Arabian Journal of Geosciences最新文献

{"title":"Comparative evaluation of machine learning classifiers and predictive modelling of human-induced land degradation in Aowin District, Ghana","authors":"Michael Stanley Peprah,&nbsp;Abdul-Wadood Moomen,&nbsp;Ambrose Derzu,&nbsp;Cynthia Awewura Abavare","doi":"10.1007/s12517-026-12605-1","DOIUrl":"10.1007/s12517-026-12605-1","url":null,"abstract":"<div><p>Human-induced land degradation associated with mining, agricultural expansion, logging, and settlement development is transforming tropical rainforest landscapes in Ghana. This regional case study assessed historical land-cover transitions and explored potential future degradation trajectories in the Aowin District using established machine-learning and land-change modelling approaches. Landsat imagery for 2005, 2015, and 2025 provided a temporally consistent 20-year record, although its 30 m spatial resolution limits the detection of fine-scale disturbances. Seven supervised classifiers Support Vector Machine, Random Forest, K-Nearest Neighbour, Extreme Gradient Boosting, Classification and Regression Tree, Minimum Distance, and Naïve Bayes were evaluated using 360 stratified reference samples and multiple accuracy metrics. Support Vector Machine achieved the highest observed performance, with an overall accuracy of 81.7% and a Kappa coefficient of 0.773; however, differences among classifiers were interpreted descriptively rather than as statistically significant. Between 2005 and 2025, dense vegetation declined by 16.49%, while bare land and built-up areas increased by 1,672.90% and 96.40%, respectively; mapped mined land expanded from 0.10 to 5.26 km². Land-cover transitions were interpreted as degradation, persistence, or potential recovery rather than assuming that all change represented degradation. Conditional scenarios for 2035, 2045, and 2055 were generated using an MLP–CA–Markov framework with elevation, slope, land surface temperature, soil moisture index, distance to roads, and distance to rivers. Hindcasting of the known 2025 landscape was used to evaluate the predictive framework. Under continuation of the historical relationships represented in the model, dense vegetation is projected to decline by 20.35% between 2025 and 2055, while bare land and built-up areas are projected to increase. These scenarios provide regional evidence to support environmental monitoring, restoration prioritisation, and land-use planning.</p></div>","PeriodicalId":476,"journal":{"name":"Arabian Journal of Geosciences","volume":"19 10","pages":""},"PeriodicalIF":1.827,"publicationDate":"2026-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149132064","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Flood hazard characterisation in a data-scarce region: a case study of the Mbaka catchment, Southern Tanzania","authors":"Abubakari Amanzi,&nbsp;Lina Friberg Njau,&nbsp;Eunice Makungu","doi":"10.1007/s12517-026-12602-4","DOIUrl":"10.1007/s12517-026-12602-4","url":null,"abstract":"<div><p>Rating curve saturation during high flows reduces the accuracy of streamflow estimation from various gauging stations in sub-Saharan Africa, especially for flood hazard assessment. This study investigates a stage-based HBV model to overcome censoring of Mbaka at Mwaya gauging station by reconstruction of a 51-year flood record and ranking of historical floods for this catchment. Simulated streamflow was converted to stage using a geometric Manning-Strickler relationship derived from measured channel geometry. The model showed a satisfactory performance with NSh = 0.78 and NSlog(h) = 0.82 during calibration and NSh = 0.74 and NSlog(h) = 0.79 during validation. Sentinel-1 SAR analysis found five flood events between the year 2015 and 2025 with flood inundation extents ranging from 18.5 to 44.0 km². A Spearman rank correlation (ρ = 0.70, <i>p</i> = 0.188, <i>n</i> = 5) between simulated peak streamflow and Synthetic Aperture Radar (SAR)-derived flood extent indicates a consistent rank-order relationship between independent datasets though the correlation is not statistically significant due limited sample size. Height Above the Nearest Drainage (HAND)-based zonation shows 87% to 91% of inundation occurs within HAND &lt; 5 m class, representing topographic control. This method used can be directly adopted to other data-scarce catchments to address censored gauging station records.</p></div>","PeriodicalId":476,"journal":{"name":"Arabian Journal of Geosciences","volume":"19 10","pages":""},"PeriodicalIF":1.827,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149132246","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Flood hazard assessment using attraction model-enhanced DEMs, morphometric features, and remote sensing indices","authors":"Marzieh Mokarram,&nbsp;Abdol Rassoul Zarei,&nbsp;Mohammad Mehdi Ghasemi","doi":"10.1007/s12517-026-12599-w","DOIUrl":"10.1007/s12517-026-12599-w","url":null,"abstract":"<div><p>Understanding the morphological characteristics of river channels through high-spatial-resolution imagery is essential for accurate watershed delineation and for improving flood management in floodplain environments. This study applied the Attraction Model (ATM) to refine the spatial resolution of a Digital Elevation Model (DEM) for more precise river channel extraction, focusing on flood-prone regions of Jahrom, Fars Province, Iran. The ATM was implemented using three scale factors (2, 3, and 4). Two neighborhood configurations, including touching and quadrant methods, were evaluated separately. Enhanced DEMs were generated using these approaches. These DEMs were then used to extract river channels. Subsequently, morphometric indicators such as curvature coefficient and central angle of meander rotation were calculated and compared with those derived from the original 10 m DEM. In parallel, several remote sensing (RS) indices, including the Automated Water Extraction Index (AWEI), Soil Water Index (SWI), Water Ratio Index (WRI), Modified Normalized Difference Water Index (MNDWI), and Normalized Difference Vegetation Index (NDVI), were employed to identify and map flood-prone areas. Findings showed that the spatial ATM with a scale factor of 2 produced the highest-quality results, with the touching neighborhood method outperforming the quadrant method. Meander rotation analysis revealed that 58.2% of the study area is highly susceptible to flooding. Among the RS indices tested, SWI demonstrated the greatest accuracy (R² = 0.88) in detecting vulnerable zones. Approximately 70% of these flood-prone zones were located in the southern and northwestern sectors of the study area. The riverbed was found to be narrow and steep, limiting water conveyance and increasing flood hazards. To reduce flood risk, proposed measures include channel widening and dredging, bank stabilization along meanders, planting water-absorbent vegetation, and improving overall channel capacity.</p></div>","PeriodicalId":476,"journal":{"name":"Arabian Journal of Geosciences","volume":"19 10","pages":""},"PeriodicalIF":1.827,"publicationDate":"2026-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148986818","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Assessment of suitable land for surface irrigation development in an Ardy watershed, Upper Blue Nile Basin, Ethiopia","authors":"Abebe Fentahun,&nbsp;Arega Mulu,&nbsp;Samuel Berihun Kassa,&nbsp;Kassahun Birhanu","doi":"10.1007/s12517-026-12589-y","DOIUrl":"10.1007/s12517-026-12589-y","url":null,"abstract":"<div><p>Despite its abundant water and land resources, Ethiopia has not fully harnessed its potential for irrigation development. This study assessed suitable land for surface irrigation in the Ardy watershed, Ethiopia, with the goal of utilizing the country’s abundant yet underexploited water resources to increase agricultural productivity and sustainability. The data used in this study included soil, land use/land cover, meteorological, and streamflow data. The methodology integrates the Food and Agriculture Organization (FAO) land evaluation framework with Geographic Information Systems (GIS) and multicriteria analysis (Analytic Hierarchy Process, AHP) techniques. Land suitability was determined using factors such as slope, soil, land use/land cover, and proximity to water sources, with slope identified as the most influential parameter (47.09% weight). The Soil and Water Assessment Tool (SWAT) model was used to analyze surface water potential and demonstrated strong reliability during calibration (R² = 0.90, NSE = 0.86, PBIAS = -18.9%) and validation (R² = 0.85, NSE = 0.84, PBIAS = -14.9%). The results indicated that 99% (21,509 ha) of the watershed is suitable for surface irrigation, with 53% classified as moderately suitable. However, CROPWAT 8.0 modeling revealed that during peak irrigation demand periods (December, January, and February), the gross irrigation water requirement for the total suitable land exceeds the available river flow. On the basis of minimum available flows, the critical command area that can be reliably irrigated ranges from 918 ha (maize) to 1,057 ha (potato) across the subbasins. These findings provide policymakers with valuable guidance for developing targeted and effective management strategies to ensure sustainable resource use in the Ardy watershed.</p></div>","PeriodicalId":476,"journal":{"name":"Arabian Journal of Geosciences","volume":"19 10","pages":""},"PeriodicalIF":1.827,"publicationDate":"2026-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s12517-026-12589-y.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148986511","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Carbonate polymorph evolution in historic lime–clay geomaterials from Lamayuru Monastery, Ladakh: evidence from a cold-arid Himalayan environment","authors":"Deepak Anand Gupta,&nbsp;Manager Rajdeo Singh,&nbsp;Madhuri Sawant","doi":"10.1007/s12517-026-12601-5","DOIUrl":"10.1007/s12517-026-12601-5","url":null,"abstract":"<div><p>Lime–clay plasters exposed to cold–arid Himalayan environments experience low temperatures, pronounced thermal variability, intermittent moisture availability, freeze–thaw cycling, and reduced atmospheric CO₂ partial pressure, conditions capable of influencing carbonate phase evolution in historic lime-based materials. This study aims to characterize the mineralogical and microstructural composition of historic lime–clay plasters from Lamayuru Monastery (Ladakh, India), with particular emphasis on the occurrence of CaCO₃ polymorphs and their variability within the plaster matrix, using integrated optical, mineralogical, thermal, and microstructural characterization. Optical microscopy reveals a stratified plaster system comprising distinct lime–clay and clay-rich layers with preserved interfacial continuity. X-ray diffraction and FTIR analyses identify coexistence of calcite, aragonite, and vaterite, indicating persistence of multiple CaCO₃ polymorphs within the plaster matrix. Variability in relative peak prominence among samples suggests heterogeneous carbonate distribution associated with compositionally variable microenvironments. Thermogravimetric analysis identifies a principal carbonate decomposition event near ~ 710 °C, lower than temperatures commonly reported for well-crystallised geological calcite, consistent with heterogeneous carbonate-bearing domains within clay-rich and organic-containing microenvironments. SEM–EDS observations demonstrate a heterogeneous lime–clay–silicate architecture comprising distributed carbonate phases, aluminosilicate materials, and pore-rich microstructural domains. The combined observations indicate that the Lamayuru plasters represent compositionally heterogeneous geomaterials shaped by interaction among carbonate phases, clay minerals, organic constituents, and prolonged environmental exposure. The study provides integrated archaeometric data on carbonate polymorphism and microstructural variability in historic Himalayan lime-based plasters and contributes to a broader understanding of environmentally exposed lime-based geomaterials from cold-region environments worldwide.</p></div>","PeriodicalId":476,"journal":{"name":"Arabian Journal of Geosciences","volume":"19 10","pages":""},"PeriodicalIF":1.827,"publicationDate":"2026-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148986450","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Jurassic–early cretaceous alkaline seamount volcanism in the batain nappes, eastern oman: petrology, geochemistry, and tectonic significance of the Ruwaydah volcanic unit","authors":"Sobhi Nasir","doi":"10.1007/s12517-026-12600-6","DOIUrl":"10.1007/s12517-026-12600-6","url":null,"abstract":"<div><p>The Ruwaydah Unit of the Batain Nappe, eastern Oman, preserves a complete record of Jurassic–Early Cretaceous intraplate seamount volcanism within the Eastern Oman Ophiolites. The Late Jurassic-Early Cretaceous unit comprises a volcanic-sedimentary sequence consisting of alkaline basaltic to trachytic lavas, pillow lavas, pillow breccias, pyroclastic deposits, volcanic breccias, radiolarian cherts, pelagic limestones, crinoidal limestones, Ammonitico Rosso facies, and shallow-water carbonate deposits. Field relationships indicate the development of submarine volcanic structures that evolved through repeated volcanic and sedimentary cycles within the Proto-Indian Ocean. The rocks display aphyric, porphyritic, pilotaxitic, and doleritic textures. Geochemically, the Ruwaydah volcanic suite exhibits alkaline affinities characterized by enrichment in high-field-strength elements (HFSE) and light rare earth elements (LREE), positive Nb-Ta anomalies, and strongly fractionated chondrite-normalized REE patterns. The volcanic rocks define a coherent differentiation trend from alkali basalt to trachyte through fractional crystallization of olivine, clinopyroxene, Fe-Ti oxides, apatite, and feldspar. Trace-element discrimination diagrams consistently place the Ruwaydah volcanic rocks within the ocean island basalt (OIB) and within-plate basalt fields. Rare earth element modelling indicates derivation from approximately 2–6% of partial melting of an enriched garnet-bearing lherzolite mantle source, with minor contributions from the garnet-spinel transition zone. The OIB-like geochemical characteristics, together with the absence of subduction-related signatures, indicate generation within an intraplate oceanic setting. The Ruwaydah volcanic complex formed during an Early Cretaceous tectono-magmatic event that affected the Eastern Oman Ophiolites approximately 10–15 million years after formation of the Jurassic oceanic crust. Volcanism was associated with lithospheric extension and localized asthenospheric upwelling, possibly influenced by the peripheral effects of the Marion hotspot in the southwestern Proto-Indian Ocean. However, the geochemistry demonstrates enriched intraplate mantle melting more securely than it demonstrates a particular plume source. The Ruwaydah Unit therefore represents a well-preserved example of oceanic seamount volcanism developed on mature oceanic lithosphere and provides insights into intraplate magmatism, mantle dynamics, and seamount evolution within the Proto-Indian Ocean prior to obduction of the Eastern Oman Ophiolites onto the Arabian continental margin.</p></div>","PeriodicalId":476,"journal":{"name":"Arabian Journal of Geosciences","volume":"19 10","pages":""},"PeriodicalIF":1.827,"publicationDate":"2026-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148986449","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Geospatial-based evaluation of energy generation from municipal solid waste and green house gas mitigation potential","authors":"Emmanuel C. Chukwuma,&nbsp;John O. Ojediran,&nbsp;Joseph I. Ubah,&nbsp;Chukwuemeka O. Umobi,&nbsp;Toyin F. Adeniyi,&nbsp;Emmanuel C. Nwanna","doi":"10.1007/s12517-026-12598-x","DOIUrl":"10.1007/s12517-026-12598-x","url":null,"abstract":"<div><p>Socio-economic well-being of people depends largely on access to clean energy sources. The current situation in most developing countries indicates that municipal waste is not properly harnessed. Exploitation of the organic portion of Municipal Solid Waste (MSW_org) remains a valuable energy resource to achieve sustainable clean energy. This study therefore seeks to estimate the potential energy resource that is untapped, and the possible mitigation estimate of Green House Gas (GHG) emissions resulting from the failure of utilizing MSW_org. The result of the study shows that a maximum value of about 9489 tons of MSW_org is generated from Lagos State daily, while a minimum value of 1121 tons is generated from Kwara State. On the electrical energy potential, about 1234 MW, 10,720 MW, and 5005 MW of electricity can be generated in Ekiti, Lagos and Oyo states respectively. On the GHG emission, the estimated CO₂ emission from the 7 states ranges from 47,057,161 m³ to 408,658,874 m³, the maximum CO₂ emission is from Lagos state, while the minimum value is from Kwara State. The methane emission potential shows that Kwara state has the least annual Methane emission of 30,280,260 m³, while Lagos state has the highest annual methane emission potential of about 262,963,101 m^3, followed by Oyo state with about 122,780,404 m³. The annual methane emission potential of other states are in the value of 51,217,557 m³, 69,045,992 m³, 55,638,450 m³, and 47,739,582 m³ for Ekiti, Ogun, Ondo and Osun states respectively. It is suggested that the adoption of a circular economy in the states will eliminate deficiencies in energy supply, with attendant environmental advantages of reduced climate hazards. (It is also suggested that a regional renewable energy program will help to curb the scorching effects of energy poverty in the region.)</p></div>","PeriodicalId":476,"journal":{"name":"Arabian Journal of Geosciences","volume":"19 10","pages":""},"PeriodicalIF":1.827,"publicationDate":"2026-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148986500","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Factors controlling production heterogeneity in the Eocene fractured limestone reservoir, Northwest October Field, Gulf of Suez, Egypt","authors":"Ghada Farrag,&nbsp;Mohamed Ali,&nbsp;A. Hamid Gewaily","doi":"10.1007/s12517-026-12594-1","DOIUrl":"10.1007/s12517-026-12594-1","url":null,"abstract":"&lt;div&gt;&lt;p&gt;The Eocene Formation represents one of the most prolific hydrocarbon reservoirs in the Gulf of Suez, with the October, Northwest October, and North Amer fields serving as prime examples in the central region. Although the primary oil productivity in these fields relies heavily on this Eocene reservoir, its production dynamics remain a subject of ongoing debate among researchers. Evaluating and sustaining Eocene economic oil production remains a significant geoscience challenge due to the intense spatial heterogeneity of its sub-seismic fracture networks, localized diagenetic overprints, and complex block-faulted compartmentalization. These parameters result in severe localized variations in initial flow efficiency and erratic depletion pathways that cannot be accurately predicted using standard structural or clastic logging workflows. In the studied field, the Eocene Formation consists of tight, fractured limestone characterized by significant production heterogeneity. This study focuses on evaluating the various factors contributing to this production heterogeneity within the Eocene tight fractured limestone. This study utilizes a newly reprocessed, three-dimensional pre-stack depth migration (PSDM) seismic survey integrated with dipmeter data, conventional well logs, oil-based mud formation micro-imager (OBMI) logs, and physical core samples from field wells. An integrated workflow was executed by constructing a NW–SE structural-petrophysical correlation cross-section, modeling fault networks, analyzing pressure-depth plots to determine fluid gradients, and conducting quantitative fracture-density and core microfacies characterizations. Structural mapping reveals that primary Oligo-Miocene clysmic rift faults (NW–SE) were dissected by Late Miocene cross-faults (NE–SW), causing structural omissions of 100 to 500 ft at the top of the Eocene. Pressure data prove intense compartment isolation with highly variable Oil-Down-To (ODT) levels ranging from − 6,400 to -6,990 ft TVDSS. OBMI logs show that conductive, open fractures strike exclusively NE–SW parallel to cross-fault FC1, with fracture densities exceeding 1.0 fractures/meter near the fault zone, which artificially drives matrix porosity up to 22–24%. Conversely, areas away from the cross-fault display purely resistive, mineralized fractures and a low fracture density (~ 0.2 fractures/meter), restricting reservoir storage to an unenhanced 18–20% depositional matrix porosity. Production analysis demonstrates that open-hole completion is the optimum strategy for these fractured carbonates, as validated by an elite Productivity Index (PI) of 2.5 bbl/d/psi in the ENWO-1x well. Where cased-hole completions are structurally mandatory, deploying non-explosive hydro-mechanical perforation effectively bypasses the 3–5 ft near-wellbore critical matrix damage zone with a 5.41-ft penetration depth, completely eliminating explosive-induced skin effect and doubling the well Productivity Index (1.0 bb","PeriodicalId":476,"journal":{"name":"Arabian Journal of Geosciences","volume":"19 10","pages":""},"PeriodicalIF":1.827,"publicationDate":"2026-09-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148927004","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Impact assessment of aquifer recharge through multi-arch check dams in an overexploited crystalline aquifer, Gauribidanur, India","authors":"Caroline Louis,&nbsp;Rahul R. Shende,&nbsp;Baby Swetha S,&nbsp;Abdul Rajik Khan,&nbsp;Catherine Louis,&nbsp;Anakha Ajai,&nbsp;Suchetana Biswas","doi":"10.1007/s12517-026-12596-z","DOIUrl":"10.1007/s12517-026-12596-z","url":null,"abstract":"<div><p>Ensuring sustainable groundwater availability in semi-arid crystalline aquifers remains a major challenge due to long-term overexploitation, pronounced climatic variability, and limited natural recharge. This study evaluates the hydrogeological, ecological, and socio-economic impacts of 34 Multi-Arch Check Dams (MACDs) implemented in a semi-arid region where groundwater abstraction exceeded 213% of annual replenishable resources. To assess these impacts, this research employs a pre–post intervention framework, comparing data from 2016–2020 with 2021–2024. The methodology integrates multiple remotely sensed and field-based indicators derived from cluster-scale groundwater monitoring, Sentinel-2 Normalised Difference Vegetation Index (NDVI), NASA Soil Moisture Active Passive (SMAP) surface soil moisture, village-level agricultural statistics, and structured farmer surveys. Cumulative Rainfall Departure (CRD) analysis isolated MACD-driven storage changes from inter-annual climatic fluctuations. Results reveal a statistically significant increase in groundwater levels (<i>p</i> &lt; 0.05). During the severe 2023–2024 drought, indicator wells sustained water levels at 5.82–12.48 m bgl, whereas pre-project groundwater depths ranged from 76.18 to 90.26 m bgl under similar rainfall. Ecological improvements included an increased NDVI healthy vegetation ratio (0.18 to 0.27, <i>p</i> = 0.038), dense vegetation expansion by 42.5%, a rise in mean monthly soil moisture (17.75% to 20.25%, <i>p</i> = 0.032), and a 109% expansion in summer irrigated areas. This study concludes that MACDs significantly enhance aquifer resilience in overexploited crystalline settings, providing a transferable model for evaluating managed aquifer recharge programs globally by integrating multi-source data to decouple true groundwater recovery from transient climatic factors.</p></div>","PeriodicalId":476,"journal":{"name":"Arabian Journal of Geosciences","volume":"19 10","pages":""},"PeriodicalIF":1.827,"publicationDate":"2026-09-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148927005","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Evaluation of soil detachment capacity in rills of forestlands using artificial neural networks and multiple regressions","authors":"Misagh Parhizkar,&nbsp;Manuel Esteban Lucas-Borja,&nbsp;Demetrio Antonio Zema","doi":"10.1007/s12517-026-12597-y","DOIUrl":"10.1007/s12517-026-12597-y","url":null,"abstract":"<div><p>Artificial Neural Networks (ANN) and multiple regression (MLR) models have been widely used to predict soil erosion. Few investigations have, however, explored the prediction accuracy of soil detachment capacity in rills (D_c). This study has tested whether ANN and MLR can predict D_c in forestlands affected by deforestation and fire, and subjected to soil conservation techniques as anti-erosive measures. We used a database (partly published in recent articles) of more than 1000 measurements of D_c under variable soil slopes (S) and flow discharges (Q) by flume experiments on soil samples collected in Northern Iran. S and Q, the shear stress (τ) or stream power (Ω) together with a ‘dummy’ categorical variable related to the soil condition were adopted as predictors of D_c. The modelling exercise showed a low prediction capacity of ANN (coefficients of efficiency of Nash and Sutcliffe, NSE, &lt; 0.48). In contrast, the MLR equations were reliable and accurate models (NSE up to 0.88 for calibration and to 0.87 for validation) to simulate D_c. S and Q were better predictors of D_c compared to τ or Ω. The MLR equations accurately predicted the mean D_c with errors lower than 5%, while the peak error for the maximum D_c was 39%. These results confirm that MLR provides reliable predictions of soil detachment capacity in various forest soil conditions. The main limitations of this study are the small scale as well as only one sampling site for D_c measurements. In spite of these limitations, the accuracy of MLR models in predicting soil erodibility in rills of forestlands affected by deforestation and fire, and subjected to soil conservation techniques is promising for other applications.</p></div>","PeriodicalId":476,"journal":{"name":"Arabian Journal of Geosciences","volume":"19 10","pages":""},"PeriodicalIF":1.827,"publicationDate":"2026-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148903267","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}