The objective of this research was the groundwater quality analysis in the Mygdonia basin in order to contribute to the protection of groundwater quality and the sustainable management of water resources in the study area, which is characterized by intensive agriculture activities. The interval of June and August-September 2014 was selected for measurements to be carried out. More specifically, during these two periods, in situ physicochemical parameters (pH, EC, T) were determined, showing in some cases high temperature values, which may be due to the mixing of the groundwater with geothermal fluids, and alongside water samples were collected from 17 and 26 boreholes, respectively. All samples were analyzed in the laboratory of Engineering Geology and Hydrogeology in the Aristotle University of Thessaloniki for major ions (Ca²⁺, Na⁺, Mg²⁺, K⁺, NO₃^-, SO₄^2-, Cl^-, HCO₃^-) regarding the qualitative groundwater status in the research area. The final conclusions and the spatial allocation for both ions and physicochemical parameters were depicted in maps, created in ArcGIS environment for both June and August-September. Furthermore, Piper and Durov diagrams were created by the use of AquaChem software, in order to define the hydrochemical type of groundwater. In general, the water quality is considered suitable for irrigation purposes. Finally, some recommendations are proposed to protect the quality and improve the sustainability of the groundwater resources of the basin.

Rainfall-Runoff (RR) modelling is a key component of water resource management. Because the relationship between rainfall and runoff is a complex natural event, it necessitates extensive computation. Runoff response is influenced by several interconnected phenomena, including rainfall intensity, geomorphology, climate, and landscape. The present study examines the rainfall-runoff-based model development by using artificial neural networks (ANNs) models in the Yerli sub-catchment of the upper Tapi basin for a period of 36 years, i.e., from 1981–2016. The created ANN models were capable of establishing the correlation between input and output data sets. The rainfall and runoff models that were built have been calibrated and validated. For predicting runoff, Feed Forward Back Propagation Neural Network (FFBPNN) and Cascade Forward Back Propagation Neural Network (CFBPNN) models are used. The efficacy of the model is evaluated by various measures such as mean square error (MSE), root mean square error (RMSE), and coefficient of correlation (R). With MSE, RMSE, and R values of 0.4982, 0.7056, and 0.96213, respectively, FFBPNN outperforms two networks with model architectures of 6-4-1 and Transig transfer function. Also, in this study, the Levenberg-Marquardt (LM), Bayesian Regularization (BR), and Conjugate Gradient Scaled (CGS) algorithms are used to train the ANN rainfall-runoff models. The results show that LM creates the most accurate model. It performs better than BR and CGS. The best model is the LM trained method with 30 neurons, which has MSE values of 0.7279, RMSE values of 0.8531, and R values of 0.95057. It is concluded that the constructed neural network model was capable of quite accurately predicting runoff for the Yerli sub-catchment.

Ecological stability and soil degradation are the main driving elements associated with climate change in terms of their strong impact on reducing or increasing the acceleration of climate change. Many studies have been devoted to examining the development of the ecological stability of a selected locality as well as assessing the intensity of erosion processes, but studies dealing with the relationship and connections between the development of ecological stability and the intensity of erosion are insufficient. Analyses of ecological stability and assessments of soil water erosion were performed for a catchment located in the Slovak Republic covering the years 1990, 2006, 2012, and 2018. The assessment of ecological stability was performed using several methods according to various approaches. The soil water erosion was derived using long-term simulations in an erosion model. A precipitation input data model was derived based on the Community Land Model (the CLM model). The aim of the article is to determine the relationship and connections between the ecological stability of the selected area and the intensity of its soil water erosion. The study also points to assessing changes in the development of ecological stability during the period selected together with changes in the intensity of the soil water erosion.

Availability of the portable water in sufficient quantity and standard quality is one of the basic requirements of civilization. The study area of the present work is the University of Kashmir where the existing Water Distribution System(WDS) has served its function for more than 60 years. The UOK is a growing campus and the existing WDS is unable to meet the demands of a growing population and suffers from excessive leakage losses and insufficient pressure heads at the nodes. In this work, it has been attempted to replace the existing WDS with an optimally designed WDS for the University of Kashmir, so that the objective functions of maximum efficiency of performance in terms of hydraulics and minimum capital cost in terms of pipe diameters are met. From the literature available EPANET has been found to perform hydraulic and water quality modeling with a fair degree of accuracy, ease of use, and free of cost. Therefore, for the hydraulic analysis in the current work, EPANET 2.0 has been used. To augment the assessment of the efficiency of performance of the WDS, the TPI (technical performance indices) for pressure and velocity at the hour of peak demand have been evaluated. Finally, the cost optimality of the network in terms of the pipe diameters has been validated by using the JAL-TANTRA web system developed by IIT Bombay.

The emergence of Machine learning (ML) algorithms have shown competency in a variety of fields and are growing in popularity in their application to geospatial science issues. Most recently and notably, ML algorithms have been applied to flood susceptibility (FS) mapping. Leveraging high power computing systems and existing ML algorithms with national datasets of Canada, this project has explored methods to create a national FS layer across a geographically large and diverse country with limited training data. First, approaches were considered on how to generate a map of FS for Canada, (i) national, which combined all training data into one model, and (ii) regional, where multiple models were created, based on regional similarities, and the results were mosaicked to generate a FS map. The second experiment explored the predictive capability of several ML algorithms across the geographically large and diverse landscape. Results indicate that the national approach provides a better prediction of FS, with 95.7 % of the test points, 91.5% of the pixels in the training sites, and 89.6% of the pixels across the country correctly predicted as flooded, compared to 65.5%, 80.6% and 75.6% respectively in the regional approach. ML models applied across the country found that support vector machine (svmRadial) and Neural Network (nnet) performed poorly in areas away from the training sites, while random forest (parRF) and Multivariate Adaptive Regression Spline (earth) performed better. A national ensemble model was ultimately selected as this blend of models compensated for the biases found in the individual models in geographic areas far removed training sites.

One of humanity's greatest and most significant challenges facing humanity is climate change mitigation. Despite the existing challenges in forecasting climate change effects on Earth, there is scientific agreement on its detrimental consequences. The climate change effects have been identified as adversely affecting ecosystems, soil erosion, reducing biodiversity, rising sea levels, extreme temperature changes, and global warming. Additionally, a significant impact is expected on food security, human health, energy consumption, and the economy. Forecasting air temperatures, in particular, has become an increasingly crucial climatic aspect in many fields, including tourism, energy, agriculture, industry, tourism and so on. There are several applications for air temperature forecasting, including forecasting cooling and energy consumption for residential buildings, controlling greenhouse temperatures adaptively, and predicting natural hazards. As a result, there is a need to reliably anticipate air temperature since they would assist in a planning horizon for constructing a business development, an energy policy, an insurance policy, and infrastructure upgrades when combined with the stud analysis of additional elements in the topic of interest. This research aims to develop a novel technique and explore the potential of new data intelligent models based on the optimally pruned extreme learning machine (OPELM) to predict the hourly air temperature in Quebec City, Canada. OPELM is a new development of the Extreme Learning Machine (ELM). It is a new non-tuned rapid algorithm for single-layer feed-forward neural networks. Different lags are selected as the input parameters, and various models are defined based on the time series concept. The performance of the OPELM is also compared with the ELM.

As a fundamental hydrological variable, rainfall contributes significantly to the land surface and atmospheric processes. There have been many applications for rainfall forecasting, such as drought monitoring and optimizing irrigation water management, flood forecasting and early warning systems, reservoir water level management for hydropower generation and more. Forecasting rainfall is challenging for meteorologists due to the variability in rainfall timing and quantity. As a result of its persistence and complexity, rainfall forecasting has piqued the interest of academics. Moreover, heavy rainfall can result in severe floods, causing significant fatalities and economic damage. Therefore, a more accurate rainfall forecast can aid in the formulation of appropriate measures that can reduce the risk of flooding and/or provide advanced early warning to people whose property may be within the flood damage zone. It is recognized that existing models use complex statistical models, which are often neither computationally nor technically feasible for many jurisdictions, hence downstream applications are unaffected by them. It is therefore being explored as a possible solution to these shortcomings to employ machine learning algorithms in combination with real-time monitoring technology. Accordingly, the current study presents a comparative analysis using different extreme learning machines suitable for specific downstream applications. Various models are defined for multi-step ahead forecasting of the rainfall in Quebec City, Canada. A simple mathematical formula derived from the current study could be applied to practical engineering problems.

To meet population growth, the excessive abstraction of water resources for irrigating water-intensive crops has become an increasing crisis in arid regions of Northwest China. This dynamic, typical of drylands, contains the perennial contradiction between development and desertification, which we also find within the United Nations Convention to Combat Desertification (UNCCD), where some interpret it as a developmental convention and others as a conservationist one. Land Degradation Neutrality (LDN) concept, has been set up as the primary tool to combat desertification by the UNCCD and is included in SDG 15.3. LDN refers to a state of zero net land degradation, where “the amount and quality of land resources necessary to support ecosystem functions and services and enhance food security remain stable or increase within specified temporal and spatial scales and ecosystems”.

Under the LDN framework, we apply two of its main pillars, prevention, and land planning. The aim is to understand the underlying biophysical and socio-economic mechanisms of oasis expansion in NW China, a phenomenon known as oasification. The objective is to detect under what conditions oasification tackles desertification and when it triggers land and economic degradation. From this knowledge, it will be possible to propose guidelines of action to balance land use and comply with LDN.

Drought is a natural phenomenon that is controlled by different factors such as natural climate, catchment controls, and in many worldwide regions it is now driven by human activities (i.e. reservoirs, irrigation, groundwater abstractions). Reservoirs initially ensure water availability and cope with drought, especially in semi-arid regions; however, this human modification to the environment may lead both positive and negative effects over the hydrological cycle. This involves a better understanding of hydrological processes and incorporating human interactions to improve drought management within coupled human-natural systems

This study is focused on a strongly irrigated area located in the northeast of the Iberian Peninsula, the northern part of the Aragón and Cataluña district supplied by the Barasona reservoir. We implemented a simple water management model to simulate the reservoir operation (human-influenced scenario) and examine the contribution of human activities, associated with irrigation, on the water budget and drought propagation. For this purpose, we use simulation performed by the SASER model which provided a natural scenario (without human influence) to contrast with the human-influenced scenario.

Here, we explore the linkages between agricultural drought, associated with evapotranspiration, and hydrological drought, thus we applied standardized indices to identify droughts, then we compare them to each other and assess changes induced by human activities. The first results demonstrated satisfactory performance to simulate reservoir storage and outflows against observed data, KGE values of 0.4 and 0.82, respectively. The human modifications modulate the hydrological response of the catchment, and alter the intensity of hydrological drought, while human activities reduced the intensity of agricultural droughts.

The combination formed by the intense transformation of the territory and climate change in the Spanish Mediterranean basins has configured an explosive cocktail from the point of view of the risk of flooding in these areas. Climate change is making intense DANA (Spanish acronym of depresión aislada en niveles altos) type rains a more frequent phenomenon in Mediterranean basins. The vulnerability of the coastal territory in these areas as a consequence of the DANA phenomenon now requires the authorities to implement new strategies and policies that reduce the significant economic damage and loss of human life suffered in recent years. However, correlating these two phenomena with the increased risk of flooding is a difficult problem to diagnose, and even more difficult to solve. For this reason, a structured GIS methodology is proposed based on a geostatistical indicators analysis that correlates the transformation of the territory with the increase in vulnerability due to floods. The assessment of the validity of a proposed methodology is applied to the case study of the Campo de Cartagena watershed located around the Mar Menor, a Mediterranean coastal lagoon in Southeastern Spain. This area has suffered three catastrophic floods derived from the DANA phenomenon between 2016 and 2019.The results show that apart from the effects derived from climate change, the real issue that amplifies the damage caused by floods is the diffuse anthropization process in the area, which has caused the loss of the natural hydrographic network that traditionally existed in the basin. Based on the results obtained in this analysis, risk mitigation actions will be proposed through the improvement of land management by using of nature-based solutions.