Spatiotemporal assessment of sustainable groundwater management using process-based and remote sensing indices: A novel approach.

This study aims to develop a process-based method for evaluating groundwater sustainability and use the results in an archetypal analysis to fundamentally frame and understand sustainable development interactions in a river basin scale and sub-basin resolution. This method was applied in the Tashk-Bakhtegan-Maharloo (TBM) basin of Iran between 2003 and 2018; anthropogenic and natural factors were considered. With its 31 aquifers in 27 sub-basins, the TBM basin has repeatedly suffered severe droughts and water shortages over the past half a century, highlighting the importance of sustainable groundwater management. This study quantified anthropogenic and natural factors affecting groundwater dynamics to address sustainability and defined representative and relative indices, including climatological and drought conditions, vegetation cover, land cover, and population, to assess groundwater sustainability (GWS). Relative indices, prepared using measured data and remote sensing analysis, were chosen to explain groundwater-related situations, whereas representative indices, such as groundwater level and total dissolved solids, were used to explain the groundwater situation. GWS was spatially monitored using a couple-indicator trend-line slope comparison method to analyze process-based indices. Then, archetypal interaction patterns and their drivers in the groundwater system were investigated using results from process-based indices analyses results. The results showed that the TBM basin has moved towards unsustainable levels because of drought, increased irrigated croplands, unbalanced development of the sub-basins up- and downstream in the river's path, and over-exploitation of groundwater. These findings indicate that a deeper understanding of groundwater problems and stakeholder associations is required in order to adapt to the changing groundwater conditions.

Urban expansion simulation and scenario prediction using cellular automata: comparison between individual and multiple influencing factors.

Quantifying the contribution of driving factors is crucial to urban expansion modeling based on cellular automata (CA). The objective of this study is to compare individual-factor-based (IFB) models and multi-factor-based (MFB) models as well as examine the impacts of each factor on future urban scenarios. We quantified the contribution of driving factors using a generalized additive model (GAM), and calibrated six IFB-DE-CA models and fifteen MFB-DE-CA models using a differential evolution (DE) algorithm. The six IFB-DE-CA models and five MFB-DE-CA models were selected to simulate the 2005-2015 urban expansion of Hangzhou, China, and all IFB-DE-CA models were applied to project future urban scenarios out to the year 2030. Our results show that terrain (DEM) and population density (POP) are the two most influential factors affecting urban expansion of Hangzhou, indicating the dominance of biophysical and demographic drivers. All DE-CA models produced defensible simulations for 2015, with overall accuracy exceeding 89%. The IFB-DE-CA models based on DEM and POP outperformed some MFB-DE-CA models, suggesting that multiple factors are not necessarily more effective than a single factor in simulating present urban patterns. The future scenarios produced by the IFB-DE-CA models are substantially shaped by the corresponding factors. These scenarios can inform urban modelers and policy-makers as to how Hangzhou city will evolve if the corresponding factors are individually focused. This study improves our understanding of the effects of driving factors on urban expansion and future scenarios when incorporating the factors separately.

Novel forecasting approaches using combination of machine learning and statistical models for flood susceptibility mapping.

In this research, eight individual machine learning and statistical models are implemented and compared, and based on their results, seven ensemble models for flood susceptibility assessment are introduced. The individual models included artificial neural networks, classification and regression trees, flexible discriminant analysis, generalized linear model, generalized additive model, boosted regression trees, multivariate adaptive regression splines, and maximum entropy, and the ensemble models were Ensemble Model committee averaging (EMca), Ensemble Model confidence interval Inferior (EMciInf), Ensemble Model confidence interval Superior (EMciSup), Ensemble Model to estimate the coefficient of variation (EMcv), Ensemble Model to estimate the mean (EMmean), Ensemble Model to estimate the median (EMmedian), and Ensemble Model based on weighted mean (EMwmean). The data set covered 201 flood events in the Haraz watershed (Mazandaran province in Iran) and 10,000 randomly selected non-occurrence points. Among the individual models, the Area Under the Receiver Operating Characteristic (AUROC), which showed the highest value, belonged to boosted regression trees (0.975) and the lowest value was recorded for generalized linear model (0.642). On the other hand, the proposed EMmedian resulted in the highest accuracy (0.976) among all models. In spite of the outstanding performance of some models, nevertheless, variability among the prediction of individual models was considerable. Therefore, to reduce uncertainty, creating more generalizable, more stable, and less sensitive models, ensemble forecasting approaches and in particular the EMmedian is recommended for flood susceptibility assessment.

Transition index maps for urban growth simulation: application of artificial neural networks, weight of evidence and fuzzy multi-criteria evaluation.

Transition index maps (TIMs) are key products in urban growth simulation models. However, their operationalization is still conflicting. Our aim was to compare the prediction accuracy of three TIM-based spatially explicit land cover change (LCC) models in the mega city of Mumbai, India. These LCC models include two data-driven approaches, namely artificial neural networks (ANNs) and weight of evidence (WOE), and one knowledge-based approach which integrates an analytical hierarchical process with fuzzy membership functions (FAHP). Using the relative operating characteristics (ROC), the performance of these three LCC models were evaluated. The results showed 85%, 75%, and 73% accuracy for the ANN, FAHP, and WOE. The ANN was clearly superior compared to the other LCC models when simulating urban growth for the year 2010; hence, ANN was used to predict urban growth for 2020 and 2030. Projected urban growth maps were assessed using statistical measures, including figure of merit, average spatial distance deviation, producer accuracy, and overall accuracy. Based on our findings, we recomend ANNs as an and accurate method for simulating future patterns of urban growth.