natural hazards and is expected to occur more frequently due to the impacts of climate change. By altering the hydrologic characteristics of basins, urbanization exacerbates flooding risk. The Storm Water Management Model (SWMM) is a widely used urban hydrology model. Discretization of sub-catchments is a critical phase in constructing an SWMM model because it influences the effort of model construction and accuracy of the results. In the current review, we examine sub-catchment delineation in SWMM-based modeling by analyzing information from 43 journal papers encompassing 123 modeling scenarios. A novel categorization of sub-catchment delineation techniques with five primary and three secondary methods is proposed. Despite the hydrologic significance of external routing, where runoff is routed between sub-catchments, and internal routing, where runoff is routed from impervious to pervious portion of a sub-catchment or vice versa, such connections were considered in only 27% and 15% of the reviewed scenarios, respectively. Results also show that 69% of the scenarios used sub-catchments that were coarser than parcel (CP) resolution, but they were outperformed in calibration and validation by scenarios that used sub-catchments that were finer than or equal to parcel (FEP) resolution. Calibration was attempted in only 40% of the scenarios. Depression storage and surface roughness were commonly calibrated model parameters, but the tuned values suggest that in many cases calibration was achieved at the expense of physical plausibility. Future SWMM modeling applications can be improved by (i) delineating sub-catchments at FEP scale, (ii) utilizing SWMM routing options to better represent the physical flow pathways, and (iii) avoiding physically unrealistic calibration scenarios. Continued research on secondary delineation methods, which combine multiple primary methods, is recommended to offset the limitations of primary methods.