Groundwater level prediction using an improved SVR model integrated with hybrid particle swarm optimization and firefly algorithm

Cleaner Water, Journal Year: 2024, Volume and Issue: 1, P. 100003 - 100003

Published: Jan. 21, 2024

The demand for water resources has increased due to rapid increase of metropolitan areas brought on by growth in population and industrialisation. In addition, the groundwater recharge is being afftected shifting land use pattern caused urban development. Using precise trustworthy estimates level vital sustainable management face changing climatic circumstances. this context, machine learning (ML) methods offer a new promising approach accurately forecasting long-term changes (GWL) without computational effort developing comprehensive flow model. order simulate GWL, five data-driven (DD) models, including hybridization support vector regression (SVR) with two optimisation algorithms i.e., firefly algorithm particle swarm (FFAPSO), SVR-FFA, SVR-PSO, SVR Multilayer perception (MLP), have been examined present study. Spatial clustering was utilised choose four observation wells within Cuttack district study assess levels. Six scenarios were created incorporating numerous variables, such as GWL previous months, evapotranspiration, temperature, precipitation, river discharge. goal identify variables that most efficient predicting GWL. SVR-FFAPSO model performs best Khuntuni station, according quantitative analysis correlation coefficient (R) = 0.9978, Nash–Sutcliffe efficiency (NSE) 0.9933, mean absolute error (MAE) 0.00025 (m), root squared (RMSE) 0.00775 (m) during training phase. It advised monitoring network data collecting system are strengthen India ensuring effective modelling resources.

Language: Английский

---

Revolutionizing Groundwater Management with Hybrid AI Models: A Practical Review

Water, Journal Year: 2023, Volume and Issue: 15(9), P. 1750 - 1750

Published: May 2, 2023

Developing precise soft computing methods for groundwater management, which includes quality and quantity, is crucial improving water resources planning management. In the past 20 years, significant progress has been made in management using hybrid machine learning (ML) models as artificial intelligence (AI). Although various review articles have reported advances this field, existing literature must cover ML. This article aims to understand current state-of-the-art ML used achievements domain. It most cited employed from 2009 2022. summarises reviewed papers, highlighting their strengths weaknesses, performance criteria employed, highly identified. worth noting that accuracy was significantly enhanced, resulting a substantial improvement demonstrating robust outcome. Additionally, outlines recommendations future research directions enhance of including prediction related knowledge.

Language: Английский

---

Predicting Future Land Use Utilizing Economic and Land Surface Parameters with ANN and Markov Chain Models

Earth, Journal Year: 2023, Volume and Issue: 4(3), P. 728 - 751

Published: Sept. 15, 2023

The main aim of this study is to comprehensively analyze the dynamics land use and cover (LULC) changes in Bathinda region Punjab, India, encompassing historical, current, future trends. To forecast LULC, Cellular Automaton–Markov Chain (CA) based on artificial neural network (ANN) concepts was used using cartographic variables such as environmental, economic, cultural. For segmenting a combination ML models, support vector machine (SVM) Maximum Likelihood Classifier (MLC). empirical nature, it employs quantitative analyses shed light LULC variations through time. result indicates that barren expected shrink from 55.2 km2 1990 5.6 2050, signifying better management or increasing human activity. Vegetative expanses, other hand, are rise 81.3 205.6 reflecting balance between urbanization ecological conservation. Agricultural fields increase 2597.4 2859.6 2020 before stabilizing at 2898.4 2050. Water landscapes 13.4 providing possible issues for water resources. Wetland regions decrease, thus complicating irrigation groundwater reservoir sustainability. These findings confirmed by strong statistical indices, with study’s high kappa coefficients Kno (0.97), Kstandard (0.95), Klocation (0.97) indicating reasonable level accuracy CA prediction. From F1 score, significant issue found MLC vegetation, resolved SVM classification. can be inform policy plans sustainable development beyond.

Language: Английский

---

Enhancing Flood Susceptibility Modeling: a Hybrid Deep Neural Network with Statistical Learning Algorithms for Predicting Flood Prone Areas

Water Resources Management, Journal Year: 2024, Volume and Issue: 38(8), P. 2687 - 2710

Published: March 18, 2024

Language: Английский

---

Prediction of groundwater level variations using deep learning methods and GMS numerical model

Earth Science Informatics, Journal Year: 2023, Volume and Issue: 16(4), P. 3227 - 3241

Published: Aug. 31, 2023

Language: Английский

---

Modeling of wetlands storage instabilities using an optimized slffn machine learning using evolutionary computation considering the RROC analysis and PDF techniques

Earth Science Informatics, Journal Year: 2024, Volume and Issue: 17(4), P. 3137 - 3148

Published: May 25, 2024

Language: Английский

---

Soft computing assessment of current and future groundwater resources under CMIP6 scenarios in northwestern Iran

Agricultural Water Management, Journal Year: 2023, Volume and Issue: 285, P. 108369 - 108369

Published: May 26, 2023

Excessive use of water resources in combination with climate change threaten to significantly reduce groundwater arid and semiarid regions. We studied the effects on level for important Dehgolan Aquifer northwestern Iran. The this aquifer has dropped by about 35 m during last 30 years. Soft computing techniques were used together projections three methodological steps estimate drop 2045. Firstly, MODFLOW was simulate flow movement. Secondly, simulation results, support vector regression (SVR), least-squares SVR (LSSVR) machine learning models predict levels future 20-year period (2026–2045). whale optimization algorithm (WOA) improve prediction results optimizing parameters. Thirdly, CMIP6 (ACCESS-CM2, BCC-CSM2-MR, CMCC-ESM2) changes precipitation (2026–2045) using SSP 2.6 8.5 scenarios. showed that MODFLOW-LSSVR model predicted more accurately than MODFLOW-SVR MODFLOW-SVR-WOA. calculation scenario containing previous month level, monthly withdrawal, had highest performance predicting root mean square error (RMSE), absolute percentage (MAPE), Nash Sutcliffe efficiency (NSE) equal 0.305 m, 0.014 0.998, respectively. indicate may decrease (about 6% compared reference 1987–2005). This decrease, along continuation current will cause a 36 (during 28 years) (1.3 per year). reveal could be reduced 12 adopting 25% reduction withdrawal. findings show necessity providing suitable management approach prevent exhaustion due withdrawal situation region.

Language: Английский

---

Unraveling aquifer dynamics: Time series evaluation for informed groundwater management

Groundwater for Sustainable Development, Journal Year: 2024, Volume and Issue: 25, P. 101174 - 101174

Published: April 10, 2024

Language: Английский

---

Automatic Detection of Subglacial Water Bodies in the AGAP Region, East Antarctica, Based on Short-Time Fourier Transform

Remote Sensing, Journal Year: 2023, Volume and Issue: 15(2), P. 363 - 363

Published: Jan. 6, 2023

Subglacial water bodies are critical components in analyzing the instability of Antarctic ice sheet. Their detection and identification normally rely on geophysical remote sensing methods such as airborne radar echo sounding (RES), ground seismic, satellite/airborne altimetry gravity surveys. In particular, RES surveys able to detect basal terrain with a relatively high accuracy that can assist mapping subglacial hydrology systems. Traditional processing for mostly their brightness radargrams hydraulic flatness. this study, we propose an automatic method main objective differentiate materials by quantitatively evaluating shape A-scope waveform near interface radargrams, which has been seldom investigated. We develop algorithm mainly based traditional short-time Fourier transform (STFT) quantify radargrams. Specifically, appropriate window width applied peak each radargram, STFT shows distinct contrasting frequency responses at ice-water ice-rock interface, is largely dependent upon different reflection characteristics interface. apply 882 collected Antarctic’s Gamburtsev Province (AGAP) East Antarctica. There 8822 identified A-scopes calculated value larger than set threshold, out overall 1,515,065 valid these Although only takes 0.58% population, they show exceptionally continuous distribution represent bodies. Through comprehensive comparison existing inventories lakes, successfully verify validity advantages our identifying using probability other theoretically highest along-track resolution. The signature obtained proposed joint time–frequency analysis provides new corridor investigation be further expanded multivariable deep learning approaches englacial material characterization, well mapping.

Language: Английский

---

Towards Cleaner Ports: Predictive Modeling of Sulfur Dioxide Shipping Emissions in Maritime Facilities Using Machine Learning

Sustainability, Journal Year: 2023, Volume and Issue: 15(16), P. 12171 - 12171

Published: Aug. 9, 2023

Maritime ports play a pivotal role in fostering the growth of domestic and international trade economies. As continue to expand size capacity, impact their operations on air quality climate change becomes increasingly significant. While nearby regions may experience economic benefits, there are significant concerns regarding emission atmospheric pollutants, which have adverse effects both human health change. Predictive modeling port emissions can serve as valuable tool identifying areas concern, evaluating effectiveness reduction strategies, promoting sustainable development within ports. The primary objective this research is utilize machine learning frameworks estimate SO2 from ships during various activities, including hoteling, maneuvering, cruising. By employing these models, we aim gain insights into patterns explore strategies mitigate impact. Through our analysis, identified most effective models for estimating emissions. AutoML TPOT framework emerges top-performing model, followed by Non-Linear Regression with interaction effects. On other hand, Linear exhibited lowest performance among evaluated. advanced techniques, contribute body knowledge surrounding foster practices maritime industry.

Language: Английский

---