JSJamel Seidu

Jamel Seidu

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Jamel Seidu has a total of 62 co-authors. Here are the top co-authors by number of shared publications:

  • Anthony Ewusi

    Department of Geological Engineering, Faculty of Geoscience and Environmental Studies, University of Mines and Technology

    7 shared publications
  • Jerry Samuel Yaw Kuma

    Department of Geological Engineering, Faculty of Geoscience and Environmental Studies, University of Mines and Technology (UMaT), Tarkwa, Ghana

    4 shared publications
  • Abid A. Memon

    Department of Mathematics and Social Sciences, Sukkur IBA University, Sukkur, 65200, Sindh, Pakistan

    3 shared publications
  • Anthony Ewusi

    Department of Geological Engineering, Faculty of Geoscience and Environmental Studies, University of Mines and Technology (UMaT), Tarkwa, Ghana

    3 shared publications
  • Jacob Kavikumar

    Department of Mathematics and Statistics, Faculty of Applied Sciences and Technology, Universiti Tun Hussein Onn Malaysia, Pagoh, Malaysia

    3 shared publications
  • Muhammad Asif Memon

    Department of Mathematics, Basic Sciences and Humanities Sukkur IBA University, Sukkur, Pakistan

    3 shared publications

Latest Publications

  • Open Access

    Understanding the hydrogeochemistry of groundwater quality in Wassa East area of Ghana: implications for health risk assessment

    by Jamel Seidu, Kingchrist Joe Obosu, Kwame Atta Frimpong, Sylvanus Sebbeh-Newton, Emmanuel Kofahene Mensah, Anthony Ewusi, Ellen Benyarku

    Groundwater is the primary drinking water source for communities in Wassa East, Ghana, but faces increasing threats from intensive farming, illegal artisanal mining, and improper waste disposal. This study integrates hydrogeochemical characterisation (Piper, Chadha, Gaillardet, Gibbs diagrams), multivariate statistics (PCA, hierarchical clustering), an Integrated Weighted Water Quality Index (IWQI) combining entropy and CRITIC weighting, and a USEPA-based human health risk assessment (carcinogenic and non-carcinogenic) to evaluate groundwater quality from 139 boreholes across the district. Results show that groundwater chemistry is controlled primarily by silicate weathering, with Ca-Mg-HCO₃, Ca-Mg-Cl, and Na-HCO₃-Cl as the dominant water types. Most physicochemical parameters are within WHO permissible limits, but iron (48 samples, mean 0.53 mg/L) and chromium (8 samples) exceed thresholds in localized areas. The IWQI classifies most groundwater as “good” for drinking, with poorer quality in southern/central zones driven by geogenic iron and colour. However, health risk assessment reveals a more concerning pattern: 12 samples exceed the non-carcinogenic hazard quotient for cyanide (maximum HQ = 4.56), 5 for arsenic, and 8 for chromium. Carcinogenic risks for arsenic and chromium exceed the 1 × 10 −4 threshold in several samples. Microbial contamination is widespread (89 samples positive for total coliforms; 35 for E. coli ). Hazard Index mapping shows unacceptable cumulative non-carcinogenic risks (HI > 1) concentrated in northern and central communities where mining and agricultural land use intensity is highest. The study concludes that while Wassa East’s groundwater is generally suitable for drinking based on IWQI classification, significant localised health risks exist from cyanide (artisanal mining), chromium, arsenic, and microbial pathogens. Routine monitoring and targeted interventions in hotspot communities (Daboase, Odumase, New Subri Zongo) are recommended.

  • Open Access

    Forecasting daily rainfall in Coastal Ghana with machine learning: an evaluation analysis of gated residual learning framework

    by Jamel Seidu, Ernest Kwame Ampomah, Emmanuel Asante, Kwame Atta Frimpong

    Accurate rainfall prediction is critical for climate adaptation, agriculture, water resource management, and other arenas, especially in tropical monsoon regions like Coastal Ghana. Rainfall exhibits strong seasonality, non-linearity, and high interannual variability in these areas. Traditional statistical models such as ARIMA struggle to capture these complex dynamics, especially during extreme rainfall events. Machine learning (ML) approaches offer flexible, non-parametric alternatives that can capture non-linear relationships. However, hybrid ML approaches that incorporate gated mechanisms for noise filtering and residual correction remain largely unexplored, especially in West African rainfall forecasting. This study proposed a gated residual learning (GRL) framework that integrates sequence-based models [long-short-term memory (LSTM) and Temporal Convolutional Network (TCNs)] and non-sequence learners [Support Vector Regression (SVR), Random Forest, XGBoost, and Gradient Boosting] with a secondary residual learner to correct systematic prediction errors and improve daily rainfall prediction accuracy. Using Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and R 2 metrics, the performance of the models was evaluated across the full test dataset, specifically during extreme rainfall events (≥90th percentile). Results from the study show that integration of gated residual learning (GRL) significantly enhances the performance of baseline models. Overall, Gated-SVR achieved the best RMSE and the highest R 2 . Gated-XGBoost attained the best MAE. Integration of GRL improves alignment with rainfall events and mitigates systematic bias. This underscores the framework's capacity to refine both classical and deep learning models.

  • Open Access

    Optimised stacking generalisation methodology for groundwater level prediction

    by Jamel Seidu, Anthony Ewusi, Jerry Samuel Yaw Kuma, Yao Yevenyo Ziggah

    Global climate change has brought a paradigm shift in groundwater management practices. For a sustainable groundwater supply, there is a need to predict the availability of groundwater due to the intermittent fluctuations in the discharge and recharge phenomena. This study proposes an optimised stacking generalisation methodology for groundwater level prediction. The proposed methodology had the Particle Swarm Optimization-Artificial Neural Network (PSO-ANN), Genetic Algorithm-Artificial Neural Network (GA-ANN) and Self-Adaptive Differential Evolutionary Extreme Learning Machine (SaDE-ELM) as the base learners. Intercomparison revealed the prediction strength of SaDE-ELM over the other base learners. Hence, the stacked model was formed when the SaDE-ELM was used as the meta learner. Statistical analyses show that the proposed stacking method performed better than the standalone hybrid PSO-ANN, GA-ANN and SaDE-ELM with average RMSE and R values of 0.2209 m and 0.78, respectively. The superiority of the stacking method was further revealed using the Taylor diagram to present the statistical comparison with the observations of all models used.


  • Open Access

    Advancing hydrogeological studies with particle swarm optimization-based inversion of VES data: a study in Lake Tana Sub-Basin, Ethiopia

    by Lingerew Nebere Kassie, Sileshi Mamo Fantaye, Sabrine Zaghdoudi, Adugnaw Tadesse, Jamel Seidu, Melkamu Meseret Alemu, Daniel Mengistu, Dejene Sahlu

    Groundwater occurrences, movements and interactions are governed by hydrogeological features, whose spatial variability in thickness and continuity introduces uncertainties in models derived from limited discrete data points. Integrating geophysical methods with borehole data is essential for accurately characterizing these features. However, conventional deterministic geophysical inversion approaches often suffer from trapping in local minima and strong dependence on initial models potentially resulting in inaccurate hydrogeological interpretations. This study applies a Particle Swarm Optimization (PSO)-based inversion algorithm to interpret 20 vertical electrical sounding (VES) datasets acquired using the Schlumberger configuration in the Lake Tana Sub-Basin, Ethiopia. Results reveal an 8–40 m thick low-resistivity sedimentary layer (4–12 Ω.m), underlain by a moderate-resistivity unit (9–20 Ω.m) of variable thickness. At approximately 40 m depth, another low-resistivity layer appears, followed by a very high-resistivity zone (>20 Ω.m) beginning around 70 m depth. To enhance structural interpretation, cubic interpolation was employed to construct 2D resistivity cross-sections along selected transects. The resulting models show strong correlation with borehole lithological logs, validating the inversion results. Through cooperative interpretation, hydrogeological units such as aquitard, leaky aquifer, aquitard and aquifer from top to bottom were identified. The surface low-resistivity layer is interpreted as an aquitard that thickens toward Lake Tana. Beneath it lies a leaky aquifer, which extends upland but not toward the lake. This is underlain by a regional aquitard that confines the main leaky aquifer at ~70 m depth. This study provides valuable insights into hydrostratigraphic characterization in the Sub-Basin. The successful application of PSO-based inversion also highlights its potential as a robust alternative to conventional techniques, paving the way for future hydrogeophysical investigations in the Lake Tana Sub-Basin.


  • Open Access

    Optimised Stacking Generalisation Methodology for Groundwater Level Prediction

    by Jamel Seidu, Anthony Ewusi, Jerry Samuel Yaw Kuma, Yao Yevenyo Ziggah

    In this current study, the stacking method, which is based on optimised artificial intelligence (AI), was employed. The methods used are the particle swarm optimization-artificial neural network (PSO-ANN), genetic algorithm-artificial neural network (GA-ANN) and self-adaptive differential evolutionary extreme learning machine (SaDE-ELM). In the first level classifier, rainfall, temperature and evaporation were used as the input parameters to predict groundwater level (GWL), which was the output parameter. The best predictor among the PSO-ANN, GA-ANN and SaDE-ELM was selected and used as the meta-learner. At this stage, the predicted GWL data from the optimal base-learners was used as input. Statistical analyses show that the proposed stacking method performed better than the standalone hybrid AI models, with average RMSE and R values of 0.2209 m and 0.78, respectively. The superiority of the stacking method was further revealed using the Taylor diagram to present the statistical comparison with the observations of all models used.

  • Open Access

    Assessing groundwater vulnerability to contamination using the SINTACS model and GIS: a case study at Akoon, Tarkwa-Ghana

    by Yamoa Munko-Abo, Joyceline Philomina, Jamel Seidu

    Even though groundwater constitutes the main water supply sources, it is fraught with challenges of water quality issues such as groundwater contamination in which most of these contaminants are external in origin. Groundwater has been treated as an important source of water supply due to relatively low vulnerability to contamination compared to surface waters. Groundwater in hard- rock aquifers, particularly in mining areas are known to be vulnerable to quality problems that may have serious impact on the human health as reported by Smedley et al. (in: British Geological Survey, Technical Report, WC/95/43. p 122, 1995). For the purpose of this study, the most widely used method for delineating areas where groundwater is more vulnerable to contamination is used which is the SINTACS model with the integration of a computerbased GIS. The SINTACS model uses seven environmental parameters to characterize the hydrogeological setting and evaluate the aquifer vulnerability to contamination. In this study the SINTACS model was applied to a part of Tarkwa specifically at Akoon to generate a vulnerability map for the area. The whole area was classified as medium, high and very high vulnerability zones which were characterized by the rock types of Tarkwa phyllite, Huni sandstone and the Banket series respectively. This suggests that boreholes within the Banket series which is mainly dominated by quartzite is more vulnerable to contamination. From the analysis, depth to water table and effective infiltration inflicted the greatest impact on the intrinsic vulnerability of the aquifer.


  • Open Access

    Hydrogeological risk assessment for mineral exploration in Ghana: A brief overview

    by Anthony Ewusi, Samuel Edem Kodzo Tetteh, Jamel Seidu, Isaac Ahenkorah


  • Open Access

    Looming Earthquake Threat in Ghana

    by Albert Kafui Klu, Edmund Nana Asare, Jamel Seidu, Nicholas Opoku

    The earthquake catalogue of Ghana since the first recorded earthquake and recently collected seismic data in earthquake-prone locations of the Greater Accra Region, Ghanas capital, are covered in this chapter. This study lists the locations, periods, magnitudes, and intensities of 44 earthquakes and tremors from the past 408years using the countrys seismic inventory. The HoliSurface approach was also utilised to conduct Multichannel Analysis of Surface Wave (MASW) surveys to collect seismic data from 22 sites in the southern Greater Accra Regions seismically active zones. The earthquake catalogue data and geophysical data will help determine if Ghana is facing a major earthquake. This analysis shows that consistent earth tremors over the previous few years increase the risk of a devastating earthquake. The Gutenberg-Richter relations further emphasise the probabilities of a looming earthquake threat in the Ghana. The geophysical investigations also revealed areas with relatively weak subsurface materials that might contribute to more destruction in the event of an earthquake. The fundamental periods of the sites being in resonance with the buildings is also another disaster in waiting.